From 85d7066867537b24ef13798e85bfdb4325ecdc43 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 12:39:05 +0800
Subject: [PATCH 01/42] updated README

---
 README-zh.md | 4 ++--
 README.md    | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/README-zh.md b/README-zh.md
index b6319a07..9ed4c2ae 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -1,10 +1,10 @@
 # Crawlab
 
 <p>
-  <a href="https://hub.docker.com/r/tikazyq/crawlab" target="_blank">
+  <a href="https://hub.docker.com/r/tikazyq/crawlab/builds" target="_blank">
     <img src="https://img.shields.io/docker/cloud/build/tikazyq/crawlab.svg?label=build&logo=docker">
   </a>
-  <a href="https://hub.docker.com/r/tikazyq/crawlab/builds" target="_blank">
+  <a href="https://hub.docker.com/r/tikazyq/crawlab" target="_blank">
     <img src="https://img.shields.io/docker/pulls/tikazyq/crawlab?label=pulls&logo=docker">
   </a>
   <a href="https://github.com/crawlab-team/crawlab/releases" target="_blank">
diff --git a/README.md b/README.md
index 20bc9e4b..e90df67f 100644
--- a/README.md
+++ b/README.md
@@ -1,10 +1,10 @@
 # Crawlab
 
 <p>
-  <a href="https://hub.docker.com/r/tikazyq/crawlab" target="_blank">
+  <a href="https://hub.docker.com/r/tikazyq/crawlab/builds" target="_blank">
     <img src="https://img.shields.io/docker/cloud/build/tikazyq/crawlab.svg?label=build&logo=docker">
   </a>
-  <a href="https://hub.docker.com/r/tikazyq/crawlab/builds" target="_blank">
+  <a href="https://hub.docker.com/r/tikazyq/crawlab" target="_blank">
     <img src="https://img.shields.io/docker/pulls/tikazyq/crawlab?label=pulls&logo=docker">
   </a>
   <a href="https://github.com/crawlab-team/crawlab/releases" target="_blank">

From b4004ad9deda147da13b612339e78f9aa3c76025 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 13:11:50 +0800
Subject: [PATCH 02/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 9 +++++++++
 CHANGELOG.md    | 8 ++++++++
 2 files changed, 17 insertions(+)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index 85c964d4..b9b8cc09 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -1,3 +1,12 @@
+# 0.4.4 (未知)
+
+### 功能 / 优化
+- **API 地址优化**. 在前端加入相对路径，因此用户不需要特别注明 `CRAWLAB_API_ADDRESS`.
+- **SDK 兼容**. 允许用户通过 Crawlab SDK 与 Scrapy 或通用爬虫集成.
+
+### Bug 修复
+- **`nil retuened` 错误**.
+
 # 0.4.3 (2020-01-07)
 
 ### 功能 / 优化
diff --git a/CHANGELOG.md b/CHANGELOG.md
index ebfed440..8e0a5ef1 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,11 @@
+# 0.4.4 (unknown)
+### Features / Enhancement
+- **API Address Optimization**. Added relative URL path in frontend so that users don't have to specify `CRAWLAB_API_ADDRESS` explicitly.
+- **SDK Compatiblity**. Allow users to integrate Scrapy or general spiders with Crawlab SDK.
+
+### Bug Fixes
+- **`nil retuened` error**.
+
 # 0.4.3 (2020-01-07)
 
 ### Features / Enhancement

From a9820bdba291a9341415677bbe0a6a61dde5a581 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 14:49:37 +0800
Subject: [PATCH 03/42] updated README.md

---
 README-zh.md | 60 +++++++++++++++++++++++++++++-----------------------
 README.md    | 52 ++++++++++++++++++++++++++-------------------
 2 files changed, 63 insertions(+), 49 deletions(-)

diff --git a/README-zh.md b/README-zh.md
index 9ed4c2ae..57f4af47 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -193,37 +193,43 @@ Redis是非常受欢迎的Key-Value数据库，在Crawlab中主要实现节点
 
 ## 与其他框架的集成
 
+[Crawlab SDK](https://github.com/crawlab-team/crawlab-sdk) 提供了一些 `helper` 方法来让您的爬虫更好的集成到 Crawlab 中，例如保存结果数据到 Crawlab 中等等。
+
+### 集成 Scrapy
+
+在 `settings.py` 中找到 `ITEM_PIPELINES`（`dict` 类型的变量），在其中添加如下内容。
+
+```python
+ITEM_PIPELINES = {
+    'crawlab.pipelines.CrawlabMongoPipeline': 888,
+}
+```
+
+然后，启动 Scrapy 爬虫，运行完成之后，您就应该能看到抓取结果出现在 **任务详情-结果** 里。
+
+### 通用 Python 爬虫
+
+将下列代码加入到您爬虫中的结果保存部分。
+
+```python
+# 引入保存结果方法
+from crawlab import save_item
+
+# 这是一个结果，需要为 dict 类型
+result = {'name': 'crawlab'}
+
+# 调用保存结果方法
+save_item(result)
+```
+
+然后，启动爬虫，运行完成之后，您就应该能看到抓取结果出现在 **任务详情-结果** 里。
+
+### 其他框架和语言
+
 爬虫任务本质上是由一个shell命令来实现的。任务ID将以环境变量`CRAWLAB_TASK_ID`的形式存在于爬虫任务运行的进程中，并以此来关联抓取数据。另外，`CRAWLAB_COLLECTION`是Crawlab传过来的所存放collection的名称。
 
 在爬虫程序中，需要将`CRAWLAB_TASK_ID`的值以`task_id`作为可以存入数据库中`CRAWLAB_COLLECTION`的collection中。这样Crawlab就知道如何将爬虫任务与抓取数据关联起来了。当前，Crawlab只支持MongoDB。
 
-### 集成Scrapy
-
-以下是Crawlab跟Scrapy集成的例子，利用了Crawlab传过来的task_id和collection_name。
-
-```python
-import os
-from pymongo import MongoClient
-
-MONGO_HOST = '192.168.99.100'
-MONGO_PORT = 27017
-MONGO_DB = 'crawlab_test'
-
-# scrapy example in the pipeline
-class JuejinPipeline(object):
-    mongo = MongoClient(host=MONGO_HOST, port=MONGO_PORT)
-    db = mongo[MONGO_DB]
-    col_name = os.environ.get('CRAWLAB_COLLECTION')
-    if not col_name:
-        col_name = 'test'
-    col = db[col_name]
-
-    def process_item(self, item, spider):
-        item['task_id'] = os.environ.get('CRAWLAB_TASK_ID')
-        self.col.save(item)
-        return item
-```
-
 ## 与其他框架比较
 
 现在已经有一些爬虫管理框架了，因此为啥还要用Crawlab？
diff --git a/README.md b/README.md
index e90df67f..3f6224a8 100644
--- a/README.md
+++ b/README.md
@@ -192,35 +192,43 @@ Frontend is a SPA based on
 
 ## Integration with Other Frameworks
 
-A crawling task is actually executed through a shell command. The Task ID will be passed to the crawling task process in the form of environment variable named `CRAWLAB_TASK_ID`. By doing so, the data can be related to a task. Also, another environment variable `CRAWLAB_COLLECTION` is passed by Crawlab as the name of the collection to store results data.
+[Crawlab SDK](https://github.com/crawlab-team/crawlab-sdk) provides some `helper` methods to make it easier for you to integrate your spiders into Crawlab, e.g. saving results.
+
+⚠️Note: make sure you have already installed `crawlab-sdk` using pip.
 
 ### Scrapy
 
-Below is an example to integrate Crawlab with Scrapy in pipelines. 
+In `settings.py` in your Scrapy project, find the variable named `ITEM_PIPELINES` (a `dict` variable). Add content below.
 
 ```python
-import os
-from pymongo import MongoClient
-
-MONGO_HOST = '192.168.99.100'
-MONGO_PORT = 27017
-MONGO_DB = 'crawlab_test'
-
-# scrapy example in the pipeline
-class JuejinPipeline(object):
-    mongo = MongoClient(host=MONGO_HOST, port=MONGO_PORT)
-    db = mongo[MONGO_DB]
-    col_name = os.environ.get('CRAWLAB_COLLECTION')
-    if not col_name:
-        col_name = 'test'
-    col = db[col_name]
-
-    def process_item(self, item, spider):
-        item['task_id'] = os.environ.get('CRAWLAB_TASK_ID')
-        self.col.save(item)
-        return item
+ITEM_PIPELINES = {
+    'crawlab.pipelines.CrawlabMongoPipeline': 888,
+}
 ```
 
+Then, start the Scrapy spider. After it's done, you should be able to see scraped results in **Task Detail -> Result**
+
+### General Python Spider
+
+Please add below content to your spider files to save results.
+
+```python
+# import result saving method
+from crawlab import save_item
+
+# this is a result record, must be dict type
+result = {'name': 'crawlab'}
+
+# call result saving method
+save_item(result)
+```
+
+Then, start the spider. After it's done, you should be able to see scraped results in **Task Detail -> Result**
+
+### Other Frameworks / Languages
+
+A crawling task is actually executed through a shell command. The Task ID will be passed to the crawling task process in the form of environment variable named `CRAWLAB_TASK_ID`. By doing so, the data can be related to a task. Also, another environment variable `CRAWLAB_COLLECTION` is passed by Crawlab as the name of the collection to store results data.
+
 ## Comparison with Other Frameworks
 
 There are existing spider management frameworks. So why use Crawlab? 

From 117ae581defbfd590297edb6518a7711d87a5938 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 15:57:29 +0800
Subject: [PATCH 04/42] updated vendor

---
 backend/go.mod                                |    3 +
 backend/go.sum                                |   31 +
 backend/services/notification/mail.go         |    7 +
 .../github.com/Masterminds/semver/.travis.yml |   27 +
 .../Masterminds/semver/CHANGELOG.md           |   86 +
 .../github.com/Masterminds/semver/LICENSE.txt |   20 +
 .../github.com/Masterminds/semver/Makefile    |   36 +
 .../github.com/Masterminds/semver/README.md   |  165 ++
 .../Masterminds/semver/appveyor.yml           |   44 +
 .../Masterminds/semver/collection.go          |   24 +
 .../Masterminds/semver/constraints.go         |  426 +++
 .../github.com/Masterminds/semver/doc.go      |  115 +
 .../github.com/Masterminds/semver/version.go  |  421 +++
 .../github.com/Masterminds/sprig/.gitignore   |    2 +
 .../github.com/Masterminds/sprig/.travis.yml  |   24 +
 .../github.com/Masterminds/sprig/CHANGELOG.md |  153 ++
 .../github.com/Masterminds/sprig/LICENSE.txt  |   20 +
 .../github.com/Masterminds/sprig/Makefile     |   13 +
 .../github.com/Masterminds/sprig/README.md    |   81 +
 .../github.com/Masterminds/sprig/appveyor.yml |   26 +
 .../github.com/Masterminds/sprig/crypto.go    |  435 +++
 .../github.com/Masterminds/sprig/date.go      |   76 +
 .../github.com/Masterminds/sprig/defaults.go  |   83 +
 .../github.com/Masterminds/sprig/dict.go      |   97 +
 .../github.com/Masterminds/sprig/doc.go       |   19 +
 .../github.com/Masterminds/sprig/functions.go |  285 ++
 .../github.com/Masterminds/sprig/glide.yaml   |   15 +
 .../github.com/Masterminds/sprig/list.go      |  291 ++
 .../github.com/Masterminds/sprig/numeric.go   |  159 ++
 .../github.com/Masterminds/sprig/reflect.go   |   28 +
 .../github.com/Masterminds/sprig/regex.go     |   35 +
 .../github.com/Masterminds/sprig/semver.go    |   23 +
 .../github.com/Masterminds/sprig/strings.go   |  210 ++
 .../github.com/aokoli/goutils/.travis.yml     |   18 +
 .../github.com/aokoli/goutils/CHANGELOG.md    |    8 +
 .../github.com/aokoli/goutils/LICENSE.txt     |  202 ++
 .../github.com/aokoli/goutils/README.md       |   70 +
 .../github.com/aokoli/goutils/appveyor.yml    |   21 +
 .../aokoli/goutils/randomstringutils.go       |  268 ++
 .../github.com/aokoli/goutils/stringutils.go  |  224 ++
 .../github.com/aokoli/goutils/wordutils.go    |  356 +++
 .../vendor/github.com/google/uuid/.travis.yml |    9 +
 .../github.com/google/uuid/CONTRIBUTING.md    |   10 +
 .../github.com/google/uuid/CONTRIBUTORS       |    9 +
 backend/vendor/github.com/google/uuid/LICENSE |   27 +
 .../vendor/github.com/google/uuid/README.md   |   19 +
 backend/vendor/github.com/google/uuid/dce.go  |   80 +
 backend/vendor/github.com/google/uuid/doc.go  |   12 +
 backend/vendor/github.com/google/uuid/go.mod  |    1 +
 backend/vendor/github.com/google/uuid/hash.go |   53 +
 .../vendor/github.com/google/uuid/marshal.go  |   37 +
 backend/vendor/github.com/google/uuid/node.go |   90 +
 .../vendor/github.com/google/uuid/node_js.go  |   12 +
 .../vendor/github.com/google/uuid/node_net.go |   33 +
 backend/vendor/github.com/google/uuid/sql.go  |   59 +
 backend/vendor/github.com/google/uuid/time.go |  123 +
 backend/vendor/github.com/google/uuid/util.go |   43 +
 backend/vendor/github.com/google/uuid/uuid.go |  245 ++
 .../vendor/github.com/google/uuid/version1.go |   44 +
 .../vendor/github.com/google/uuid/version4.go |   38 +
 .../github.com/huandu/xstrings/.gitignore     |   24 +
 .../github.com/huandu/xstrings/.travis.yml    |    1 +
 .../huandu/xstrings/CONTRIBUTING.md           |   23 +
 .../vendor/github.com/huandu/xstrings/LICENSE |   22 +
 .../github.com/huandu/xstrings/README.md      |  115 +
 .../github.com/huandu/xstrings/common.go      |   25 +
 .../github.com/huandu/xstrings/convert.go     |  400 +++
 .../github.com/huandu/xstrings/count.go       |  120 +
 .../vendor/github.com/huandu/xstrings/doc.go  |    8 +
 .../github.com/huandu/xstrings/format.go      |  170 ++
 .../vendor/github.com/huandu/xstrings/go.mod  |    1 +
 .../github.com/huandu/xstrings/manipulate.go  |  217 ++
 .../github.com/huandu/xstrings/translate.go   |  547 ++++
 .../github.com/imdario/mergo/.gitignore       |   33 +
 .../github.com/imdario/mergo/.travis.yml      |    7 +
 .../imdario/mergo/CODE_OF_CONDUCT.md          |   46 +
 .../vendor/github.com/imdario/mergo/LICENSE   |   28 +
 .../vendor/github.com/imdario/mergo/README.md |  222 ++
 .../vendor/github.com/imdario/mergo/doc.go    |   44 +
 .../vendor/github.com/imdario/mergo/map.go    |  174 ++
 .../vendor/github.com/imdario/mergo/merge.go  |  252 ++
 .../vendor/github.com/imdario/mergo/mergo.go  |   97 +
 .../github.com/jaytaylor/html2text/.gitignore |   24 +
 .../jaytaylor/html2text/.travis.yml           |   14 +
 .../github.com/jaytaylor/html2text/LICENSE    |   22 +
 .../github.com/jaytaylor/html2text/README.md  |  137 +
 .../jaytaylor/html2text/html2text.go          |  473 ++++
 .../github.com/matcornic/hermes/.gitignore    |   28 +
 .../github.com/matcornic/hermes/.travis.yml   |   20 +
 .../matcornic/hermes/CONTRIBUTING.md          |  347 +++
 .../github.com/matcornic/hermes/Gopkg.toml    |   42 +
 .../github.com/matcornic/hermes/LICENSE       |  201 ++
 .../github.com/matcornic/hermes/README.md     |  373 +++
 .../github.com/matcornic/hermes/default.go    |  495 ++++
 .../github.com/matcornic/hermes/flat.go       |  495 ++++
 .../matcornic/hermes/gometalinter.json        |   17 +
 .../github.com/matcornic/hermes/hermes.go     |  200 ++
 .../github.com/mattn/go-runewidth/.travis.yml |    8 +
 .../github.com/mattn/go-runewidth/LICENSE     |   21 +
 .../github.com/mattn/go-runewidth/README.mkd  |   27 +
 .../mattn/go-runewidth/runewidth.go           | 1235 +++++++++
 .../mattn/go-runewidth/runewidth_js.go        |    8 +
 .../mattn/go-runewidth/runewidth_posix.go     |   77 +
 .../mattn/go-runewidth/runewidth_windows.go   |   25 +
 .../olekukonko/tablewriter/.gitignore         |   15 +
 .../olekukonko/tablewriter/.travis.yml        |   14 +
 .../olekukonko/tablewriter/LICENSE.md         |   19 +
 .../olekukonko/tablewriter/README.md          |  277 ++
 .../github.com/olekukonko/tablewriter/csv.go  |   52 +
 .../olekukonko/tablewriter/table.go           |  839 ++++++
 .../tablewriter/table_with_color.go           |  134 +
 .../github.com/olekukonko/tablewriter/util.go |   93 +
 .../github.com/olekukonko/tablewriter/wrap.go |   99 +
 .../sanitized_anchor_name/.travis.yml         |   16 +
 .../shurcooL/sanitized_anchor_name/LICENSE    |   21 +
 .../shurcooL/sanitized_anchor_name/README.md  |   36 +
 .../shurcooL/sanitized_anchor_name/go.mod     |    1 +
 .../shurcooL/sanitized_anchor_name/main.go    |   29 +
 .../vendor/github.com/ssor/bom/.travis.yml    |   14 +
 backend/vendor/github.com/ssor/bom/LICENSE    |   21 +
 backend/vendor/github.com/ssor/bom/README.md  |   23 +
 backend/vendor/github.com/ssor/bom/bom.go     |   34 +
 backend/vendor/golang.org/x/crypto/AUTHORS    |    3 +
 .../vendor/golang.org/x/crypto/CONTRIBUTORS   |    3 +
 backend/vendor/golang.org/x/crypto/LICENSE    |   27 +
 backend/vendor/golang.org/x/crypto/PATENTS    |   22 +
 .../golang.org/x/crypto/pbkdf2/pbkdf2.go      |   77 +
 .../golang.org/x/crypto/scrypt/scrypt.go      |  213 ++
 backend/vendor/golang.org/x/net/AUTHORS       |    3 +
 backend/vendor/golang.org/x/net/CONTRIBUTORS  |    3 +
 backend/vendor/golang.org/x/net/LICENSE       |   27 +
 backend/vendor/golang.org/x/net/PATENTS       |   22 +
 .../vendor/golang.org/x/net/html/atom/atom.go |   78 +
 .../golang.org/x/net/html/atom/table.go       |  783 ++++++
 backend/vendor/golang.org/x/net/html/const.go |  112 +
 backend/vendor/golang.org/x/net/html/doc.go   |  106 +
 .../vendor/golang.org/x/net/html/doctype.go   |  156 ++
 .../vendor/golang.org/x/net/html/entity.go    | 2253 +++++++++++++++
 .../vendor/golang.org/x/net/html/escape.go    |  258 ++
 .../vendor/golang.org/x/net/html/foreign.go   |  226 ++
 backend/vendor/golang.org/x/net/html/node.go  |  220 ++
 backend/vendor/golang.org/x/net/html/parse.go | 2417 +++++++++++++++++
 .../vendor/golang.org/x/net/html/render.go    |  271 ++
 backend/vendor/golang.org/x/net/html/token.go | 1219 +++++++++
 .../russross/blackfriday.v2/.gitignore        |    8 +
 .../russross/blackfriday.v2/.travis.yml       |   18 +
 .../russross/blackfriday.v2/LICENSE.txt       |   29 +
 .../russross/blackfriday.v2/README.md         |  283 ++
 .../gopkg.in/russross/blackfriday.v2/block.go | 1549 +++++++++++
 .../gopkg.in/russross/blackfriday.v2/doc.go   |   18 +
 .../gopkg.in/russross/blackfriday.v2/esc.go   |   34 +
 .../gopkg.in/russross/blackfriday.v2/html.go  |  940 +++++++
 .../russross/blackfriday.v2/inline.go         | 1214 +++++++++
 .../russross/blackfriday.v2/markdown.go       |  940 +++++++
 .../gopkg.in/russross/blackfriday.v2/node.go  |  354 +++
 .../russross/blackfriday.v2/smartypants.go    |  457 ++++
 backend/vendor/modules.txt                    |   32 +
 157 files changed, 28568 insertions(+)
 create mode 100644 backend/services/notification/mail.go
 create mode 100644 backend/vendor/github.com/Masterminds/semver/.travis.yml
 create mode 100644 backend/vendor/github.com/Masterminds/semver/CHANGELOG.md
 create mode 100644 backend/vendor/github.com/Masterminds/semver/LICENSE.txt
 create mode 100644 backend/vendor/github.com/Masterminds/semver/Makefile
 create mode 100644 backend/vendor/github.com/Masterminds/semver/README.md
 create mode 100644 backend/vendor/github.com/Masterminds/semver/appveyor.yml
 create mode 100644 backend/vendor/github.com/Masterminds/semver/collection.go
 create mode 100644 backend/vendor/github.com/Masterminds/semver/constraints.go
 create mode 100644 backend/vendor/github.com/Masterminds/semver/doc.go
 create mode 100644 backend/vendor/github.com/Masterminds/semver/version.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/.gitignore
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/.travis.yml
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/CHANGELOG.md
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/LICENSE.txt
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/Makefile
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/README.md
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/appveyor.yml
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/crypto.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/date.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/defaults.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/dict.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/doc.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/functions.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/glide.yaml
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/list.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/numeric.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/reflect.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/regex.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/semver.go
 create mode 100644 backend/vendor/github.com/Masterminds/sprig/strings.go
 create mode 100644 backend/vendor/github.com/aokoli/goutils/.travis.yml
 create mode 100644 backend/vendor/github.com/aokoli/goutils/CHANGELOG.md
 create mode 100644 backend/vendor/github.com/aokoli/goutils/LICENSE.txt
 create mode 100644 backend/vendor/github.com/aokoli/goutils/README.md
 create mode 100644 backend/vendor/github.com/aokoli/goutils/appveyor.yml
 create mode 100644 backend/vendor/github.com/aokoli/goutils/randomstringutils.go
 create mode 100644 backend/vendor/github.com/aokoli/goutils/stringutils.go
 create mode 100644 backend/vendor/github.com/aokoli/goutils/wordutils.go
 create mode 100644 backend/vendor/github.com/google/uuid/.travis.yml
 create mode 100644 backend/vendor/github.com/google/uuid/CONTRIBUTING.md
 create mode 100644 backend/vendor/github.com/google/uuid/CONTRIBUTORS
 create mode 100644 backend/vendor/github.com/google/uuid/LICENSE
 create mode 100644 backend/vendor/github.com/google/uuid/README.md
 create mode 100644 backend/vendor/github.com/google/uuid/dce.go
 create mode 100644 backend/vendor/github.com/google/uuid/doc.go
 create mode 100644 backend/vendor/github.com/google/uuid/go.mod
 create mode 100644 backend/vendor/github.com/google/uuid/hash.go
 create mode 100644 backend/vendor/github.com/google/uuid/marshal.go
 create mode 100644 backend/vendor/github.com/google/uuid/node.go
 create mode 100644 backend/vendor/github.com/google/uuid/node_js.go
 create mode 100644 backend/vendor/github.com/google/uuid/node_net.go
 create mode 100644 backend/vendor/github.com/google/uuid/sql.go
 create mode 100644 backend/vendor/github.com/google/uuid/time.go
 create mode 100644 backend/vendor/github.com/google/uuid/util.go
 create mode 100644 backend/vendor/github.com/google/uuid/uuid.go
 create mode 100644 backend/vendor/github.com/google/uuid/version1.go
 create mode 100644 backend/vendor/github.com/google/uuid/version4.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/.gitignore
 create mode 100644 backend/vendor/github.com/huandu/xstrings/.travis.yml
 create mode 100644 backend/vendor/github.com/huandu/xstrings/CONTRIBUTING.md
 create mode 100644 backend/vendor/github.com/huandu/xstrings/LICENSE
 create mode 100644 backend/vendor/github.com/huandu/xstrings/README.md
 create mode 100644 backend/vendor/github.com/huandu/xstrings/common.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/convert.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/count.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/doc.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/format.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/go.mod
 create mode 100644 backend/vendor/github.com/huandu/xstrings/manipulate.go
 create mode 100644 backend/vendor/github.com/huandu/xstrings/translate.go
 create mode 100644 backend/vendor/github.com/imdario/mergo/.gitignore
 create mode 100644 backend/vendor/github.com/imdario/mergo/.travis.yml
 create mode 100644 backend/vendor/github.com/imdario/mergo/CODE_OF_CONDUCT.md
 create mode 100644 backend/vendor/github.com/imdario/mergo/LICENSE
 create mode 100644 backend/vendor/github.com/imdario/mergo/README.md
 create mode 100644 backend/vendor/github.com/imdario/mergo/doc.go
 create mode 100644 backend/vendor/github.com/imdario/mergo/map.go
 create mode 100644 backend/vendor/github.com/imdario/mergo/merge.go
 create mode 100644 backend/vendor/github.com/imdario/mergo/mergo.go
 create mode 100644 backend/vendor/github.com/jaytaylor/html2text/.gitignore
 create mode 100644 backend/vendor/github.com/jaytaylor/html2text/.travis.yml
 create mode 100644 backend/vendor/github.com/jaytaylor/html2text/LICENSE
 create mode 100644 backend/vendor/github.com/jaytaylor/html2text/README.md
 create mode 100644 backend/vendor/github.com/jaytaylor/html2text/html2text.go
 create mode 100644 backend/vendor/github.com/matcornic/hermes/.gitignore
 create mode 100644 backend/vendor/github.com/matcornic/hermes/.travis.yml
 create mode 100644 backend/vendor/github.com/matcornic/hermes/CONTRIBUTING.md
 create mode 100644 backend/vendor/github.com/matcornic/hermes/Gopkg.toml
 create mode 100644 backend/vendor/github.com/matcornic/hermes/LICENSE
 create mode 100644 backend/vendor/github.com/matcornic/hermes/README.md
 create mode 100644 backend/vendor/github.com/matcornic/hermes/default.go
 create mode 100644 backend/vendor/github.com/matcornic/hermes/flat.go
 create mode 100644 backend/vendor/github.com/matcornic/hermes/gometalinter.json
 create mode 100644 backend/vendor/github.com/matcornic/hermes/hermes.go
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/.travis.yml
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/LICENSE
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/README.mkd
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/runewidth.go
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/runewidth_js.go
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/runewidth_posix.go
 create mode 100644 backend/vendor/github.com/mattn/go-runewidth/runewidth_windows.go
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/.gitignore
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/.travis.yml
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/LICENSE.md
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/README.md
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/csv.go
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/table.go
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/table_with_color.go
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/util.go
 create mode 100644 backend/vendor/github.com/olekukonko/tablewriter/wrap.go
 create mode 100644 backend/vendor/github.com/shurcooL/sanitized_anchor_name/.travis.yml
 create mode 100644 backend/vendor/github.com/shurcooL/sanitized_anchor_name/LICENSE
 create mode 100644 backend/vendor/github.com/shurcooL/sanitized_anchor_name/README.md
 create mode 100644 backend/vendor/github.com/shurcooL/sanitized_anchor_name/go.mod
 create mode 100644 backend/vendor/github.com/shurcooL/sanitized_anchor_name/main.go
 create mode 100644 backend/vendor/github.com/ssor/bom/.travis.yml
 create mode 100644 backend/vendor/github.com/ssor/bom/LICENSE
 create mode 100644 backend/vendor/github.com/ssor/bom/README.md
 create mode 100644 backend/vendor/github.com/ssor/bom/bom.go
 create mode 100644 backend/vendor/golang.org/x/crypto/AUTHORS
 create mode 100644 backend/vendor/golang.org/x/crypto/CONTRIBUTORS
 create mode 100644 backend/vendor/golang.org/x/crypto/LICENSE
 create mode 100644 backend/vendor/golang.org/x/crypto/PATENTS
 create mode 100644 backend/vendor/golang.org/x/crypto/pbkdf2/pbkdf2.go
 create mode 100644 backend/vendor/golang.org/x/crypto/scrypt/scrypt.go
 create mode 100644 backend/vendor/golang.org/x/net/AUTHORS
 create mode 100644 backend/vendor/golang.org/x/net/CONTRIBUTORS
 create mode 100644 backend/vendor/golang.org/x/net/LICENSE
 create mode 100644 backend/vendor/golang.org/x/net/PATENTS
 create mode 100644 backend/vendor/golang.org/x/net/html/atom/atom.go
 create mode 100644 backend/vendor/golang.org/x/net/html/atom/table.go
 create mode 100644 backend/vendor/golang.org/x/net/html/const.go
 create mode 100644 backend/vendor/golang.org/x/net/html/doc.go
 create mode 100644 backend/vendor/golang.org/x/net/html/doctype.go
 create mode 100644 backend/vendor/golang.org/x/net/html/entity.go
 create mode 100644 backend/vendor/golang.org/x/net/html/escape.go
 create mode 100644 backend/vendor/golang.org/x/net/html/foreign.go
 create mode 100644 backend/vendor/golang.org/x/net/html/node.go
 create mode 100644 backend/vendor/golang.org/x/net/html/parse.go
 create mode 100644 backend/vendor/golang.org/x/net/html/render.go
 create mode 100644 backend/vendor/golang.org/x/net/html/token.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/.gitignore
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/.travis.yml
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/LICENSE.txt
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/README.md
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/block.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/doc.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/esc.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/html.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/inline.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/markdown.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/node.go
 create mode 100644 backend/vendor/gopkg.in/russross/blackfriday.v2/smartypants.go

diff --git a/backend/go.mod b/backend/go.mod
index 89bbdbbc..497c5b4d 100644
--- a/backend/go.mod
+++ b/backend/go.mod
@@ -13,10 +13,13 @@ require (
 	github.com/gomodule/redigo v2.0.0+incompatible
 	github.com/imroc/req v0.2.4
 	github.com/leodido/go-urn v1.1.0 // indirect
+	github.com/matcornic/hermes v1.2.0
+	github.com/matcornic/hermes/v2 v2.0.2 // indirect
 	github.com/pkg/errors v0.8.1
 	github.com/satori/go.uuid v1.2.0
 	github.com/smartystreets/goconvey v0.0.0-20190731233626-505e41936337
 	github.com/spf13/viper v1.4.0
 	gopkg.in/go-playground/validator.v9 v9.29.1
+	gopkg.in/russross/blackfriday.v2 v2.0.0 // indirect
 	gopkg.in/yaml.v2 v2.2.2
 )
diff --git a/backend/go.sum b/backend/go.sum
index e4386cc9..5a186edd 100644
--- a/backend/go.sum
+++ b/backend/go.sum
@@ -1,9 +1,15 @@
 cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
 github.com/BurntSushi/toml v0.3.1 h1:WXkYYl6Yr3qBf1K79EBnL4mak0OimBfB0XUf9Vl28OQ=
 github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
+github.com/Masterminds/semver v1.4.2 h1:WBLTQ37jOCzSLtXNdoo8bNM8876KhNqOKvrlGITgsTc=
+github.com/Masterminds/semver v1.4.2/go.mod h1:MB6lktGJrhw8PrUyiEoblNEGEQ+RzHPF078ddwwvV3Y=
+github.com/Masterminds/sprig v2.16.0+incompatible h1:QZbMUPxRQ50EKAq3LFMnxddMu88/EUUG3qmxwtDmPsY=
+github.com/Masterminds/sprig v2.16.0+incompatible/go.mod h1:y6hNFY5UBTIWBxnzTeuNhlNS5hqE0NB0E6fgfo2Br3o=
 github.com/OneOfOne/xxhash v1.2.2/go.mod h1:HSdplMjZKSmBqAxg5vPj2TmRDmfkzw+cTzAElWljhcU=
 github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=
 github.com/alecthomas/units v0.0.0-20151022065526-2efee857e7cf/go.mod h1:ybxpYRFXyAe+OPACYpWeL0wqObRcbAqCMya13uyzqw0=
+github.com/aokoli/goutils v1.0.1 h1:7fpzNGoJ3VA8qcrm++XEE1QUe0mIwNeLa02Nwq7RDkg=
+github.com/aokoli/goutils v1.0.1/go.mod h1:SijmP0QR8LtwsmDs8Yii5Z/S4trXFGFC2oO5g9DP+DQ=
 github.com/apex/log v1.1.1 h1:BwhRZ0qbjYtTob0I+2M+smavV0kOC8XgcnGZcyL9liA=
 github.com/apex/log v1.1.1/go.mod h1:Ls949n1HFtXfbDcjiTTFQqkVUrte0puoIBfO3SVgwOA=
 github.com/aphistic/golf v0.0.0-20180712155816-02c07f170c5a/go.mod h1:3NqKYiepwy8kCu4PNA+aP7WUV72eXWJeP9/r3/K9aLE=
@@ -56,6 +62,8 @@ github.com/gomodule/redigo v2.0.0+incompatible h1:K/R+8tc58AaqLkqG2Ol3Qk+DR/TlNu
 github.com/gomodule/redigo v2.0.0+incompatible/go.mod h1:B4C85qUVwatsJoIUNIfCRsp7qO0iAmpGFZ4EELWSbC4=
 github.com/google/btree v1.0.0/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
 github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
+github.com/google/uuid v1.0.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/google/uuid v1.1.1 h1:Gkbcsh/GbpXz7lPftLA3P6TYMwjCLYm83jiFQZF/3gY=
 github.com/google/uuid v1.1.1/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/gopherjs/gopherjs v0.0.0-20181017120253-0766667cb4d1 h1:EGx4pi6eqNxGaHF6qqu48+N2wcFQ5qg5FXgOdqsJ5d8=
 github.com/gopherjs/gopherjs v0.0.0-20181017120253-0766667cb4d1/go.mod h1:wJfORRmW1u3UXTncJ5qlYoELFm8eSnnEO6hX4iZ3EWY=
@@ -66,8 +74,14 @@ github.com/grpc-ecosystem/grpc-gateway v1.9.0/go.mod h1:vNeuVxBJEsws4ogUvrchl83t
 github.com/hashicorp/hcl v1.0.0 h1:0Anlzjpi4vEasTeNFn2mLJgTSwt0+6sfsiTG8qcWGx4=
 github.com/hashicorp/hcl v1.0.0/go.mod h1:E5yfLk+7swimpb2L/Alb/PJmXilQ/rhwaUYs4T20WEQ=
 github.com/hpcloud/tail v1.0.0/go.mod h1:ab1qPbhIpdTxEkNHXyeSf5vhxWSCs/tWer42PpOxQnU=
+github.com/huandu/xstrings v1.2.0 h1:yPeWdRnmynF7p+lLYz0H2tthW9lqhMJrQV/U7yy4wX0=
+github.com/huandu/xstrings v1.2.0/go.mod h1:DvyZB1rfVYsBIigL8HwpZgxHwXozlTgGqn63UyNX5k4=
+github.com/imdario/mergo v0.3.6 h1:xTNEAn+kxVO7dTZGu0CegyqKZmoWFI0rF8UxjlB2d28=
+github.com/imdario/mergo v0.3.6/go.mod h1:2EnlNZ0deacrJVfApfmtdGgDfMuh/nq6Ok1EcJh5FfA=
 github.com/imroc/req v0.2.4 h1:8XbvaQpERLAJV6as/cB186DtH5f0m5zAOtHEaTQ4ac0=
 github.com/imroc/req v0.2.4/go.mod h1:J9FsaNHDTIVyW/b5r6/Df5qKEEEq2WzZKIgKSajd1AE=
+github.com/jaytaylor/html2text v0.0.0-20180606194806-57d518f124b0 h1:xqgexXAGQgY3HAjNPSaCqn5Aahbo5TKsmhp8VRfr1iQ=
+github.com/jaytaylor/html2text v0.0.0-20180606194806-57d518f124b0/go.mod h1:CVKlgaMiht+LXvHG173ujK6JUhZXKb2u/BQtjPDIvyk=
 github.com/jmespath/go-jmespath v0.0.0-20180206201540-c2b33e8439af/go.mod h1:Nht3zPeWKUH0NzdCt2Blrr5ys8VGpn0CEB0cQHVjt7k=
 github.com/jonboulle/clockwork v0.1.0/go.mod h1:Ii8DK3G1RaLaWxj9trq07+26W01tbo22gdxWY5EU2bo=
 github.com/jpillora/backoff v0.0.0-20180909062703-3050d21c67d7/go.mod h1:2iMrUgbbvHEiQClaW2NsSzMyGHqN+rDFqY705q49KG0=
@@ -89,12 +103,17 @@ github.com/leodido/go-urn v1.1.0 h1:Sm1gr51B1kKyfD2BlRcLSiEkffoG96g6TPv6eRoEiB8=
 github.com/leodido/go-urn v1.1.0/go.mod h1:+cyI34gQWZcE1eQU7NVgKkkzdXDQHr1dBMtdAPozLkw=
 github.com/magiconair/properties v1.8.0 h1:LLgXmsheXeRoUOBOjtwPQCWIYqM/LU1ayDtDePerRcY=
 github.com/magiconair/properties v1.8.0/go.mod h1:PppfXfuXeibc/6YijjN8zIbojt8czPbwD3XqdrwzmxQ=
+github.com/matcornic/hermes v1.2.0 h1:AuqZpYcTOtTB7cahdevLfnhIpfzmpqw5Czv8vpdnFDU=
+github.com/matcornic/hermes v1.2.0/go.mod h1:lujJomb016Xjv8wBnWlNvUdtmvowjjfkqri5J/+1hYc=
+github.com/matcornic/hermes/v2 v2.0.2/go.mod h1:iVsJWSIS4NtMNtgan22sy6lt7pImok7bATGPWCoaKNY=
 github.com/mattn/go-colorable v0.1.1/go.mod h1:FuOcm+DKB9mbwrcAfNl7/TZVBZ6rcnceauSikq3lYCQ=
 github.com/mattn/go-colorable v0.1.2/go.mod h1:U0ppj6V5qS13XJ6of8GYAs25YV2eR4EVcfRqFIhoBtE=
 github.com/mattn/go-isatty v0.0.5/go.mod h1:Iq45c/XA43vh69/j3iqttzPXn0bhXyGjM0Hdxcsrc5s=
 github.com/mattn/go-isatty v0.0.7/go.mod h1:Iq45c/XA43vh69/j3iqttzPXn0bhXyGjM0Hdxcsrc5s=
 github.com/mattn/go-isatty v0.0.8 h1:HLtExJ+uU2HOZ+wI0Tt5DtUDrx8yhUqDcp7fYERX4CE=
 github.com/mattn/go-isatty v0.0.8/go.mod h1:Iq45c/XA43vh69/j3iqttzPXn0bhXyGjM0Hdxcsrc5s=
+github.com/mattn/go-runewidth v0.0.3 h1:a+kO+98RDGEfo6asOGMmpodZq4FNtnGP54yps8BzLR4=
+github.com/mattn/go-runewidth v0.0.3/go.mod h1:LwmH8dsx7+W8Uxz3IHJYH5QSwggIsqBzpuz5H//U1FU=
 github.com/matttproud/golang_protobuf_extensions v1.0.1/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
 github.com/mgutz/ansi v0.0.0-20170206155736-9520e82c474b/go.mod h1:01TrycV0kFyexm33Z7vhZRXopbI8J3TDReVlkTgMUxE=
 github.com/mitchellh/mapstructure v1.1.2 h1:fmNYVwqnSfB9mZU6OS2O6GsXM+wcskZDuKQzvN1EDeE=
@@ -105,6 +124,8 @@ github.com/modern-go/reflect2 v1.0.1 h1:9f412s+6RmYXLWZSEzVVgPGK7C2PphHj5RJrvfx9
 github.com/modern-go/reflect2 v1.0.1/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/mwitkow/go-conntrack v0.0.0-20161129095857-cc309e4a2223/go.mod h1:qRWi+5nqEBWmkhHvq77mSJWrCKwh8bxhgT7d/eI7P4U=
 github.com/oklog/ulid v1.3.1/go.mod h1:CirwcVhetQ6Lv90oh/F+FBtV6XMibvdAFo93nm5qn4U=
+github.com/olekukonko/tablewriter v0.0.1 h1:b3iUnf1v+ppJiOfNX4yxxqfWKMQPZR5yoh8urCTFX88=
+github.com/olekukonko/tablewriter v0.0.1/go.mod h1:vsDQFd/mU46D+Z4whnwzcISnGGzXWMclvtLoiIKAKIo=
 github.com/onsi/ginkgo v1.6.0/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+WWjE=
 github.com/onsi/gomega v1.5.0/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
 github.com/pelletier/go-toml v1.2.0 h1:T5zMGML61Wp+FlcbWjRDT7yAxhJNAiPPLOFECq181zc=
@@ -125,9 +146,12 @@ github.com/prometheus/procfs v0.0.0-20190507164030-5867b95ac084/go.mod h1:TjEm7z
 github.com/prometheus/tsdb v0.7.1/go.mod h1:qhTCs0VvXwvX/y3TZrWD7rabWM+ijKTux40TwIPHuXU=
 github.com/rogpeppe/fastuuid v0.0.0-20150106093220-6724a57986af/go.mod h1:XWv6SoW27p1b0cqNHllgS5HIMJraePCO15w5zCzIWYg=
 github.com/rogpeppe/fastuuid v1.1.0/go.mod h1:jVj6XXZzXRy/MSR5jhDC/2q6DgLz+nrA6LYCDYWNEvQ=
+github.com/russross/blackfriday/v2 v2.0.1/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
 github.com/satori/go.uuid v1.2.0 h1:0uYX9dsZ2yD7q2RtLRtPSdGDWzjeM3TbMJP9utgA0ww=
 github.com/satori/go.uuid v1.2.0/go.mod h1:dA0hQrYB0VpLJoorglMZABFdXlWrHn1NEOzdhQKdks0=
 github.com/sergi/go-diff v1.0.0/go.mod h1:0CfEIISq7TuYL3j771MWULgwwjU+GofnZX9QAmXWZgo=
+github.com/shurcooL/sanitized_anchor_name v1.0.0 h1:PdmoCO6wvbs+7yrJyMORt4/BmY5IYyJwS/kOiWx8mHo=
+github.com/shurcooL/sanitized_anchor_name v1.0.0/go.mod h1:1NzhyTcUVG4SuEtjjoZeVRXNmyL/1OwPU0+IJeTBvfc=
 github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPxbbu5VWo=
 github.com/smartystreets/assertions v0.0.0-20180927180507-b2de0cb4f26d/go.mod h1:OnSkiWE9lh6wB0YB77sQom3nweQdgAjqCqsofrRNTgc=
 github.com/smartystreets/assertions v1.0.0 h1:UVQPSSmc3qtTi+zPPkCXvZX9VvW/xT/NsRvKfwY81a8=
@@ -148,6 +172,8 @@ github.com/spf13/pflag v1.0.3 h1:zPAT6CGy6wXeQ7NtTnaTerfKOsV6V6F8agHXFiazDkg=
 github.com/spf13/pflag v1.0.3/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
 github.com/spf13/viper v1.4.0 h1:yXHLWeravcrgGyFSyCgdYpXQ9dR9c/WED3pg1RhxqEU=
 github.com/spf13/viper v1.4.0/go.mod h1:PTJ7Z/lr49W6bUbkmS1V3by4uWynFiR9p7+dSq/yZzE=
+github.com/ssor/bom v0.0.0-20170718123548-6386211fdfcf h1:pvbZ0lM0XWPBqUKqFU8cmavspvIl9nulOYwdy6IFRRo=
+github.com/ssor/bom v0.0.0-20170718123548-6386211fdfcf/go.mod h1:RJID2RhlZKId02nZ62WenDCkgHFerpIOmW0iT7GKmXM=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.1.1/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
@@ -167,12 +193,15 @@ go.uber.org/atomic v1.4.0/go.mod h1:gD2HeocX3+yG+ygLZcrzQJaqmWj9AIm7n08wl/qW/PE=
 go.uber.org/multierr v1.1.0/go.mod h1:wR5kodmAFQ0UK8QlbwjlSNy0Z68gJhDJUG5sjR94q/0=
 go.uber.org/zap v1.10.0/go.mod h1:vwi/ZaCAaUcBkycHslxD9B2zi4UTXhF60s6SWpuDF0Q=
 golang.org/x/crypto v0.0.0-20180904163835-0709b304e793/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
+golang.org/x/crypto v0.0.0-20181029175232-7e6ffbd03851/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
+golang.org/x/crypto v0.0.0-20190426145343-a29dc8fdc734 h1:p/H982KKEjUnLJkM3tt/LemDnOc1GiZL5FCVlORJ5zo=
 golang.org/x/crypto v0.0.0-20190426145343-a29dc8fdc734/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
 golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
 golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20180906233101-161cd47e91fd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20181029044818-c44066c5c816/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20181220203305-927f97764cc3/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
@@ -217,6 +246,8 @@ gopkg.in/go-playground/validator.v8 v8.18.2/go.mod h1:RX2a/7Ha8BgOhfk7j780h4/u/R
 gopkg.in/go-playground/validator.v9 v9.29.1 h1:SvGtYmN60a5CVKTOzMSyfzWDeZRxRuGvRQyEAKbw1xc=
 gopkg.in/go-playground/validator.v9 v9.29.1/go.mod h1:+c9/zcJMFNgbLvly1L1V+PpxWdVbfP1avr/N00E2vyQ=
 gopkg.in/resty.v1 v1.12.0/go.mod h1:mDo4pnntr5jdWRML875a/NmxYqAlA73dVijT2AXvQQo=
+gopkg.in/russross/blackfriday.v2 v2.0.0 h1:+FlnIV8DSQnT7NZ43hcVKcdJdzZoeCmJj4Ql8gq5keA=
+gopkg.in/russross/blackfriday.v2 v2.0.0/go.mod h1:6sSBNz/GtOm/pJTuh5UmBK2ZHfmnxGbl2NZg1UliSOI=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7/go.mod h1:dt/ZhP58zS4L8KSrWDmTeBkI65Dw0HsyUHuEVlX15mw=
 gopkg.in/yaml.v2 v2.0.0-20170812160011-eb3733d160e7/go.mod h1:JAlM8MvJe8wmxCU4Bli9HhUf9+ttbYbLASfIpnQbh74=
 gopkg.in/yaml.v2 v2.2.1/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
diff --git a/backend/services/notification/mail.go b/backend/services/notification/mail.go
new file mode 100644
index 00000000..dd1490dc
--- /dev/null
+++ b/backend/services/notification/mail.go
@@ -0,0 +1,7 @@
+package notification
+
+import "github.com/matcornic/hermes"
+
+func SendMail() error {
+	hermes
+}
diff --git a/backend/vendor/github.com/Masterminds/semver/.travis.yml b/backend/vendor/github.com/Masterminds/semver/.travis.yml
new file mode 100644
index 00000000..3d9ebadb
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/.travis.yml
@@ -0,0 +1,27 @@
+language: go
+
+go:
+  - 1.6.x
+  - 1.7.x
+  - 1.8.x
+  - 1.9.x
+  - 1.10.x
+  - tip
+
+# Setting sudo access to false will let Travis CI use containers rather than
+# VMs to run the tests. For more details see:
+# - http://docs.travis-ci.com/user/workers/container-based-infrastructure/
+# - http://docs.travis-ci.com/user/workers/standard-infrastructure/
+sudo: false
+
+script:
+  - make setup
+  - make test
+
+notifications:
+  webhooks:
+    urls:
+      - https://webhooks.gitter.im/e/06e3328629952dabe3e0
+    on_success: change  # options: [always|never|change] default: always
+    on_failure: always  # options: [always|never|change] default: always
+    on_start: never     # options: [always|never|change] default: always
diff --git a/backend/vendor/github.com/Masterminds/semver/CHANGELOG.md b/backend/vendor/github.com/Masterminds/semver/CHANGELOG.md
new file mode 100644
index 00000000..b888e20a
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/CHANGELOG.md
@@ -0,0 +1,86 @@
+# 1.4.2 (2018-04-10)
+
+## Changed
+- #72: Updated the docs to point to vert for a console appliaction
+- #71: Update the docs on pre-release comparator handling
+
+## Fixed
+- #70: Fix the handling of pre-releases and the 0.0.0 release edge case
+
+# 1.4.1 (2018-04-02)
+
+## Fixed
+- Fixed #64: Fix pre-release precedence issue (thanks @uudashr)
+
+# 1.4.0 (2017-10-04)
+
+## Changed
+- #61: Update NewVersion to parse ints with a 64bit int size (thanks @zknill)
+
+# 1.3.1 (2017-07-10)
+
+## Fixed
+- Fixed #57: number comparisons in prerelease sometimes inaccurate
+
+# 1.3.0 (2017-05-02)
+
+## Added
+- #45: Added json (un)marshaling support (thanks @mh-cbon)
+- Stability marker. See https://masterminds.github.io/stability/
+
+## Fixed
+- #51: Fix handling of single digit tilde constraint (thanks @dgodd)
+
+## Changed
+- #55: The godoc icon moved from png to svg
+
+# 1.2.3 (2017-04-03)
+
+## Fixed
+- #46: Fixed 0.x.x and 0.0.x in constraints being treated as *
+
+# Release 1.2.2 (2016-12-13)
+
+## Fixed
+- #34: Fixed issue where hyphen range was not working with pre-release parsing.
+
+# Release 1.2.1 (2016-11-28)
+
+## Fixed
+- #24: Fixed edge case issue where constraint "> 0" does not handle "0.0.1-alpha"
+  properly.
+
+# Release 1.2.0 (2016-11-04)
+
+## Added
+- #20: Added MustParse function for versions (thanks @adamreese)
+- #15: Added increment methods on versions (thanks @mh-cbon)
+
+## Fixed
+- Issue #21: Per the SemVer spec (section 9) a pre-release is unstable and
+  might not satisfy the intended compatibility. The change here ignores pre-releases
+  on constraint checks (e.g., ~ or ^) when a pre-release is not part of the
+  constraint. For example, `^1.2.3` will ignore pre-releases while
+  `^1.2.3-alpha` will include them.
+
+# Release 1.1.1 (2016-06-30)
+
+## Changed
+- Issue #9: Speed up version comparison performance (thanks @sdboyer)
+- Issue #8: Added benchmarks (thanks @sdboyer)
+- Updated Go Report Card URL to new location
+- Updated Readme to add code snippet formatting (thanks @mh-cbon)
+- Updating tagging to v[SemVer] structure for compatibility with other tools.
+
+# Release 1.1.0 (2016-03-11)
+
+- Issue #2: Implemented validation to provide reasons a versions failed a
+  constraint.
+
+# Release 1.0.1 (2015-12-31)
+
+- Fixed #1: * constraint failing on valid versions.
+
+# Release 1.0.0 (2015-10-20)
+
+- Initial release
diff --git a/backend/vendor/github.com/Masterminds/semver/LICENSE.txt b/backend/vendor/github.com/Masterminds/semver/LICENSE.txt
new file mode 100644
index 00000000..0da4aead
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/LICENSE.txt
@@ -0,0 +1,20 @@
+The Masterminds
+Copyright (C) 2014-2015, Matt Butcher and Matt Farina
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/backend/vendor/github.com/Masterminds/semver/Makefile b/backend/vendor/github.com/Masterminds/semver/Makefile
new file mode 100644
index 00000000..a7a1b4e3
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/Makefile
@@ -0,0 +1,36 @@
+.PHONY: setup
+setup:
+	go get -u gopkg.in/alecthomas/gometalinter.v1
+	gometalinter.v1 --install
+
+.PHONY: test
+test: validate lint
+	@echo "==> Running tests"
+	go test -v
+
+.PHONY: validate
+validate:
+	@echo "==> Running static validations"
+	@gometalinter.v1 \
+	  --disable-all \
+	  --enable deadcode \
+	  --severity deadcode:error \
+	  --enable gofmt \
+	  --enable gosimple \
+	  --enable ineffassign \
+	  --enable misspell \
+	  --enable vet \
+	  --tests \
+	  --vendor \
+	  --deadline 60s \
+	  ./... || exit_code=1
+
+.PHONY: lint
+lint:
+	@echo "==> Running linters"
+	@gometalinter.v1 \
+	  --disable-all \
+	  --enable golint \
+	  --vendor \
+	  --deadline 60s \
+	  ./... || :
diff --git a/backend/vendor/github.com/Masterminds/semver/README.md b/backend/vendor/github.com/Masterminds/semver/README.md
new file mode 100644
index 00000000..3e934ed7
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/README.md
@@ -0,0 +1,165 @@
+# SemVer
+
+The `semver` package provides the ability to work with [Semantic Versions](http://semver.org) in Go. Specifically it provides the ability to:
+
+* Parse semantic versions
+* Sort semantic versions
+* Check if a semantic version fits within a set of constraints
+* Optionally work with a `v` prefix
+
+[![Stability:
+Active](https://masterminds.github.io/stability/active.svg)](https://masterminds.github.io/stability/active.html)
+[![Build Status](https://travis-ci.org/Masterminds/semver.svg)](https://travis-ci.org/Masterminds/semver) [![Build status](https://ci.appveyor.com/api/projects/status/jfk66lib7hb985k8/branch/master?svg=true&passingText=windows%20build%20passing&failingText=windows%20build%20failing)](https://ci.appveyor.com/project/mattfarina/semver/branch/master) [![GoDoc](https://godoc.org/github.com/Masterminds/semver?status.svg)](https://godoc.org/github.com/Masterminds/semver) [![Go Report Card](https://goreportcard.com/badge/github.com/Masterminds/semver)](https://goreportcard.com/report/github.com/Masterminds/semver)
+
+## Parsing Semantic Versions
+
+To parse a semantic version use the `NewVersion` function. For example,
+
+```go
+    v, err := semver.NewVersion("1.2.3-beta.1+build345")
+```
+
+If there is an error the version wasn't parseable. The version object has methods
+to get the parts of the version, compare it to other versions, convert the
+version back into a string, and get the original string. For more details
+please see the [documentation](https://godoc.org/github.com/Masterminds/semver).
+
+## Sorting Semantic Versions
+
+A set of versions can be sorted using the [`sort`](https://golang.org/pkg/sort/)
+package from the standard library. For example,
+
+```go
+    raw := []string{"1.2.3", "1.0", "1.3", "2", "0.4.2",}
+    vs := make([]*semver.Version, len(raw))
+	for i, r := range raw {
+		v, err := semver.NewVersion(r)
+		if err != nil {
+			t.Errorf("Error parsing version: %s", err)
+		}
+
+		vs[i] = v
+	}
+
+	sort.Sort(semver.Collection(vs))
+```
+
+## Checking Version Constraints
+
+Checking a version against version constraints is one of the most featureful
+parts of the package.
+
+```go
+    c, err := semver.NewConstraint(">= 1.2.3")
+    if err != nil {
+        // Handle constraint not being parseable.
+    }
+
+    v, _ := semver.NewVersion("1.3")
+    if err != nil {
+        // Handle version not being parseable.
+    }
+    // Check if the version meets the constraints. The a variable will be true.
+    a := c.Check(v)
+```
+
+## Basic Comparisons
+
+There are two elements to the comparisons. First, a comparison string is a list
+of comma separated and comparisons. These are then separated by || separated or
+comparisons. For example, `">= 1.2, < 3.0.0 || >= 4.2.3"` is looking for a
+comparison that's greater than or equal to 1.2 and less than 3.0.0 or is
+greater than or equal to 4.2.3.
+
+The basic comparisons are:
+
+* `=`: equal (aliased to no operator)
+* `!=`: not equal
+* `>`: greater than
+* `<`: less than
+* `>=`: greater than or equal to
+* `<=`: less than or equal to
+
+_Note, according to the Semantic Version specification pre-releases may not be
+API compliant with their release counterpart. It says,_
+
+> _A pre-release version indicates that the version is unstable and might not satisfy the intended compatibility requirements as denoted by its associated normal version._
+
+_SemVer comparisons without a pre-release value will skip pre-release versions.
+For example, `>1.2.3` will skip pre-releases when looking at a list of values
+while `>1.2.3-alpha.1` will evaluate pre-releases._
+
+## Hyphen Range Comparisons
+
+There are multiple methods to handle ranges and the first is hyphens ranges.
+These look like:
+
+* `1.2 - 1.4.5` which is equivalent to `>= 1.2, <= 1.4.5`
+* `2.3.4 - 4.5` which is equivalent to `>= 2.3.4, <= 4.5`
+
+## Wildcards In Comparisons
+
+The `x`, `X`, and `*` characters can be used as a wildcard character. This works
+for all comparison operators. When used on the `=` operator it falls
+back to the pack level comparison (see tilde below). For example,
+
+* `1.2.x` is equivalent to `>= 1.2.0, < 1.3.0`
+* `>= 1.2.x` is equivalent to `>= 1.2.0`
+* `<= 2.x` is equivalent to `<= 3`
+* `*` is equivalent to `>= 0.0.0`
+
+## Tilde Range Comparisons (Patch)
+
+The tilde (`~`) comparison operator is for patch level ranges when a minor
+version is specified and major level changes when the minor number is missing.
+For example,
+
+* `~1.2.3` is equivalent to `>= 1.2.3, < 1.3.0`
+* `~1` is equivalent to `>= 1, < 2`
+* `~2.3` is equivalent to `>= 2.3, < 2.4`
+* `~1.2.x` is equivalent to `>= 1.2.0, < 1.3.0`
+* `~1.x` is equivalent to `>= 1, < 2`
+
+## Caret Range Comparisons (Major)
+
+The caret (`^`) comparison operator is for major level changes. This is useful
+when comparisons of API versions as a major change is API breaking. For example,
+
+* `^1.2.3` is equivalent to `>= 1.2.3, < 2.0.0`
+* `^1.2.x` is equivalent to `>= 1.2.0, < 2.0.0`
+* `^2.3` is equivalent to `>= 2.3, < 3`
+* `^2.x` is equivalent to `>= 2.0.0, < 3`
+
+# Validation
+
+In addition to testing a version against a constraint, a version can be validated
+against a constraint. When validation fails a slice of errors containing why a
+version didn't meet the constraint is returned. For example,
+
+```go
+    c, err := semver.NewConstraint("<= 1.2.3, >= 1.4")
+    if err != nil {
+        // Handle constraint not being parseable.
+    }
+
+    v, _ := semver.NewVersion("1.3")
+    if err != nil {
+        // Handle version not being parseable.
+    }
+
+    // Validate a version against a constraint.
+    a, msgs := c.Validate(v)
+    // a is false
+    for _, m := range msgs {
+        fmt.Println(m)
+
+        // Loops over the errors which would read
+        // "1.3 is greater than 1.2.3"
+        // "1.3 is less than 1.4"
+    }
+```
+
+# Contribute
+
+If you find an issue or want to contribute please file an [issue](https://github.com/Masterminds/semver/issues)
+or [create a pull request](https://github.com/Masterminds/semver/pulls).
diff --git a/backend/vendor/github.com/Masterminds/semver/appveyor.yml b/backend/vendor/github.com/Masterminds/semver/appveyor.yml
new file mode 100644
index 00000000..b2778df1
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/appveyor.yml
@@ -0,0 +1,44 @@
+version: build-{build}.{branch}
+
+clone_folder: C:\gopath\src\github.com\Masterminds\semver
+shallow_clone: true
+
+environment:
+  GOPATH: C:\gopath
+
+platform:
+  - x64
+
+install:
+  - go version
+  - go env
+  - go get -u gopkg.in/alecthomas/gometalinter.v1
+  - set PATH=%PATH%;%GOPATH%\bin
+  - gometalinter.v1.exe --install
+
+build_script:
+  - go install -v ./...
+
+test_script:
+  - "gometalinter.v1 \
+    --disable-all \
+    --enable deadcode \
+    --severity deadcode:error \
+    --enable gofmt \
+    --enable gosimple \
+    --enable ineffassign \
+    --enable misspell \
+    --enable vet \
+    --tests \
+    --vendor \
+    --deadline 60s \
+    ./... || exit_code=1"
+  - "gometalinter.v1 \
+    --disable-all \
+    --enable golint \
+    --vendor \
+    --deadline 60s \
+    ./... || :"
+  - go test -v
+
+deploy: off
diff --git a/backend/vendor/github.com/Masterminds/semver/collection.go b/backend/vendor/github.com/Masterminds/semver/collection.go
new file mode 100644
index 00000000..a7823589
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/collection.go
@@ -0,0 +1,24 @@
+package semver
+
+// Collection is a collection of Version instances and implements the sort
+// interface. See the sort package for more details.
+// https://golang.org/pkg/sort/
+type Collection []*Version
+
+// Len returns the length of a collection. The number of Version instances
+// on the slice.
+func (c Collection) Len() int {
+	return len(c)
+}
+
+// Less is needed for the sort interface to compare two Version objects on the
+// slice. If checks if one is less than the other.
+func (c Collection) Less(i, j int) bool {
+	return c[i].LessThan(c[j])
+}
+
+// Swap is needed for the sort interface to replace the Version objects
+// at two different positions in the slice.
+func (c Collection) Swap(i, j int) {
+	c[i], c[j] = c[j], c[i]
+}
diff --git a/backend/vendor/github.com/Masterminds/semver/constraints.go b/backend/vendor/github.com/Masterminds/semver/constraints.go
new file mode 100644
index 00000000..a41a6a7a
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/constraints.go
@@ -0,0 +1,426 @@
+package semver
+
+import (
+	"errors"
+	"fmt"
+	"regexp"
+	"strings"
+)
+
+// Constraints is one or more constraint that a semantic version can be
+// checked against.
+type Constraints struct {
+	constraints [][]*constraint
+}
+
+// NewConstraint returns a Constraints instance that a Version instance can
+// be checked against. If there is a parse error it will be returned.
+func NewConstraint(c string) (*Constraints, error) {
+
+	// Rewrite - ranges into a comparison operation.
+	c = rewriteRange(c)
+
+	ors := strings.Split(c, "||")
+	or := make([][]*constraint, len(ors))
+	for k, v := range ors {
+		cs := strings.Split(v, ",")
+		result := make([]*constraint, len(cs))
+		for i, s := range cs {
+			pc, err := parseConstraint(s)
+			if err != nil {
+				return nil, err
+			}
+
+			result[i] = pc
+		}
+		or[k] = result
+	}
+
+	o := &Constraints{constraints: or}
+	return o, nil
+}
+
+// Check tests if a version satisfies the constraints.
+func (cs Constraints) Check(v *Version) bool {
+	// loop over the ORs and check the inner ANDs
+	for _, o := range cs.constraints {
+		joy := true
+		for _, c := range o {
+			if !c.check(v) {
+				joy = false
+				break
+			}
+		}
+
+		if joy {
+			return true
+		}
+	}
+
+	return false
+}
+
+// Validate checks if a version satisfies a constraint. If not a slice of
+// reasons for the failure are returned in addition to a bool.
+func (cs Constraints) Validate(v *Version) (bool, []error) {
+	// loop over the ORs and check the inner ANDs
+	var e []error
+	for _, o := range cs.constraints {
+		joy := true
+		for _, c := range o {
+			if !c.check(v) {
+				em := fmt.Errorf(c.msg, v, c.orig)
+				e = append(e, em)
+				joy = false
+			}
+		}
+
+		if joy {
+			return true, []error{}
+		}
+	}
+
+	return false, e
+}
+
+var constraintOps map[string]cfunc
+var constraintMsg map[string]string
+var constraintRegex *regexp.Regexp
+
+func init() {
+	constraintOps = map[string]cfunc{
+		"":   constraintTildeOrEqual,
+		"=":  constraintTildeOrEqual,
+		"!=": constraintNotEqual,
+		">":  constraintGreaterThan,
+		"<":  constraintLessThan,
+		">=": constraintGreaterThanEqual,
+		"=>": constraintGreaterThanEqual,
+		"<=": constraintLessThanEqual,
+		"=<": constraintLessThanEqual,
+		"~":  constraintTilde,
+		"~>": constraintTilde,
+		"^":  constraintCaret,
+	}
+
+	constraintMsg = map[string]string{
+		"":   "%s is not equal to %s",
+		"=":  "%s is not equal to %s",
+		"!=": "%s is equal to %s",
+		">":  "%s is less than or equal to %s",
+		"<":  "%s is greater than or equal to %s",
+		">=": "%s is less than %s",
+		"=>": "%s is less than %s",
+		"<=": "%s is greater than %s",
+		"=<": "%s is greater than %s",
+		"~":  "%s does not have same major and minor version as %s",
+		"~>": "%s does not have same major and minor version as %s",
+		"^":  "%s does not have same major version as %s",
+	}
+
+	ops := make([]string, 0, len(constraintOps))
+	for k := range constraintOps {
+		ops = append(ops, regexp.QuoteMeta(k))
+	}
+
+	constraintRegex = regexp.MustCompile(fmt.Sprintf(
+		`^\s*(%s)\s*(%s)\s*$`,
+		strings.Join(ops, "|"),
+		cvRegex))
+
+	constraintRangeRegex = regexp.MustCompile(fmt.Sprintf(
+		`\s*(%s)\s+-\s+(%s)\s*`,
+		cvRegex, cvRegex))
+}
+
+// An individual constraint
+type constraint struct {
+	// The callback function for the restraint. It performs the logic for
+	// the constraint.
+	function cfunc
+
+	msg string
+
+	// The version used in the constraint check. For example, if a constraint
+	// is '<= 2.0.0' the con a version instance representing 2.0.0.
+	con *Version
+
+	// The original parsed version (e.g., 4.x from != 4.x)
+	orig string
+
+	// When an x is used as part of the version (e.g., 1.x)
+	minorDirty bool
+	dirty      bool
+	patchDirty bool
+}
+
+// Check if a version meets the constraint
+func (c *constraint) check(v *Version) bool {
+	return c.function(v, c)
+}
+
+type cfunc func(v *Version, c *constraint) bool
+
+func parseConstraint(c string) (*constraint, error) {
+	m := constraintRegex.FindStringSubmatch(c)
+	if m == nil {
+		return nil, fmt.Errorf("improper constraint: %s", c)
+	}
+
+	ver := m[2]
+	orig := ver
+	minorDirty := false
+	patchDirty := false
+	dirty := false
+	if isX(m[3]) {
+		ver = "0.0.0"
+		dirty = true
+	} else if isX(strings.TrimPrefix(m[4], ".")) || m[4] == "" {
+		minorDirty = true
+		dirty = true
+		ver = fmt.Sprintf("%s.0.0%s", m[3], m[6])
+	} else if isX(strings.TrimPrefix(m[5], ".")) {
+		dirty = true
+		patchDirty = true
+		ver = fmt.Sprintf("%s%s.0%s", m[3], m[4], m[6])
+	}
+
+	con, err := NewVersion(ver)
+	if err != nil {
+
+		// The constraintRegex should catch any regex parsing errors. So,
+		// we should never get here.
+		return nil, errors.New("constraint Parser Error")
+	}
+
+	cs := &constraint{
+		function:   constraintOps[m[1]],
+		msg:        constraintMsg[m[1]],
+		con:        con,
+		orig:       orig,
+		minorDirty: minorDirty,
+		patchDirty: patchDirty,
+		dirty:      dirty,
+	}
+	return cs, nil
+}
+
+// Constraint functions
+func constraintNotEqual(v *Version, c *constraint) bool {
+	if c.dirty {
+
+		// If there is a pre-release on the version but the constraint isn't looking
+		// for them assume that pre-releases are not compatible. See issue 21 for
+		// more details.
+		if v.Prerelease() != "" && c.con.Prerelease() == "" {
+			return false
+		}
+
+		if c.con.Major() != v.Major() {
+			return true
+		}
+		if c.con.Minor() != v.Minor() && !c.minorDirty {
+			return true
+		} else if c.minorDirty {
+			return false
+		}
+
+		return false
+	}
+
+	return !v.Equal(c.con)
+}
+
+func constraintGreaterThan(v *Version, c *constraint) bool {
+
+	// An edge case the constraint is 0.0.0 and the version is 0.0.0-someprerelease
+	// exists. This that case.
+	if !isNonZero(c.con) && isNonZero(v) {
+		return true
+	}
+
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	return v.Compare(c.con) == 1
+}
+
+func constraintLessThan(v *Version, c *constraint) bool {
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	if !c.dirty {
+		return v.Compare(c.con) < 0
+	}
+
+	if v.Major() > c.con.Major() {
+		return false
+	} else if v.Minor() > c.con.Minor() && !c.minorDirty {
+		return false
+	}
+
+	return true
+}
+
+func constraintGreaterThanEqual(v *Version, c *constraint) bool {
+	// An edge case the constraint is 0.0.0 and the version is 0.0.0-someprerelease
+	// exists. This that case.
+	if !isNonZero(c.con) && isNonZero(v) {
+		return true
+	}
+
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	return v.Compare(c.con) >= 0
+}
+
+func constraintLessThanEqual(v *Version, c *constraint) bool {
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	if !c.dirty {
+		return v.Compare(c.con) <= 0
+	}
+
+	if v.Major() > c.con.Major() {
+		return false
+	} else if v.Minor() > c.con.Minor() && !c.minorDirty {
+		return false
+	}
+
+	return true
+}
+
+// ~*, ~>* --> >= 0.0.0 (any)
+// ~2, ~2.x, ~2.x.x, ~>2, ~>2.x ~>2.x.x --> >=2.0.0, <3.0.0
+// ~2.0, ~2.0.x, ~>2.0, ~>2.0.x --> >=2.0.0, <2.1.0
+// ~1.2, ~1.2.x, ~>1.2, ~>1.2.x --> >=1.2.0, <1.3.0
+// ~1.2.3, ~>1.2.3 --> >=1.2.3, <1.3.0
+// ~1.2.0, ~>1.2.0 --> >=1.2.0, <1.3.0
+func constraintTilde(v *Version, c *constraint) bool {
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	if v.LessThan(c.con) {
+		return false
+	}
+
+	// ~0.0.0 is a special case where all constraints are accepted. It's
+	// equivalent to >= 0.0.0.
+	if c.con.Major() == 0 && c.con.Minor() == 0 && c.con.Patch() == 0 &&
+		!c.minorDirty && !c.patchDirty {
+		return true
+	}
+
+	if v.Major() != c.con.Major() {
+		return false
+	}
+
+	if v.Minor() != c.con.Minor() && !c.minorDirty {
+		return false
+	}
+
+	return true
+}
+
+// When there is a .x (dirty) status it automatically opts in to ~. Otherwise
+// it's a straight =
+func constraintTildeOrEqual(v *Version, c *constraint) bool {
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	if c.dirty {
+		c.msg = constraintMsg["~"]
+		return constraintTilde(v, c)
+	}
+
+	return v.Equal(c.con)
+}
+
+// ^* --> (any)
+// ^2, ^2.x, ^2.x.x --> >=2.0.0, <3.0.0
+// ^2.0, ^2.0.x --> >=2.0.0, <3.0.0
+// ^1.2, ^1.2.x --> >=1.2.0, <2.0.0
+// ^1.2.3 --> >=1.2.3, <2.0.0
+// ^1.2.0 --> >=1.2.0, <2.0.0
+func constraintCaret(v *Version, c *constraint) bool {
+	// If there is a pre-release on the version but the constraint isn't looking
+	// for them assume that pre-releases are not compatible. See issue 21 for
+	// more details.
+	if v.Prerelease() != "" && c.con.Prerelease() == "" {
+		return false
+	}
+
+	if v.LessThan(c.con) {
+		return false
+	}
+
+	if v.Major() != c.con.Major() {
+		return false
+	}
+
+	return true
+}
+
+var constraintRangeRegex *regexp.Regexp
+
+const cvRegex string = `v?([0-9|x|X|\*]+)(\.[0-9|x|X|\*]+)?(\.[0-9|x|X|\*]+)?` +
+	`(-([0-9A-Za-z\-]+(\.[0-9A-Za-z\-]+)*))?` +
+	`(\+([0-9A-Za-z\-]+(\.[0-9A-Za-z\-]+)*))?`
+
+func isX(x string) bool {
+	switch x {
+	case "x", "*", "X":
+		return true
+	default:
+		return false
+	}
+}
+
+func rewriteRange(i string) string {
+	m := constraintRangeRegex.FindAllStringSubmatch(i, -1)
+	if m == nil {
+		return i
+	}
+	o := i
+	for _, v := range m {
+		t := fmt.Sprintf(">= %s, <= %s", v[1], v[11])
+		o = strings.Replace(o, v[0], t, 1)
+	}
+
+	return o
+}
+
+// Detect if a version is not zero (0.0.0)
+func isNonZero(v *Version) bool {
+	if v.Major() != 0 || v.Minor() != 0 || v.Patch() != 0 || v.Prerelease() != "" {
+		return true
+	}
+
+	return false
+}
diff --git a/backend/vendor/github.com/Masterminds/semver/doc.go b/backend/vendor/github.com/Masterminds/semver/doc.go
new file mode 100644
index 00000000..e00f65eb
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/doc.go
@@ -0,0 +1,115 @@
+/*
+Package semver provides the ability to work with Semantic Versions (http://semver.org) in Go.
+
+Specifically it provides the ability to:
+
+    * Parse semantic versions
+    * Sort semantic versions
+    * Check if a semantic version fits within a set of constraints
+    * Optionally work with a `v` prefix
+
+Parsing Semantic Versions
+
+To parse a semantic version use the `NewVersion` function. For example,
+
+    v, err := semver.NewVersion("1.2.3-beta.1+build345")
+
+If there is an error the version wasn't parseable. The version object has methods
+to get the parts of the version, compare it to other versions, convert the
+version back into a string, and get the original string. For more details
+please see the documentation at https://godoc.org/github.com/Masterminds/semver.
+
+Sorting Semantic Versions
+
+A set of versions can be sorted using the `sort` package from the standard library.
+For example,
+
+    raw := []string{"1.2.3", "1.0", "1.3", "2", "0.4.2",}
+    vs := make([]*semver.Version, len(raw))
+	for i, r := range raw {
+		v, err := semver.NewVersion(r)
+		if err != nil {
+			t.Errorf("Error parsing version: %s", err)
+		}
+
+		vs[i] = v
+	}
+
+	sort.Sort(semver.Collection(vs))
+
+Checking Version Constraints
+
+Checking a version against version constraints is one of the most featureful
+parts of the package.
+
+    c, err := semver.NewConstraint(">= 1.2.3")
+    if err != nil {
+        // Handle constraint not being parseable.
+    }
+
+    v, _ := semver.NewVersion("1.3")
+    if err != nil {
+        // Handle version not being parseable.
+    }
+    // Check if the version meets the constraints. The a variable will be true.
+    a := c.Check(v)
+
+Basic Comparisons
+
+There are two elements to the comparisons. First, a comparison string is a list
+of comma separated and comparisons. These are then separated by || separated or
+comparisons. For example, `">= 1.2, < 3.0.0 || >= 4.2.3"` is looking for a
+comparison that's greater than or equal to 1.2 and less than 3.0.0 or is
+greater than or equal to 4.2.3.
+
+The basic comparisons are:
+
+    * `=`: equal (aliased to no operator)
+    * `!=`: not equal
+    * `>`: greater than
+    * `<`: less than
+    * `>=`: greater than or equal to
+    * `<=`: less than or equal to
+
+Hyphen Range Comparisons
+
+There are multiple methods to handle ranges and the first is hyphens ranges.
+These look like:
+
+    * `1.2 - 1.4.5` which is equivalent to `>= 1.2, <= 1.4.5`
+    * `2.3.4 - 4.5` which is equivalent to `>= 2.3.4, <= 4.5`
+
+Wildcards In Comparisons
+
+The `x`, `X`, and `*` characters can be used as a wildcard character. This works
+for all comparison operators. When used on the `=` operator it falls
+back to the pack level comparison (see tilde below). For example,
+
+    * `1.2.x` is equivalent to `>= 1.2.0, < 1.3.0`
+    * `>= 1.2.x` is equivalent to `>= 1.2.0`
+    * `<= 2.x` is equivalent to `<= 3`
+    * `*` is equivalent to `>= 0.0.0`
+
+Tilde Range Comparisons (Patch)
+
+The tilde (`~`) comparison operator is for patch level ranges when a minor
+version is specified and major level changes when the minor number is missing.
+For example,
+
+    * `~1.2.3` is equivalent to `>= 1.2.3, < 1.3.0`
+    * `~1` is equivalent to `>= 1, < 2`
+    * `~2.3` is equivalent to `>= 2.3, < 2.4`
+    * `~1.2.x` is equivalent to `>= 1.2.0, < 1.3.0`
+    * `~1.x` is equivalent to `>= 1, < 2`
+
+Caret Range Comparisons (Major)
+
+The caret (`^`) comparison operator is for major level changes. This is useful
+when comparisons of API versions as a major change is API breaking. For example,
+
+    * `^1.2.3` is equivalent to `>= 1.2.3, < 2.0.0`
+    * `^1.2.x` is equivalent to `>= 1.2.0, < 2.0.0`
+    * `^2.3` is equivalent to `>= 2.3, < 3`
+    * `^2.x` is equivalent to `>= 2.0.0, < 3`
+*/
+package semver
diff --git a/backend/vendor/github.com/Masterminds/semver/version.go b/backend/vendor/github.com/Masterminds/semver/version.go
new file mode 100644
index 00000000..9d22ea63
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/semver/version.go
@@ -0,0 +1,421 @@
+package semver
+
+import (
+	"bytes"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"regexp"
+	"strconv"
+	"strings"
+)
+
+// The compiled version of the regex created at init() is cached here so it
+// only needs to be created once.
+var versionRegex *regexp.Regexp
+var validPrereleaseRegex *regexp.Regexp
+
+var (
+	// ErrInvalidSemVer is returned a version is found to be invalid when
+	// being parsed.
+	ErrInvalidSemVer = errors.New("Invalid Semantic Version")
+
+	// ErrInvalidMetadata is returned when the metadata is an invalid format
+	ErrInvalidMetadata = errors.New("Invalid Metadata string")
+
+	// ErrInvalidPrerelease is returned when the pre-release is an invalid format
+	ErrInvalidPrerelease = errors.New("Invalid Prerelease string")
+)
+
+// SemVerRegex is the regular expression used to parse a semantic version.
+const SemVerRegex string = `v?([0-9]+)(\.[0-9]+)?(\.[0-9]+)?` +
+	`(-([0-9A-Za-z\-]+(\.[0-9A-Za-z\-]+)*))?` +
+	`(\+([0-9A-Za-z\-]+(\.[0-9A-Za-z\-]+)*))?`
+
+// ValidPrerelease is the regular expression which validates
+// both prerelease and metadata values.
+const ValidPrerelease string = `^([0-9A-Za-z\-]+(\.[0-9A-Za-z\-]+)*)`
+
+// Version represents a single semantic version.
+type Version struct {
+	major, minor, patch int64
+	pre                 string
+	metadata            string
+	original            string
+}
+
+func init() {
+	versionRegex = regexp.MustCompile("^" + SemVerRegex + "$")
+	validPrereleaseRegex = regexp.MustCompile(ValidPrerelease)
+}
+
+// NewVersion parses a given version and returns an instance of Version or
+// an error if unable to parse the version.
+func NewVersion(v string) (*Version, error) {
+	m := versionRegex.FindStringSubmatch(v)
+	if m == nil {
+		return nil, ErrInvalidSemVer
+	}
+
+	sv := &Version{
+		metadata: m[8],
+		pre:      m[5],
+		original: v,
+	}
+
+	var temp int64
+	temp, err := strconv.ParseInt(m[1], 10, 64)
+	if err != nil {
+		return nil, fmt.Errorf("Error parsing version segment: %s", err)
+	}
+	sv.major = temp
+
+	if m[2] != "" {
+		temp, err = strconv.ParseInt(strings.TrimPrefix(m[2], "."), 10, 64)
+		if err != nil {
+			return nil, fmt.Errorf("Error parsing version segment: %s", err)
+		}
+		sv.minor = temp
+	} else {
+		sv.minor = 0
+	}
+
+	if m[3] != "" {
+		temp, err = strconv.ParseInt(strings.TrimPrefix(m[3], "."), 10, 64)
+		if err != nil {
+			return nil, fmt.Errorf("Error parsing version segment: %s", err)
+		}
+		sv.patch = temp
+	} else {
+		sv.patch = 0
+	}
+
+	return sv, nil
+}
+
+// MustParse parses a given version and panics on error.
+func MustParse(v string) *Version {
+	sv, err := NewVersion(v)
+	if err != nil {
+		panic(err)
+	}
+	return sv
+}
+
+// String converts a Version object to a string.
+// Note, if the original version contained a leading v this version will not.
+// See the Original() method to retrieve the original value. Semantic Versions
+// don't contain a leading v per the spec. Instead it's optional on
+// impelementation.
+func (v *Version) String() string {
+	var buf bytes.Buffer
+
+	fmt.Fprintf(&buf, "%d.%d.%d", v.major, v.minor, v.patch)
+	if v.pre != "" {
+		fmt.Fprintf(&buf, "-%s", v.pre)
+	}
+	if v.metadata != "" {
+		fmt.Fprintf(&buf, "+%s", v.metadata)
+	}
+
+	return buf.String()
+}
+
+// Original returns the original value passed in to be parsed.
+func (v *Version) Original() string {
+	return v.original
+}
+
+// Major returns the major version.
+func (v *Version) Major() int64 {
+	return v.major
+}
+
+// Minor returns the minor version.
+func (v *Version) Minor() int64 {
+	return v.minor
+}
+
+// Patch returns the patch version.
+func (v *Version) Patch() int64 {
+	return v.patch
+}
+
+// Prerelease returns the pre-release version.
+func (v *Version) Prerelease() string {
+	return v.pre
+}
+
+// Metadata returns the metadata on the version.
+func (v *Version) Metadata() string {
+	return v.metadata
+}
+
+// originalVPrefix returns the original 'v' prefix if any.
+func (v *Version) originalVPrefix() string {
+
+	// Note, only lowercase v is supported as a prefix by the parser.
+	if v.original != "" && v.original[:1] == "v" {
+		return v.original[:1]
+	}
+	return ""
+}
+
+// IncPatch produces the next patch version.
+// If the current version does not have prerelease/metadata information,
+// it unsets metadata and prerelease values, increments patch number.
+// If the current version has any of prerelease or metadata information,
+// it unsets both values and keeps curent patch value
+func (v Version) IncPatch() Version {
+	vNext := v
+	// according to http://semver.org/#spec-item-9
+	// Pre-release versions have a lower precedence than the associated normal version.
+	// according to http://semver.org/#spec-item-10
+	// Build metadata SHOULD be ignored when determining version precedence.
+	if v.pre != "" {
+		vNext.metadata = ""
+		vNext.pre = ""
+	} else {
+		vNext.metadata = ""
+		vNext.pre = ""
+		vNext.patch = v.patch + 1
+	}
+	vNext.original = v.originalVPrefix() + "" + vNext.String()
+	return vNext
+}
+
+// IncMinor produces the next minor version.
+// Sets patch to 0.
+// Increments minor number.
+// Unsets metadata.
+// Unsets prerelease status.
+func (v Version) IncMinor() Version {
+	vNext := v
+	vNext.metadata = ""
+	vNext.pre = ""
+	vNext.patch = 0
+	vNext.minor = v.minor + 1
+	vNext.original = v.originalVPrefix() + "" + vNext.String()
+	return vNext
+}
+
+// IncMajor produces the next major version.
+// Sets patch to 0.
+// Sets minor to 0.
+// Increments major number.
+// Unsets metadata.
+// Unsets prerelease status.
+func (v Version) IncMajor() Version {
+	vNext := v
+	vNext.metadata = ""
+	vNext.pre = ""
+	vNext.patch = 0
+	vNext.minor = 0
+	vNext.major = v.major + 1
+	vNext.original = v.originalVPrefix() + "" + vNext.String()
+	return vNext
+}
+
+// SetPrerelease defines the prerelease value.
+// Value must not include the required 'hypen' prefix.
+func (v Version) SetPrerelease(prerelease string) (Version, error) {
+	vNext := v
+	if len(prerelease) > 0 && !validPrereleaseRegex.MatchString(prerelease) {
+		return vNext, ErrInvalidPrerelease
+	}
+	vNext.pre = prerelease
+	vNext.original = v.originalVPrefix() + "" + vNext.String()
+	return vNext, nil
+}
+
+// SetMetadata defines metadata value.
+// Value must not include the required 'plus' prefix.
+func (v Version) SetMetadata(metadata string) (Version, error) {
+	vNext := v
+	if len(metadata) > 0 && !validPrereleaseRegex.MatchString(metadata) {
+		return vNext, ErrInvalidMetadata
+	}
+	vNext.metadata = metadata
+	vNext.original = v.originalVPrefix() + "" + vNext.String()
+	return vNext, nil
+}
+
+// LessThan tests if one version is less than another one.
+func (v *Version) LessThan(o *Version) bool {
+	return v.Compare(o) < 0
+}
+
+// GreaterThan tests if one version is greater than another one.
+func (v *Version) GreaterThan(o *Version) bool {
+	return v.Compare(o) > 0
+}
+
+// Equal tests if two versions are equal to each other.
+// Note, versions can be equal with different metadata since metadata
+// is not considered part of the comparable version.
+func (v *Version) Equal(o *Version) bool {
+	return v.Compare(o) == 0
+}
+
+// Compare compares this version to another one. It returns -1, 0, or 1 if
+// the version smaller, equal, or larger than the other version.
+//
+// Versions are compared by X.Y.Z. Build metadata is ignored. Prerelease is
+// lower than the version without a prerelease.
+func (v *Version) Compare(o *Version) int {
+	// Compare the major, minor, and patch version for differences. If a
+	// difference is found return the comparison.
+	if d := compareSegment(v.Major(), o.Major()); d != 0 {
+		return d
+	}
+	if d := compareSegment(v.Minor(), o.Minor()); d != 0 {
+		return d
+	}
+	if d := compareSegment(v.Patch(), o.Patch()); d != 0 {
+		return d
+	}
+
+	// At this point the major, minor, and patch versions are the same.
+	ps := v.pre
+	po := o.Prerelease()
+
+	if ps == "" && po == "" {
+		return 0
+	}
+	if ps == "" {
+		return 1
+	}
+	if po == "" {
+		return -1
+	}
+
+	return comparePrerelease(ps, po)
+}
+
+// UnmarshalJSON implements JSON.Unmarshaler interface.
+func (v *Version) UnmarshalJSON(b []byte) error {
+	var s string
+	if err := json.Unmarshal(b, &s); err != nil {
+		return err
+	}
+	temp, err := NewVersion(s)
+	if err != nil {
+		return err
+	}
+	v.major = temp.major
+	v.minor = temp.minor
+	v.patch = temp.patch
+	v.pre = temp.pre
+	v.metadata = temp.metadata
+	v.original = temp.original
+	temp = nil
+	return nil
+}
+
+// MarshalJSON implements JSON.Marshaler interface.
+func (v *Version) MarshalJSON() ([]byte, error) {
+	return json.Marshal(v.String())
+}
+
+func compareSegment(v, o int64) int {
+	if v < o {
+		return -1
+	}
+	if v > o {
+		return 1
+	}
+
+	return 0
+}
+
+func comparePrerelease(v, o string) int {
+
+	// split the prelease versions by their part. The separator, per the spec,
+	// is a .
+	sparts := strings.Split(v, ".")
+	oparts := strings.Split(o, ".")
+
+	// Find the longer length of the parts to know how many loop iterations to
+	// go through.
+	slen := len(sparts)
+	olen := len(oparts)
+
+	l := slen
+	if olen > slen {
+		l = olen
+	}
+
+	// Iterate over each part of the prereleases to compare the differences.
+	for i := 0; i < l; i++ {
+		// Since the lentgh of the parts can be different we need to create
+		// a placeholder. This is to avoid out of bounds issues.
+		stemp := ""
+		if i < slen {
+			stemp = sparts[i]
+		}
+
+		otemp := ""
+		if i < olen {
+			otemp = oparts[i]
+		}
+
+		d := comparePrePart(stemp, otemp)
+		if d != 0 {
+			return d
+		}
+	}
+
+	// Reaching here means two versions are of equal value but have different
+	// metadata (the part following a +). They are not identical in string form
+	// but the version comparison finds them to be equal.
+	return 0
+}
+
+func comparePrePart(s, o string) int {
+	// Fastpath if they are equal
+	if s == o {
+		return 0
+	}
+
+	// When s or o are empty we can use the other in an attempt to determine
+	// the response.
+	if s == "" {
+		if o != "" {
+			return -1
+		}
+		return 1
+	}
+
+	if o == "" {
+		if s != "" {
+			return 1
+		}
+		return -1
+	}
+
+	// When comparing strings "99" is greater than "103". To handle
+	// cases like this we need to detect numbers and compare them.
+
+	oi, n1 := strconv.ParseInt(o, 10, 64)
+	si, n2 := strconv.ParseInt(s, 10, 64)
+
+	// The case where both are strings compare the strings
+	if n1 != nil && n2 != nil {
+		if s > o {
+			return 1
+		}
+		return -1
+	} else if n1 != nil {
+		// o is a string and s is a number
+		return -1
+	} else if n2 != nil {
+		// s is a string and o is a number
+		return 1
+	}
+	// Both are numbers
+	if si > oi {
+		return 1
+	}
+	return -1
+
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/.gitignore b/backend/vendor/github.com/Masterminds/sprig/.gitignore
new file mode 100644
index 00000000..5e3002f8
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/.gitignore
@@ -0,0 +1,2 @@
+vendor/
+/.glide
diff --git a/backend/vendor/github.com/Masterminds/sprig/.travis.yml b/backend/vendor/github.com/Masterminds/sprig/.travis.yml
new file mode 100644
index 00000000..482aa3cd
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/.travis.yml
@@ -0,0 +1,24 @@
+language: go
+
+go:
+  - 1.9.x
+  - 1.10.x
+  - 1.11.x
+  - tip
+
+# Setting sudo access to false will let Travis CI use containers rather than
+# VMs to run the tests. For more details see:
+# - http://docs.travis-ci.com/user/workers/container-based-infrastructure/
+# - http://docs.travis-ci.com/user/workers/standard-infrastructure/
+sudo: false
+
+script:
+  - make setup test
+
+notifications:
+  webhooks:
+    urls:
+      - https://webhooks.gitter.im/e/06e3328629952dabe3e0
+    on_success: change  # options: [always|never|change] default: always
+    on_failure: always  # options: [always|never|change] default: always
+    on_start: never     # options: [always|never|change] default: always
diff --git a/backend/vendor/github.com/Masterminds/sprig/CHANGELOG.md b/backend/vendor/github.com/Masterminds/sprig/CHANGELOG.md
new file mode 100644
index 00000000..44593713
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/CHANGELOG.md
@@ -0,0 +1,153 @@
+# Changelog
+
+## Release 2.15.0 (2018-04-02)
+
+### Added
+
+- #68 and #69: Add json helpers to docs (thanks @arunvelsriram)
+- #66: Add ternary function (thanks @binoculars)
+- #67: Allow keys function to take multiple dicts (thanks @binoculars)
+- #89: Added sha1sum to crypto function (thanks @benkeil)
+- #81: Allow customizing Root CA that used by genSignedCert (thanks @chenzhiwei)
+- #92: Add travis testing for go 1.10
+- #93: Adding appveyor config for windows testing
+
+### Changed
+
+- #90: Updating to more recent dependencies
+- #73: replace satori/go.uuid with google/uuid (thanks @petterw)
+
+### Fixed
+
+- #76: Fixed documentation typos (thanks @Thiht)
+- Fixed rounding issue on the `ago` function. Note, the removes support for Go 1.8 and older
+
+## Release 2.14.1 (2017-12-01)
+
+### Fixed
+
+- #60: Fix typo in function name documentation (thanks @neil-ca-moore)
+- #61: Removing line with {{ due to blocking github pages genertion
+- #64: Update the list functions to handle int, string, and other slices for compatibility
+
+## Release 2.14.0 (2017-10-06)
+
+This new version of Sprig adds a set of functions for generating and working with SSL certificates.
+
+- `genCA` generates an SSL Certificate Authority
+- `genSelfSignedCert` generates an SSL self-signed certificate
+- `genSignedCert` generates an SSL certificate and key based on a given CA
+
+## Release 2.13.0 (2017-09-18)
+
+This release adds new functions, including:
+
+- `regexMatch`, `regexFindAll`, `regexFind`, `regexReplaceAll`, `regexReplaceAllLiteral`, and `regexSplit` to work with regular expressions
+- `floor`, `ceil`, and `round` math functions
+- `toDate` converts a string to a date
+- `nindent` is just like `indent` but also prepends a new line
+- `ago` returns the time from `time.Now`
+
+### Added
+
+- #40: Added basic regex functionality (thanks @alanquillin)
+- #41: Added ceil floor and round functions (thanks @alanquillin)
+- #48: Added toDate function (thanks @andreynering)
+- #50: Added nindent function (thanks @binoculars)
+- #46: Added ago function (thanks @slayer)
+
+### Changed
+
+- #51: Updated godocs to include new string functions (thanks @curtisallen)
+- #49: Added ability to merge multiple dicts (thanks @binoculars)
+
+## Release 2.12.0 (2017-05-17)
+
+- `snakecase`, `camelcase`, and `shuffle` are three new string functions
+- `fail` allows you to bail out of a template render when conditions are not met
+
+## Release 2.11.0 (2017-05-02)
+
+- Added `toJson` and `toPrettyJson`
+- Added `merge`
+- Refactored documentation
+
+## Release 2.10.0 (2017-03-15)
+
+- Added `semver` and `semverCompare` for Semantic Versions
+- `list` replaces `tuple`
+- Fixed issue with `join`
+- Added `first`, `last`, `intial`, `rest`, `prepend`, `append`, `toString`, `toStrings`, `sortAlpha`, `reverse`, `coalesce`, `pluck`, `pick`, `compact`, `keys`, `omit`, `uniq`, `has`, `without`
+
+## Release 2.9.0 (2017-02-23)
+
+- Added `splitList` to split a list
+- Added crypto functions of `genPrivateKey` and `derivePassword`
+
+## Release 2.8.0 (2016-12-21)
+
+- Added access to several path functions (`base`, `dir`, `clean`, `ext`, and `abs`)
+- Added functions for _mutating_ dictionaries (`set`, `unset`, `hasKey`)
+
+## Release 2.7.0 (2016-12-01)
+
+- Added `sha256sum` to generate a hash of an input
+- Added functions to convert a numeric or string to `int`, `int64`, `float64`
+
+## Release 2.6.0 (2016-10-03)
+
+- Added a `uuidv4` template function for generating UUIDs inside of a template.
+
+## Release 2.5.0 (2016-08-19)
+
+- New `trimSuffix`, `trimPrefix`, `hasSuffix`, and `hasPrefix` functions
+- New aliases have been added for a few functions that didn't follow the naming conventions (`trimAll` and `abbrevBoth`)
+- `trimall` and `abbrevboth` (notice the case) are deprecated and will be removed in 3.0.0
+
+## Release 2.4.0 (2016-08-16)
+
+- Adds two functions: `until` and `untilStep`
+
+## Release 2.3.0 (2016-06-21)
+
+- cat: Concatenate strings with whitespace separators.
+- replace: Replace parts of a string: `replace " " "-" "Me First"` renders "Me-First"
+- plural: Format plurals: `len "foo" | plural "one foo" "many foos"` renders "many foos"
+- indent: Indent blocks of text in a way that is sensitive to "\n" characters.
+
+## Release 2.2.0 (2016-04-21)
+
+- Added a `genPrivateKey` function (Thanks @bacongobbler)
+
+## Release 2.1.0 (2016-03-30)
+
+- `default` now prints the default value when it does not receive a value down the pipeline. It is much safer now to do `{{.Foo | default "bar"}}`.
+- Added accessors for "hermetic" functions. These return only functions that, when given the same input, produce the same output.
+
+## Release 2.0.0 (2016-03-29)
+
+Because we switched from `int` to `int64` as the return value for all integer math functions, the library's major version number has been incremented.
+
+- `min` complements `max` (formerly `biggest`)
+- `empty` indicates that a value is the empty value for its type
+- `tuple` creates a tuple inside of a template: `{{$t := tuple "a", "b" "c"}}`
+- `dict` creates a dictionary inside of a template `{{$d := dict "key1" "val1" "key2" "val2"}}` 
+- Date formatters have been added for HTML dates (as used in `date` input fields)
+- Integer math functions can convert from a number of types, including `string` (via `strconv.ParseInt`).
+
+## Release 1.2.0 (2016-02-01)
+
+- Added quote and squote
+- Added b32enc and b32dec
+- add now takes varargs
+- biggest now takes varargs
+
+## Release 1.1.0 (2015-12-29)
+
+- Added #4: Added contains function. strings.Contains, but with the arguments
+  switched to simplify common pipelines. (thanks krancour)
+- Added Travis-CI testing support
+
+## Release 1.0.0 (2015-12-23)
+
+- Initial release
diff --git a/backend/vendor/github.com/Masterminds/sprig/LICENSE.txt b/backend/vendor/github.com/Masterminds/sprig/LICENSE.txt
new file mode 100644
index 00000000..5c95accc
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/LICENSE.txt
@@ -0,0 +1,20 @@
+Sprig
+Copyright (C) 2013 Masterminds
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/backend/vendor/github.com/Masterminds/sprig/Makefile b/backend/vendor/github.com/Masterminds/sprig/Makefile
new file mode 100644
index 00000000..63a93fdf
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/Makefile
@@ -0,0 +1,13 @@
+
+HAS_GLIDE := $(shell command -v glide;)
+
+.PHONY: test
+test:
+	go test -v .
+
+.PHONY: setup
+setup:
+ifndef HAS_GLIDE
+	go get -u github.com/Masterminds/glide
+endif
+	glide install
diff --git a/backend/vendor/github.com/Masterminds/sprig/README.md b/backend/vendor/github.com/Masterminds/sprig/README.md
new file mode 100644
index 00000000..25bf3d4f
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/README.md
@@ -0,0 +1,81 @@
+# Sprig: Template functions for Go templates
+[![Stability: Sustained](https://masterminds.github.io/stability/sustained.svg)](https://masterminds.github.io/stability/sustained.html)
+[![Build Status](https://travis-ci.org/Masterminds/sprig.svg?branch=master)](https://travis-ci.org/Masterminds/sprig)
+
+The Go language comes with a [built-in template
+language](http://golang.org/pkg/text/template/), but not
+very many template functions. This library provides a group of commonly
+used template functions.
+
+It is inspired by the template functions found in
+[Twig](http://twig.sensiolabs.org/documentation) and also in various
+JavaScript libraries, such as [underscore.js](http://underscorejs.org/).
+
+## Usage
+
+Template developers can read the [Sprig function documentation](http://masterminds.github.io/sprig/) to
+learn about the >100 template functions available.
+
+For Go developers wishing to include Sprig as a library in their programs,
+API documentation is available [at GoDoc.org](http://godoc.org/github.com/Masterminds/sprig), but
+read on for standard usage.
+
+### Load the Sprig library
+
+To load the Sprig `FuncMap`:
+
+```go
+
+import (
+  "github.com/Masterminds/sprig"
+  "html/template"
+)
+
+// This example illustrates that the FuncMap *must* be set before the
+// templates themselves are loaded.
+tpl := template.Must(
+  template.New("base").Funcs(sprig.FuncMap()).ParseGlob("*.html")
+)
+
+
+```
+
+### Call the functions inside of templates
+
+By convention, all functions are lowercase. This seems to follow the Go
+idiom for template functions (as opposed to template methods, which are
+TitleCase).
+
+
+Example:
+
+```
+{{ "hello!" | upper | repeat 5 }}
+```
+
+Produces:
+
+```
+HELLO!HELLO!HELLO!HELLO!HELLO!
+```
+
+## Principles:
+
+The following principles were used in deciding on which functions to add, and
+determining how to implement them.
+
+- Template functions should be used to build layout. Therefore, the following
+  types of operations are within the domain of template functions:
+  - Formatting
+  - Layout
+  - Simple type conversions
+  - Utilities that assist in handling common formatting and layout needs (e.g. arithmetic)
+- Template functions should not return errors unless there is no way to print
+  a sensible value. For example, converting a string to an integer should not
+  produce an error if conversion fails. Instead, it should display a default
+  value that can be displayed.
+- Simple math is necessary for grid layouts, pagers, and so on. Complex math
+  (anything other than arithmetic) should be done outside of templates.
+- Template functions only deal with the data passed into them. They never retrieve
+  data from a source.
+- Finally, do not override core Go template functions.
diff --git a/backend/vendor/github.com/Masterminds/sprig/appveyor.yml b/backend/vendor/github.com/Masterminds/sprig/appveyor.yml
new file mode 100644
index 00000000..d545a987
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/appveyor.yml
@@ -0,0 +1,26 @@
+
+version: build-{build}.{branch}
+
+clone_folder: C:\gopath\src\github.com\Masterminds\sprig
+shallow_clone: true
+
+environment:
+  GOPATH: C:\gopath
+
+platform:
+  - x64
+
+install:
+  - go get -u github.com/Masterminds/glide
+  - set PATH=%GOPATH%\bin;%PATH%
+  - go version
+  - go env
+
+build_script:
+  - glide install
+  - go install ./...
+
+test_script:
+  - go test -v
+
+deploy: off
diff --git a/backend/vendor/github.com/Masterminds/sprig/crypto.go b/backend/vendor/github.com/Masterminds/sprig/crypto.go
new file mode 100644
index 00000000..dc6579d6
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/crypto.go
@@ -0,0 +1,435 @@
+package sprig
+
+import (
+	"bytes"
+	"crypto/dsa"
+	"crypto/ecdsa"
+	"crypto/elliptic"
+	"crypto/hmac"
+	"crypto/rand"
+	"crypto/rsa"
+	"crypto/sha1"
+	"crypto/sha256"
+	"crypto/x509"
+	"crypto/x509/pkix"
+	"encoding/asn1"
+	"encoding/base64"
+	"encoding/binary"
+	"encoding/hex"
+	"encoding/pem"
+	"errors"
+	"fmt"
+	"math/big"
+	"net"
+	"time"
+
+	"github.com/google/uuid"
+	"golang.org/x/crypto/scrypt"
+)
+
+func sha256sum(input string) string {
+	hash := sha256.Sum256([]byte(input))
+	return hex.EncodeToString(hash[:])
+}
+
+func sha1sum(input string) string {
+	hash := sha1.Sum([]byte(input))
+	return hex.EncodeToString(hash[:])
+}
+
+// uuidv4 provides a safe and secure UUID v4 implementation
+func uuidv4() string {
+	return fmt.Sprintf("%s", uuid.New())
+}
+
+var master_password_seed = "com.lyndir.masterpassword"
+
+var password_type_templates = map[string][][]byte{
+	"maximum": {[]byte("anoxxxxxxxxxxxxxxxxx"), []byte("axxxxxxxxxxxxxxxxxno")},
+	"long": {[]byte("CvcvnoCvcvCvcv"), []byte("CvcvCvcvnoCvcv"), []byte("CvcvCvcvCvcvno"), []byte("CvccnoCvcvCvcv"), []byte("CvccCvcvnoCvcv"),
+		[]byte("CvccCvcvCvcvno"), []byte("CvcvnoCvccCvcv"), []byte("CvcvCvccnoCvcv"), []byte("CvcvCvccCvcvno"), []byte("CvcvnoCvcvCvcc"),
+		[]byte("CvcvCvcvnoCvcc"), []byte("CvcvCvcvCvccno"), []byte("CvccnoCvccCvcv"), []byte("CvccCvccnoCvcv"), []byte("CvccCvccCvcvno"),
+		[]byte("CvcvnoCvccCvcc"), []byte("CvcvCvccnoCvcc"), []byte("CvcvCvccCvccno"), []byte("CvccnoCvcvCvcc"), []byte("CvccCvcvnoCvcc"),
+		[]byte("CvccCvcvCvccno")},
+	"medium": {[]byte("CvcnoCvc"), []byte("CvcCvcno")},
+	"short":  {[]byte("Cvcn")},
+	"basic":  {[]byte("aaanaaan"), []byte("aannaaan"), []byte("aaannaaa")},
+	"pin":    {[]byte("nnnn")},
+}
+
+var template_characters = map[byte]string{
+	'V': "AEIOU",
+	'C': "BCDFGHJKLMNPQRSTVWXYZ",
+	'v': "aeiou",
+	'c': "bcdfghjklmnpqrstvwxyz",
+	'A': "AEIOUBCDFGHJKLMNPQRSTVWXYZ",
+	'a': "AEIOUaeiouBCDFGHJKLMNPQRSTVWXYZbcdfghjklmnpqrstvwxyz",
+	'n': "0123456789",
+	'o': "@&%?,=[]_:-+*$#!'^~;()/.",
+	'x': "AEIOUaeiouBCDFGHJKLMNPQRSTVWXYZbcdfghjklmnpqrstvwxyz0123456789!@#$%^&*()",
+}
+
+func derivePassword(counter uint32, password_type, password, user, site string) string {
+	var templates = password_type_templates[password_type]
+	if templates == nil {
+		return fmt.Sprintf("cannot find password template %s", password_type)
+	}
+
+	var buffer bytes.Buffer
+	buffer.WriteString(master_password_seed)
+	binary.Write(&buffer, binary.BigEndian, uint32(len(user)))
+	buffer.WriteString(user)
+
+	salt := buffer.Bytes()
+	key, err := scrypt.Key([]byte(password), salt, 32768, 8, 2, 64)
+	if err != nil {
+		return fmt.Sprintf("failed to derive password: %s", err)
+	}
+
+	buffer.Truncate(len(master_password_seed))
+	binary.Write(&buffer, binary.BigEndian, uint32(len(site)))
+	buffer.WriteString(site)
+	binary.Write(&buffer, binary.BigEndian, counter)
+
+	var hmacv = hmac.New(sha256.New, key)
+	hmacv.Write(buffer.Bytes())
+	var seed = hmacv.Sum(nil)
+	var temp = templates[int(seed[0])%len(templates)]
+
+	buffer.Truncate(0)
+	for i, element := range temp {
+		pass_chars := template_characters[element]
+		pass_char := pass_chars[int(seed[i+1])%len(pass_chars)]
+		buffer.WriteByte(pass_char)
+	}
+
+	return buffer.String()
+}
+
+func generatePrivateKey(typ string) string {
+	var priv interface{}
+	var err error
+	switch typ {
+	case "", "rsa":
+		// good enough for government work
+		priv, err = rsa.GenerateKey(rand.Reader, 4096)
+	case "dsa":
+		key := new(dsa.PrivateKey)
+		// again, good enough for government work
+		if err = dsa.GenerateParameters(&key.Parameters, rand.Reader, dsa.L2048N256); err != nil {
+			return fmt.Sprintf("failed to generate dsa params: %s", err)
+		}
+		err = dsa.GenerateKey(key, rand.Reader)
+		priv = key
+	case "ecdsa":
+		// again, good enough for government work
+		priv, err = ecdsa.GenerateKey(elliptic.P256(), rand.Reader)
+	default:
+		return "Unknown type " + typ
+	}
+	if err != nil {
+		return fmt.Sprintf("failed to generate private key: %s", err)
+	}
+
+	return string(pem.EncodeToMemory(pemBlockForKey(priv)))
+}
+
+type DSAKeyFormat struct {
+	Version       int
+	P, Q, G, Y, X *big.Int
+}
+
+func pemBlockForKey(priv interface{}) *pem.Block {
+	switch k := priv.(type) {
+	case *rsa.PrivateKey:
+		return &pem.Block{Type: "RSA PRIVATE KEY", Bytes: x509.MarshalPKCS1PrivateKey(k)}
+	case *dsa.PrivateKey:
+		val := DSAKeyFormat{
+			P: k.P, Q: k.Q, G: k.G,
+			Y: k.Y, X: k.X,
+		}
+		bytes, _ := asn1.Marshal(val)
+		return &pem.Block{Type: "DSA PRIVATE KEY", Bytes: bytes}
+	case *ecdsa.PrivateKey:
+		b, _ := x509.MarshalECPrivateKey(k)
+		return &pem.Block{Type: "EC PRIVATE KEY", Bytes: b}
+	default:
+		return nil
+	}
+}
+
+type certificate struct {
+	Cert string
+	Key  string
+}
+
+func buildCustomCertificate(b64cert string, b64key string) (certificate, error) {
+	crt := certificate{}
+
+	cert, err := base64.StdEncoding.DecodeString(b64cert)
+	if err != nil {
+		return crt, errors.New("unable to decode base64 certificate")
+	}
+
+	key, err := base64.StdEncoding.DecodeString(b64key)
+	if err != nil {
+		return crt, errors.New("unable to decode base64 private key")
+	}
+
+	decodedCert, _ := pem.Decode(cert)
+	if decodedCert == nil {
+		return crt, errors.New("unable to decode certificate")
+	}
+	_, err = x509.ParseCertificate(decodedCert.Bytes)
+	if err != nil {
+		return crt, fmt.Errorf(
+			"error parsing certificate: decodedCert.Bytes: %s",
+			err,
+		)
+	}
+
+	decodedKey, _ := pem.Decode(key)
+	if decodedKey == nil {
+		return crt, errors.New("unable to decode key")
+	}
+	_, err = x509.ParsePKCS1PrivateKey(decodedKey.Bytes)
+	if err != nil {
+		return crt, fmt.Errorf(
+			"error parsing prive key: decodedKey.Bytes: %s",
+			err,
+		)
+	}
+
+	crt.Cert = string(cert)
+	crt.Key = string(key)
+
+	return crt, nil
+}
+
+func generateCertificateAuthority(
+	cn string,
+	daysValid int,
+) (certificate, error) {
+	ca := certificate{}
+
+	template, err := getBaseCertTemplate(cn, nil, nil, daysValid)
+	if err != nil {
+		return ca, err
+	}
+	// Override KeyUsage and IsCA
+	template.KeyUsage = x509.KeyUsageKeyEncipherment |
+		x509.KeyUsageDigitalSignature |
+		x509.KeyUsageCertSign
+	template.IsCA = true
+
+	priv, err := rsa.GenerateKey(rand.Reader, 2048)
+	if err != nil {
+		return ca, fmt.Errorf("error generating rsa key: %s", err)
+	}
+
+	ca.Cert, ca.Key, err = getCertAndKey(template, priv, template, priv)
+	if err != nil {
+		return ca, err
+	}
+
+	return ca, nil
+}
+
+func generateSelfSignedCertificate(
+	cn string,
+	ips []interface{},
+	alternateDNS []interface{},
+	daysValid int,
+) (certificate, error) {
+	cert := certificate{}
+
+	template, err := getBaseCertTemplate(cn, ips, alternateDNS, daysValid)
+	if err != nil {
+		return cert, err
+	}
+
+	priv, err := rsa.GenerateKey(rand.Reader, 2048)
+	if err != nil {
+		return cert, fmt.Errorf("error generating rsa key: %s", err)
+	}
+
+	cert.Cert, cert.Key, err = getCertAndKey(template, priv, template, priv)
+	if err != nil {
+		return cert, err
+	}
+
+	return cert, nil
+}
+
+func generateSignedCertificate(
+	cn string,
+	ips []interface{},
+	alternateDNS []interface{},
+	daysValid int,
+	ca certificate,
+) (certificate, error) {
+	cert := certificate{}
+
+	decodedSignerCert, _ := pem.Decode([]byte(ca.Cert))
+	if decodedSignerCert == nil {
+		return cert, errors.New("unable to decode certificate")
+	}
+	signerCert, err := x509.ParseCertificate(decodedSignerCert.Bytes)
+	if err != nil {
+		return cert, fmt.Errorf(
+			"error parsing certificate: decodedSignerCert.Bytes: %s",
+			err,
+		)
+	}
+	decodedSignerKey, _ := pem.Decode([]byte(ca.Key))
+	if decodedSignerKey == nil {
+		return cert, errors.New("unable to decode key")
+	}
+	signerKey, err := x509.ParsePKCS1PrivateKey(decodedSignerKey.Bytes)
+	if err != nil {
+		return cert, fmt.Errorf(
+			"error parsing prive key: decodedSignerKey.Bytes: %s",
+			err,
+		)
+	}
+
+	template, err := getBaseCertTemplate(cn, ips, alternateDNS, daysValid)
+	if err != nil {
+		return cert, err
+	}
+
+	priv, err := rsa.GenerateKey(rand.Reader, 2048)
+	if err != nil {
+		return cert, fmt.Errorf("error generating rsa key: %s", err)
+	}
+
+	cert.Cert, cert.Key, err = getCertAndKey(
+		template,
+		priv,
+		signerCert,
+		signerKey,
+	)
+	if err != nil {
+		return cert, err
+	}
+
+	return cert, nil
+}
+
+func getCertAndKey(
+	template *x509.Certificate,
+	signeeKey *rsa.PrivateKey,
+	parent *x509.Certificate,
+	signingKey *rsa.PrivateKey,
+) (string, string, error) {
+	derBytes, err := x509.CreateCertificate(
+		rand.Reader,
+		template,
+		parent,
+		&signeeKey.PublicKey,
+		signingKey,
+	)
+	if err != nil {
+		return "", "", fmt.Errorf("error creating certificate: %s", err)
+	}
+
+	certBuffer := bytes.Buffer{}
+	if err := pem.Encode(
+		&certBuffer,
+		&pem.Block{Type: "CERTIFICATE", Bytes: derBytes},
+	); err != nil {
+		return "", "", fmt.Errorf("error pem-encoding certificate: %s", err)
+	}
+
+	keyBuffer := bytes.Buffer{}
+	if err := pem.Encode(
+		&keyBuffer,
+		&pem.Block{
+			Type:  "RSA PRIVATE KEY",
+			Bytes: x509.MarshalPKCS1PrivateKey(signeeKey),
+		},
+	); err != nil {
+		return "", "", fmt.Errorf("error pem-encoding key: %s", err)
+	}
+
+	return string(certBuffer.Bytes()), string(keyBuffer.Bytes()), nil
+}
+
+func getBaseCertTemplate(
+	cn string,
+	ips []interface{},
+	alternateDNS []interface{},
+	daysValid int,
+) (*x509.Certificate, error) {
+	ipAddresses, err := getNetIPs(ips)
+	if err != nil {
+		return nil, err
+	}
+	dnsNames, err := getAlternateDNSStrs(alternateDNS)
+	if err != nil {
+		return nil, err
+	}
+	serialNumberUpperBound := new(big.Int).Lsh(big.NewInt(1), 128)
+	serialNumber, err := rand.Int(rand.Reader, serialNumberUpperBound)
+	if err != nil {
+		return nil, err
+	}
+	return &x509.Certificate{
+		SerialNumber: serialNumber,
+		Subject: pkix.Name{
+			CommonName: cn,
+		},
+		IPAddresses: ipAddresses,
+		DNSNames:    dnsNames,
+		NotBefore:   time.Now(),
+		NotAfter:    time.Now().Add(time.Hour * 24 * time.Duration(daysValid)),
+		KeyUsage:    x509.KeyUsageKeyEncipherment | x509.KeyUsageDigitalSignature,
+		ExtKeyUsage: []x509.ExtKeyUsage{
+			x509.ExtKeyUsageServerAuth,
+			x509.ExtKeyUsageClientAuth,
+		},
+		BasicConstraintsValid: true,
+	}, nil
+}
+
+func getNetIPs(ips []interface{}) ([]net.IP, error) {
+	if ips == nil {
+		return []net.IP{}, nil
+	}
+	var ipStr string
+	var ok bool
+	var netIP net.IP
+	netIPs := make([]net.IP, len(ips))
+	for i, ip := range ips {
+		ipStr, ok = ip.(string)
+		if !ok {
+			return nil, fmt.Errorf("error parsing ip: %v is not a string", ip)
+		}
+		netIP = net.ParseIP(ipStr)
+		if netIP == nil {
+			return nil, fmt.Errorf("error parsing ip: %s", ipStr)
+		}
+		netIPs[i] = netIP
+	}
+	return netIPs, nil
+}
+
+func getAlternateDNSStrs(alternateDNS []interface{}) ([]string, error) {
+	if alternateDNS == nil {
+		return []string{}, nil
+	}
+	var dnsStr string
+	var ok bool
+	alternateDNSStrs := make([]string, len(alternateDNS))
+	for i, dns := range alternateDNS {
+		dnsStr, ok = dns.(string)
+		if !ok {
+			return nil, fmt.Errorf(
+				"error processing alternate dns name: %v is not a string",
+				dns,
+			)
+		}
+		alternateDNSStrs[i] = dnsStr
+	}
+	return alternateDNSStrs, nil
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/date.go b/backend/vendor/github.com/Masterminds/sprig/date.go
new file mode 100644
index 00000000..1c2c3653
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/date.go
@@ -0,0 +1,76 @@
+package sprig
+
+import (
+	"time"
+)
+
+// Given a format and a date, format the date string.
+//
+// Date can be a `time.Time` or an `int, int32, int64`.
+// In the later case, it is treated as seconds since UNIX
+// epoch.
+func date(fmt string, date interface{}) string {
+	return dateInZone(fmt, date, "Local")
+}
+
+func htmlDate(date interface{}) string {
+	return dateInZone("2006-01-02", date, "Local")
+}
+
+func htmlDateInZone(date interface{}, zone string) string {
+	return dateInZone("2006-01-02", date, zone)
+}
+
+func dateInZone(fmt string, date interface{}, zone string) string {
+	var t time.Time
+	switch date := date.(type) {
+	default:
+		t = time.Now()
+	case time.Time:
+		t = date
+	case int64:
+		t = time.Unix(date, 0)
+	case int:
+		t = time.Unix(int64(date), 0)
+	case int32:
+		t = time.Unix(int64(date), 0)
+	}
+
+	loc, err := time.LoadLocation(zone)
+	if err != nil {
+		loc, _ = time.LoadLocation("UTC")
+	}
+
+	return t.In(loc).Format(fmt)
+}
+
+func dateModify(fmt string, date time.Time) time.Time {
+	d, err := time.ParseDuration(fmt)
+	if err != nil {
+		return date
+	}
+	return date.Add(d)
+}
+
+func dateAgo(date interface{}) string {
+	var t time.Time
+
+	switch date := date.(type) {
+	default:
+		t = time.Now()
+	case time.Time:
+		t = date
+	case int64:
+		t = time.Unix(date, 0)
+	case int:
+		t = time.Unix(int64(date), 0)
+	}
+	// Drop resolution to seconds
+	duration := time.Since(t).Round(time.Second)
+	return duration.String()
+}
+
+func toDate(fmt, str string) time.Time {
+	t, _ := time.ParseInLocation(fmt, str, time.Local)
+	return t
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/defaults.go b/backend/vendor/github.com/Masterminds/sprig/defaults.go
new file mode 100644
index 00000000..ed6a8ab2
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/defaults.go
@@ -0,0 +1,83 @@
+package sprig
+
+import (
+	"encoding/json"
+	"reflect"
+)
+
+// dfault checks whether `given` is set, and returns default if not set.
+//
+// This returns `d` if `given` appears not to be set, and `given` otherwise.
+//
+// For numeric types 0 is unset.
+// For strings, maps, arrays, and slices, len() = 0 is considered unset.
+// For bool, false is unset.
+// Structs are never considered unset.
+//
+// For everything else, including pointers, a nil value is unset.
+func dfault(d interface{}, given ...interface{}) interface{} {
+
+	if empty(given) || empty(given[0]) {
+		return d
+	}
+	return given[0]
+}
+
+// empty returns true if the given value has the zero value for its type.
+func empty(given interface{}) bool {
+	g := reflect.ValueOf(given)
+	if !g.IsValid() {
+		return true
+	}
+
+	// Basically adapted from text/template.isTrue
+	switch g.Kind() {
+	default:
+		return g.IsNil()
+	case reflect.Array, reflect.Slice, reflect.Map, reflect.String:
+		return g.Len() == 0
+	case reflect.Bool:
+		return g.Bool() == false
+	case reflect.Complex64, reflect.Complex128:
+		return g.Complex() == 0
+	case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+		return g.Int() == 0
+	case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr:
+		return g.Uint() == 0
+	case reflect.Float32, reflect.Float64:
+		return g.Float() == 0
+	case reflect.Struct:
+		return false
+	}
+}
+
+// coalesce returns the first non-empty value.
+func coalesce(v ...interface{}) interface{} {
+	for _, val := range v {
+		if !empty(val) {
+			return val
+		}
+	}
+	return nil
+}
+
+// toJson encodes an item into a JSON string
+func toJson(v interface{}) string {
+	output, _ := json.Marshal(v)
+	return string(output)
+}
+
+// toPrettyJson encodes an item into a pretty (indented) JSON string
+func toPrettyJson(v interface{}) string {
+	output, _ := json.MarshalIndent(v, "", "  ")
+	return string(output)
+}
+
+// ternary returns the first value if the last value is true, otherwise returns the second value.
+func ternary(vt interface{}, vf interface{}, v bool) interface{} {
+	if v {
+		return vt
+	}
+
+	return vf
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/dict.go b/backend/vendor/github.com/Masterminds/sprig/dict.go
new file mode 100644
index 00000000..3713e58a
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/dict.go
@@ -0,0 +1,97 @@
+package sprig
+
+import "github.com/imdario/mergo"
+
+func set(d map[string]interface{}, key string, value interface{}) map[string]interface{} {
+	d[key] = value
+	return d
+}
+
+func unset(d map[string]interface{}, key string) map[string]interface{} {
+	delete(d, key)
+	return d
+}
+
+func hasKey(d map[string]interface{}, key string) bool {
+	_, ok := d[key]
+	return ok
+}
+
+func pluck(key string, d ...map[string]interface{}) []interface{} {
+	res := []interface{}{}
+	for _, dict := range d {
+		if val, ok := dict[key]; ok {
+			res = append(res, val)
+		}
+	}
+	return res
+}
+
+func keys(dicts ...map[string]interface{}) []string {
+	k := []string{}
+	for _, dict := range dicts {
+		for key := range dict {
+			k = append(k, key)
+		}
+	}
+	return k
+}
+
+func pick(dict map[string]interface{}, keys ...string) map[string]interface{} {
+	res := map[string]interface{}{}
+	for _, k := range keys {
+		if v, ok := dict[k]; ok {
+			res[k] = v
+		}
+	}
+	return res
+}
+
+func omit(dict map[string]interface{}, keys ...string) map[string]interface{} {
+	res := map[string]interface{}{}
+
+	omit := make(map[string]bool, len(keys))
+	for _, k := range keys {
+		omit[k] = true
+	}
+
+	for k, v := range dict {
+		if _, ok := omit[k]; !ok {
+			res[k] = v
+		}
+	}
+	return res
+}
+
+func dict(v ...interface{}) map[string]interface{} {
+	dict := map[string]interface{}{}
+	lenv := len(v)
+	for i := 0; i < lenv; i += 2 {
+		key := strval(v[i])
+		if i+1 >= lenv {
+			dict[key] = ""
+			continue
+		}
+		dict[key] = v[i+1]
+	}
+	return dict
+}
+
+func merge(dst map[string]interface{}, srcs ...map[string]interface{}) interface{} {
+	for _, src := range srcs {
+		if err := mergo.Merge(&dst, src); err != nil {
+			// Swallow errors inside of a template.
+			return ""
+		}
+	}
+	return dst
+}
+
+func values(dict map[string]interface{}) []interface{} {
+	values := []interface{}{}
+	for _, value := range dict {
+		values = append(values, value)
+	}
+
+	return values
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/doc.go b/backend/vendor/github.com/Masterminds/sprig/doc.go
new file mode 100644
index 00000000..8f8f1d73
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/doc.go
@@ -0,0 +1,19 @@
+/*
+Sprig: Template functions for Go.
+
+This package contains a number of utility functions for working with data
+inside of Go `html/template` and `text/template` files.
+
+To add these functions, use the `template.Funcs()` method:
+
+	t := templates.New("foo").Funcs(sprig.FuncMap())
+
+Note that you should add the function map before you parse any template files.
+
+	In several cases, Sprig reverses the order of arguments from the way they
+	appear in the standard library. This is to make it easier to pipe
+	arguments into functions.
+
+See http://masterminds.github.io/sprig/ for more detailed documentation on each of the available functions.
+*/
+package sprig
diff --git a/backend/vendor/github.com/Masterminds/sprig/functions.go b/backend/vendor/github.com/Masterminds/sprig/functions.go
new file mode 100644
index 00000000..e985e969
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/functions.go
@@ -0,0 +1,285 @@
+package sprig
+
+import (
+	"errors"
+	"html/template"
+	"os"
+	"path"
+	"strconv"
+	"strings"
+	ttemplate "text/template"
+	"time"
+
+	util "github.com/aokoli/goutils"
+	"github.com/huandu/xstrings"
+)
+
+// Produce the function map.
+//
+// Use this to pass the functions into the template engine:
+//
+// 	tpl := template.New("foo").Funcs(sprig.FuncMap()))
+//
+func FuncMap() template.FuncMap {
+	return HtmlFuncMap()
+}
+
+// HermeticTextFuncMap returns a 'text/template'.FuncMap with only repeatable functions.
+func HermeticTxtFuncMap() ttemplate.FuncMap {
+	r := TxtFuncMap()
+	for _, name := range nonhermeticFunctions {
+		delete(r, name)
+	}
+	return r
+}
+
+// HermeticHtmlFuncMap returns an 'html/template'.Funcmap with only repeatable functions.
+func HermeticHtmlFuncMap() template.FuncMap {
+	r := HtmlFuncMap()
+	for _, name := range nonhermeticFunctions {
+		delete(r, name)
+	}
+	return r
+}
+
+// TextFuncMap returns a 'text/template'.FuncMap
+func TxtFuncMap() ttemplate.FuncMap {
+	return ttemplate.FuncMap(GenericFuncMap())
+}
+
+// HtmlFuncMap returns an 'html/template'.Funcmap
+func HtmlFuncMap() template.FuncMap {
+	return template.FuncMap(GenericFuncMap())
+}
+
+// GenericFuncMap returns a copy of the basic function map as a map[string]interface{}.
+func GenericFuncMap() map[string]interface{} {
+	gfm := make(map[string]interface{}, len(genericMap))
+	for k, v := range genericMap {
+		gfm[k] = v
+	}
+	return gfm
+}
+
+// These functions are not guaranteed to evaluate to the same result for given input, because they
+// refer to the environemnt or global state.
+var nonhermeticFunctions = []string{
+	// Date functions
+	"date",
+	"date_in_zone",
+	"date_modify",
+	"now",
+	"htmlDate",
+	"htmlDateInZone",
+	"dateInZone",
+	"dateModify",
+
+	// Strings
+	"randAlphaNum",
+	"randAlpha",
+	"randAscii",
+	"randNumeric",
+	"uuidv4",
+
+	// OS
+	"env",
+	"expandenv",
+}
+
+var genericMap = map[string]interface{}{
+	"hello": func() string { return "Hello!" },
+
+	// Date functions
+	"date":           date,
+	"date_in_zone":   dateInZone,
+	"date_modify":    dateModify,
+	"now":            func() time.Time { return time.Now() },
+	"htmlDate":       htmlDate,
+	"htmlDateInZone": htmlDateInZone,
+	"dateInZone":     dateInZone,
+	"dateModify":     dateModify,
+	"ago":            dateAgo,
+	"toDate":         toDate,
+
+	// Strings
+	"abbrev":     abbrev,
+	"abbrevboth": abbrevboth,
+	"trunc":      trunc,
+	"trim":       strings.TrimSpace,
+	"upper":      strings.ToUpper,
+	"lower":      strings.ToLower,
+	"title":      strings.Title,
+	"untitle":    untitle,
+	"substr":     substring,
+	// Switch order so that "foo" | repeat 5
+	"repeat": func(count int, str string) string { return strings.Repeat(str, count) },
+	// Deprecated: Use trimAll.
+	"trimall": func(a, b string) string { return strings.Trim(b, a) },
+	// Switch order so that "$foo" | trimall "$"
+	"trimAll":      func(a, b string) string { return strings.Trim(b, a) },
+	"trimSuffix":   func(a, b string) string { return strings.TrimSuffix(b, a) },
+	"trimPrefix":   func(a, b string) string { return strings.TrimPrefix(b, a) },
+	"nospace":      util.DeleteWhiteSpace,
+	"initials":     initials,
+	"randAlphaNum": randAlphaNumeric,
+	"randAlpha":    randAlpha,
+	"randAscii":    randAscii,
+	"randNumeric":  randNumeric,
+	"swapcase":     util.SwapCase,
+	"shuffle":      xstrings.Shuffle,
+	"snakecase":    xstrings.ToSnakeCase,
+	"camelcase":    xstrings.ToCamelCase,
+	"wrap":         func(l int, s string) string { return util.Wrap(s, l) },
+	"wrapWith":     func(l int, sep, str string) string { return util.WrapCustom(str, l, sep, true) },
+	// Switch order so that "foobar" | contains "foo"
+	"contains":  func(substr string, str string) bool { return strings.Contains(str, substr) },
+	"hasPrefix": func(substr string, str string) bool { return strings.HasPrefix(str, substr) },
+	"hasSuffix": func(substr string, str string) bool { return strings.HasSuffix(str, substr) },
+	"quote":     quote,
+	"squote":    squote,
+	"cat":       cat,
+	"indent":    indent,
+	"nindent":   nindent,
+	"replace":   replace,
+	"plural":    plural,
+	"sha1sum":   sha1sum,
+	"sha256sum": sha256sum,
+	"toString":  strval,
+
+	// Wrap Atoi to stop errors.
+	"atoi":    func(a string) int { i, _ := strconv.Atoi(a); return i },
+	"int64":   toInt64,
+	"int":     toInt,
+	"float64": toFloat64,
+
+	//"gt": func(a, b int) bool {return a > b},
+	//"gte": func(a, b int) bool {return a >= b},
+	//"lt": func(a, b int) bool {return a < b},
+	//"lte": func(a, b int) bool {return a <= b},
+
+	// split "/" foo/bar returns map[int]string{0: foo, 1: bar}
+	"split":     split,
+	"splitList": func(sep, orig string) []string { return strings.Split(orig, sep) },
+	// splitn "/" foo/bar/fuu returns map[int]string{0: foo, 1: bar/fuu}
+	"splitn":    splitn,
+	"toStrings": strslice,
+
+	"until":     until,
+	"untilStep": untilStep,
+
+	// VERY basic arithmetic.
+	"add1": func(i interface{}) int64 { return toInt64(i) + 1 },
+	"add": func(i ...interface{}) int64 {
+		var a int64 = 0
+		for _, b := range i {
+			a += toInt64(b)
+		}
+		return a
+	},
+	"sub": func(a, b interface{}) int64 { return toInt64(a) - toInt64(b) },
+	"div": func(a, b interface{}) int64 { return toInt64(a) / toInt64(b) },
+	"mod": func(a, b interface{}) int64 { return toInt64(a) % toInt64(b) },
+	"mul": func(a interface{}, v ...interface{}) int64 {
+		val := toInt64(a)
+		for _, b := range v {
+			val = val * toInt64(b)
+		}
+		return val
+	},
+	"biggest": max,
+	"max":     max,
+	"min":     min,
+	"ceil":    ceil,
+	"floor":   floor,
+	"round":   round,
+
+	// string slices. Note that we reverse the order b/c that's better
+	// for template processing.
+	"join":      join,
+	"sortAlpha": sortAlpha,
+
+	// Defaults
+	"default":      dfault,
+	"empty":        empty,
+	"coalesce":     coalesce,
+	"compact":      compact,
+	"toJson":       toJson,
+	"toPrettyJson": toPrettyJson,
+	"ternary":      ternary,
+
+	// Reflection
+	"typeOf":     typeOf,
+	"typeIs":     typeIs,
+	"typeIsLike": typeIsLike,
+	"kindOf":     kindOf,
+	"kindIs":     kindIs,
+
+	// OS:
+	"env":       func(s string) string { return os.Getenv(s) },
+	"expandenv": func(s string) string { return os.ExpandEnv(s) },
+
+	// File Paths:
+	"base":  path.Base,
+	"dir":   path.Dir,
+	"clean": path.Clean,
+	"ext":   path.Ext,
+	"isAbs": path.IsAbs,
+
+	// Encoding:
+	"b64enc": base64encode,
+	"b64dec": base64decode,
+	"b32enc": base32encode,
+	"b32dec": base32decode,
+
+	// Data Structures:
+	"tuple":  list, // FIXME: with the addition of append/prepend these are no longer immutable.
+	"list":   list,
+	"dict":   dict,
+	"set":    set,
+	"unset":  unset,
+	"hasKey": hasKey,
+	"pluck":  pluck,
+	"keys":   keys,
+	"pick":   pick,
+	"omit":   omit,
+	"merge":  merge,
+	"values": values,
+
+	"append": push, "push": push,
+	"prepend": prepend,
+	"first":   first,
+	"rest":    rest,
+	"last":    last,
+	"initial": initial,
+	"reverse": reverse,
+	"uniq":    uniq,
+	"without": without,
+	"has":     has,
+	"slice":   slice,
+
+	// Crypto:
+	"genPrivateKey":     generatePrivateKey,
+	"derivePassword":    derivePassword,
+	"buildCustomCert":   buildCustomCertificate,
+	"genCA":             generateCertificateAuthority,
+	"genSelfSignedCert": generateSelfSignedCertificate,
+	"genSignedCert":     generateSignedCertificate,
+
+	// UUIDs:
+	"uuidv4": uuidv4,
+
+	// SemVer:
+	"semver":        semver,
+	"semverCompare": semverCompare,
+
+	// Flow Control:
+	"fail": func(msg string) (string, error) { return "", errors.New(msg) },
+
+	// Regex
+	"regexMatch":             regexMatch,
+	"regexFindAll":           regexFindAll,
+	"regexFind":              regexFind,
+	"regexReplaceAll":        regexReplaceAll,
+	"regexReplaceAllLiteral": regexReplaceAllLiteral,
+	"regexSplit":             regexSplit,
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/glide.yaml b/backend/vendor/github.com/Masterminds/sprig/glide.yaml
new file mode 100644
index 00000000..772ba913
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/glide.yaml
@@ -0,0 +1,15 @@
+package: github.com/Masterminds/sprig
+import:
+- package: github.com/Masterminds/goutils
+  version: ^1.0.0
+- package: github.com/google/uuid
+  version: ^0.2
+- package: golang.org/x/crypto
+  subpackages:
+  - scrypt
+- package: github.com/Masterminds/semver
+  version: v1.2.2
+- package: github.com/stretchr/testify
+- package: github.com/imdario/mergo
+  version: ~0.2.2
+- package: github.com/huandu/xstrings
diff --git a/backend/vendor/github.com/Masterminds/sprig/list.go b/backend/vendor/github.com/Masterminds/sprig/list.go
new file mode 100644
index 00000000..184c1ca1
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/list.go
@@ -0,0 +1,291 @@
+package sprig
+
+import (
+	"fmt"
+	"reflect"
+	"sort"
+)
+
+// Reflection is used in these functions so that slices and arrays of strings,
+// ints, and other types not implementing []interface{} can be worked with.
+// For example, this is useful if you need to work on the output of regexs.
+
+func list(v ...interface{}) []interface{} {
+	return v
+}
+
+func push(list interface{}, v interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		nl := make([]interface{}, l)
+		for i := 0; i < l; i++ {
+			nl[i] = l2.Index(i).Interface()
+		}
+
+		return append(nl, v)
+
+	default:
+		panic(fmt.Sprintf("Cannot push on type %s", tp))
+	}
+}
+
+func prepend(list interface{}, v interface{}) []interface{} {
+	//return append([]interface{}{v}, list...)
+
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		nl := make([]interface{}, l)
+		for i := 0; i < l; i++ {
+			nl[i] = l2.Index(i).Interface()
+		}
+
+		return append([]interface{}{v}, nl...)
+
+	default:
+		panic(fmt.Sprintf("Cannot prepend on type %s", tp))
+	}
+}
+
+func last(list interface{}) interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		if l == 0 {
+			return nil
+		}
+
+		return l2.Index(l - 1).Interface()
+	default:
+		panic(fmt.Sprintf("Cannot find last on type %s", tp))
+	}
+}
+
+func first(list interface{}) interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		if l == 0 {
+			return nil
+		}
+
+		return l2.Index(0).Interface()
+	default:
+		panic(fmt.Sprintf("Cannot find first on type %s", tp))
+	}
+}
+
+func rest(list interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		if l == 0 {
+			return nil
+		}
+
+		nl := make([]interface{}, l-1)
+		for i := 1; i < l; i++ {
+			nl[i-1] = l2.Index(i).Interface()
+		}
+
+		return nl
+	default:
+		panic(fmt.Sprintf("Cannot find rest on type %s", tp))
+	}
+}
+
+func initial(list interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		if l == 0 {
+			return nil
+		}
+
+		nl := make([]interface{}, l-1)
+		for i := 0; i < l-1; i++ {
+			nl[i] = l2.Index(i).Interface()
+		}
+
+		return nl
+	default:
+		panic(fmt.Sprintf("Cannot find initial on type %s", tp))
+	}
+}
+
+func sortAlpha(list interface{}) []string {
+	k := reflect.Indirect(reflect.ValueOf(list)).Kind()
+	switch k {
+	case reflect.Slice, reflect.Array:
+		a := strslice(list)
+		s := sort.StringSlice(a)
+		s.Sort()
+		return s
+	}
+	return []string{strval(list)}
+}
+
+func reverse(v interface{}) []interface{} {
+	tp := reflect.TypeOf(v).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(v)
+
+		l := l2.Len()
+		// We do not sort in place because the incoming array should not be altered.
+		nl := make([]interface{}, l)
+		for i := 0; i < l; i++ {
+			nl[l-i-1] = l2.Index(i).Interface()
+		}
+
+		return nl
+	default:
+		panic(fmt.Sprintf("Cannot find reverse on type %s", tp))
+	}
+}
+
+func compact(list interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		nl := []interface{}{}
+		var item interface{}
+		for i := 0; i < l; i++ {
+			item = l2.Index(i).Interface()
+			if !empty(item) {
+				nl = append(nl, item)
+			}
+		}
+
+		return nl
+	default:
+		panic(fmt.Sprintf("Cannot compact on type %s", tp))
+	}
+}
+
+func uniq(list interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		dest := []interface{}{}
+		var item interface{}
+		for i := 0; i < l; i++ {
+			item = l2.Index(i).Interface()
+			if !inList(dest, item) {
+				dest = append(dest, item)
+			}
+		}
+
+		return dest
+	default:
+		panic(fmt.Sprintf("Cannot find uniq on type %s", tp))
+	}
+}
+
+func inList(haystack []interface{}, needle interface{}) bool {
+	for _, h := range haystack {
+		if reflect.DeepEqual(needle, h) {
+			return true
+		}
+	}
+	return false
+}
+
+func without(list interface{}, omit ...interface{}) []interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		res := []interface{}{}
+		var item interface{}
+		for i := 0; i < l; i++ {
+			item = l2.Index(i).Interface()
+			if !inList(omit, item) {
+				res = append(res, item)
+			}
+		}
+
+		return res
+	default:
+		panic(fmt.Sprintf("Cannot find without on type %s", tp))
+	}
+}
+
+func has(needle interface{}, haystack interface{}) bool {
+	tp := reflect.TypeOf(haystack).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(haystack)
+		var item interface{}
+		l := l2.Len()
+		for i := 0; i < l; i++ {
+			item = l2.Index(i).Interface()
+			if reflect.DeepEqual(needle, item) {
+				return true
+			}
+		}
+
+		return false
+	default:
+		panic(fmt.Sprintf("Cannot find has on type %s", tp))
+	}
+}
+
+// $list := [1, 2, 3, 4, 5]
+// slice $list     -> list[0:5] = list[:]
+// slice $list 0 3 -> list[0:3] = list[:3]
+// slice $list 3 5 -> list[3:5]
+// slice $list 3   -> list[3:5] = list[3:]
+func slice(list interface{}, indices ...interface{}) interface{} {
+	tp := reflect.TypeOf(list).Kind()
+	switch tp {
+	case reflect.Slice, reflect.Array:
+		l2 := reflect.ValueOf(list)
+
+		l := l2.Len()
+		if l == 0 {
+			return nil
+		}
+
+		var start, end int
+		if len(indices) > 0 {
+			start = toInt(indices[0])
+		}
+		if len(indices) < 2 {
+			end = l
+		} else {
+			end = toInt(indices[1])
+		}
+
+		return l2.Slice(start, end).Interface()
+	default:
+		panic(fmt.Sprintf("list should be type of slice or array but %s", tp))
+	}
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/numeric.go b/backend/vendor/github.com/Masterminds/sprig/numeric.go
new file mode 100644
index 00000000..4bd89bf7
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/numeric.go
@@ -0,0 +1,159 @@
+package sprig
+
+import (
+	"math"
+	"reflect"
+	"strconv"
+)
+
+// toFloat64 converts 64-bit floats
+func toFloat64(v interface{}) float64 {
+	if str, ok := v.(string); ok {
+		iv, err := strconv.ParseFloat(str, 64)
+		if err != nil {
+			return 0
+		}
+		return iv
+	}
+
+	val := reflect.Indirect(reflect.ValueOf(v))
+	switch val.Kind() {
+	case reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, reflect.Int:
+		return float64(val.Int())
+	case reflect.Uint8, reflect.Uint16, reflect.Uint32:
+		return float64(val.Uint())
+	case reflect.Uint, reflect.Uint64:
+		return float64(val.Uint())
+	case reflect.Float32, reflect.Float64:
+		return val.Float()
+	case reflect.Bool:
+		if val.Bool() == true {
+			return 1
+		}
+		return 0
+	default:
+		return 0
+	}
+}
+
+func toInt(v interface{}) int {
+	//It's not optimal. Bud I don't want duplicate toInt64 code.
+	return int(toInt64(v))
+}
+
+// toInt64 converts integer types to 64-bit integers
+func toInt64(v interface{}) int64 {
+	if str, ok := v.(string); ok {
+		iv, err := strconv.ParseInt(str, 10, 64)
+		if err != nil {
+			return 0
+		}
+		return iv
+	}
+
+	val := reflect.Indirect(reflect.ValueOf(v))
+	switch val.Kind() {
+	case reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, reflect.Int:
+		return val.Int()
+	case reflect.Uint8, reflect.Uint16, reflect.Uint32:
+		return int64(val.Uint())
+	case reflect.Uint, reflect.Uint64:
+		tv := val.Uint()
+		if tv <= math.MaxInt64 {
+			return int64(tv)
+		}
+		// TODO: What is the sensible thing to do here?
+		return math.MaxInt64
+	case reflect.Float32, reflect.Float64:
+		return int64(val.Float())
+	case reflect.Bool:
+		if val.Bool() == true {
+			return 1
+		}
+		return 0
+	default:
+		return 0
+	}
+}
+
+func max(a interface{}, i ...interface{}) int64 {
+	aa := toInt64(a)
+	for _, b := range i {
+		bb := toInt64(b)
+		if bb > aa {
+			aa = bb
+		}
+	}
+	return aa
+}
+
+func min(a interface{}, i ...interface{}) int64 {
+	aa := toInt64(a)
+	for _, b := range i {
+		bb := toInt64(b)
+		if bb < aa {
+			aa = bb
+		}
+	}
+	return aa
+}
+
+func until(count int) []int {
+	step := 1
+	if count < 0 {
+		step = -1
+	}
+	return untilStep(0, count, step)
+}
+
+func untilStep(start, stop, step int) []int {
+	v := []int{}
+
+	if stop < start {
+		if step >= 0 {
+			return v
+		}
+		for i := start; i > stop; i += step {
+			v = append(v, i)
+		}
+		return v
+	}
+
+	if step <= 0 {
+		return v
+	}
+	for i := start; i < stop; i += step {
+		v = append(v, i)
+	}
+	return v
+}
+
+func floor(a interface{}) float64 {
+	aa := toFloat64(a)
+	return math.Floor(aa)
+}
+
+func ceil(a interface{}) float64 {
+	aa := toFloat64(a)
+	return math.Ceil(aa)
+}
+
+func round(a interface{}, p int, r_opt ...float64) float64 {
+	roundOn := .5
+	if len(r_opt) > 0 {
+		roundOn = r_opt[0]
+	}
+	val := toFloat64(a)
+	places := toFloat64(p)
+
+	var round float64
+	pow := math.Pow(10, places)
+	digit := pow * val
+	_, div := math.Modf(digit)
+	if div >= roundOn {
+		round = math.Ceil(digit)
+	} else {
+		round = math.Floor(digit)
+	}
+	return round / pow
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/reflect.go b/backend/vendor/github.com/Masterminds/sprig/reflect.go
new file mode 100644
index 00000000..8a65c132
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/reflect.go
@@ -0,0 +1,28 @@
+package sprig
+
+import (
+	"fmt"
+	"reflect"
+)
+
+// typeIs returns true if the src is the type named in target.
+func typeIs(target string, src interface{}) bool {
+	return target == typeOf(src)
+}
+
+func typeIsLike(target string, src interface{}) bool {
+	t := typeOf(src)
+	return target == t || "*"+target == t
+}
+
+func typeOf(src interface{}) string {
+	return fmt.Sprintf("%T", src)
+}
+
+func kindIs(target string, src interface{}) bool {
+	return target == kindOf(src)
+}
+
+func kindOf(src interface{}) string {
+	return reflect.ValueOf(src).Kind().String()
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/regex.go b/backend/vendor/github.com/Masterminds/sprig/regex.go
new file mode 100644
index 00000000..2016f663
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/regex.go
@@ -0,0 +1,35 @@
+package sprig
+
+import (
+	"regexp"
+)
+
+func regexMatch(regex string, s string) bool {
+	match, _ := regexp.MatchString(regex, s)
+	return match
+}
+
+func regexFindAll(regex string, s string, n int) []string {
+	r := regexp.MustCompile(regex)
+	return r.FindAllString(s, n)
+}
+
+func regexFind(regex string, s string) string {
+	r := regexp.MustCompile(regex)
+	return r.FindString(s)
+}
+
+func regexReplaceAll(regex string, s string, repl string) string {
+	r := regexp.MustCompile(regex)
+	return r.ReplaceAllString(s, repl)
+}
+
+func regexReplaceAllLiteral(regex string, s string, repl string) string {
+	r := regexp.MustCompile(regex)
+	return r.ReplaceAllLiteralString(s, repl)
+}
+
+func regexSplit(regex string, s string, n int) []string {
+	r := regexp.MustCompile(regex)
+	return r.Split(s, n)
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/semver.go b/backend/vendor/github.com/Masterminds/sprig/semver.go
new file mode 100644
index 00000000..c2bf8a1f
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/semver.go
@@ -0,0 +1,23 @@
+package sprig
+
+import (
+	sv2 "github.com/Masterminds/semver"
+)
+
+func semverCompare(constraint, version string) (bool, error) {
+	c, err := sv2.NewConstraint(constraint)
+	if err != nil {
+		return false, err
+	}
+
+	v, err := sv2.NewVersion(version)
+	if err != nil {
+		return false, err
+	}
+
+	return c.Check(v), nil
+}
+
+func semver(version string) (*sv2.Version, error) {
+	return sv2.NewVersion(version)
+}
diff --git a/backend/vendor/github.com/Masterminds/sprig/strings.go b/backend/vendor/github.com/Masterminds/sprig/strings.go
new file mode 100644
index 00000000..3a6967cf
--- /dev/null
+++ b/backend/vendor/github.com/Masterminds/sprig/strings.go
@@ -0,0 +1,210 @@
+package sprig
+
+import (
+	"encoding/base32"
+	"encoding/base64"
+	"fmt"
+	"reflect"
+	"strconv"
+	"strings"
+
+	util "github.com/aokoli/goutils"
+)
+
+func base64encode(v string) string {
+	return base64.StdEncoding.EncodeToString([]byte(v))
+}
+
+func base64decode(v string) string {
+	data, err := base64.StdEncoding.DecodeString(v)
+	if err != nil {
+		return err.Error()
+	}
+	return string(data)
+}
+
+func base32encode(v string) string {
+	return base32.StdEncoding.EncodeToString([]byte(v))
+}
+
+func base32decode(v string) string {
+	data, err := base32.StdEncoding.DecodeString(v)
+	if err != nil {
+		return err.Error()
+	}
+	return string(data)
+}
+
+func abbrev(width int, s string) string {
+	if width < 4 {
+		return s
+	}
+	r, _ := util.Abbreviate(s, width)
+	return r
+}
+
+func abbrevboth(left, right int, s string) string {
+	if right < 4 || left > 0 && right < 7 {
+		return s
+	}
+	r, _ := util.AbbreviateFull(s, left, right)
+	return r
+}
+func initials(s string) string {
+	// Wrap this just to eliminate the var args, which templates don't do well.
+	return util.Initials(s)
+}
+
+func randAlphaNumeric(count int) string {
+	// It is not possible, it appears, to actually generate an error here.
+	r, _ := util.RandomAlphaNumeric(count)
+	return r
+}
+
+func randAlpha(count int) string {
+	r, _ := util.RandomAlphabetic(count)
+	return r
+}
+
+func randAscii(count int) string {
+	r, _ := util.RandomAscii(count)
+	return r
+}
+
+func randNumeric(count int) string {
+	r, _ := util.RandomNumeric(count)
+	return r
+}
+
+func untitle(str string) string {
+	return util.Uncapitalize(str)
+}
+
+func quote(str ...interface{}) string {
+	out := make([]string, len(str))
+	for i, s := range str {
+		out[i] = fmt.Sprintf("%q", strval(s))
+	}
+	return strings.Join(out, " ")
+}
+
+func squote(str ...interface{}) string {
+	out := make([]string, len(str))
+	for i, s := range str {
+		out[i] = fmt.Sprintf("'%v'", s)
+	}
+	return strings.Join(out, " ")
+}
+
+func cat(v ...interface{}) string {
+	r := strings.TrimSpace(strings.Repeat("%v ", len(v)))
+	return fmt.Sprintf(r, v...)
+}
+
+func indent(spaces int, v string) string {
+	pad := strings.Repeat(" ", spaces)
+	return pad + strings.Replace(v, "\n", "\n"+pad, -1)
+}
+
+func nindent(spaces int, v string) string {
+	return "\n" + indent(spaces, v)
+}
+
+func replace(old, new, src string) string {
+	return strings.Replace(src, old, new, -1)
+}
+
+func plural(one, many string, count int) string {
+	if count == 1 {
+		return one
+	}
+	return many
+}
+
+func strslice(v interface{}) []string {
+	switch v := v.(type) {
+	case []string:
+		return v
+	case []interface{}:
+		l := len(v)
+		b := make([]string, l)
+		for i := 0; i < l; i++ {
+			b[i] = strval(v[i])
+		}
+		return b
+	default:
+		val := reflect.ValueOf(v)
+		switch val.Kind() {
+		case reflect.Array, reflect.Slice:
+			l := val.Len()
+			b := make([]string, l)
+			for i := 0; i < l; i++ {
+				b[i] = strval(val.Index(i).Interface())
+			}
+			return b
+		default:
+			return []string{strval(v)}
+		}
+	}
+}
+
+func strval(v interface{}) string {
+	switch v := v.(type) {
+	case string:
+		return v
+	case []byte:
+		return string(v)
+	case error:
+		return v.Error()
+	case fmt.Stringer:
+		return v.String()
+	default:
+		return fmt.Sprintf("%v", v)
+	}
+}
+
+func trunc(c int, s string) string {
+	if len(s) <= c {
+		return s
+	}
+	return s[0:c]
+}
+
+func join(sep string, v interface{}) string {
+	return strings.Join(strslice(v), sep)
+}
+
+func split(sep, orig string) map[string]string {
+	parts := strings.Split(orig, sep)
+	res := make(map[string]string, len(parts))
+	for i, v := range parts {
+		res["_"+strconv.Itoa(i)] = v
+	}
+	return res
+}
+
+func splitn(sep string, n int, orig string) map[string]string {
+	parts := strings.SplitN(orig, sep, n)
+	res := make(map[string]string, len(parts))
+	for i, v := range parts {
+		res["_"+strconv.Itoa(i)] = v
+	}
+	return res
+}
+
+// substring creates a substring of the given string.
+//
+// If start is < 0, this calls string[:length].
+//
+// If start is >= 0 and length < 0, this calls string[start:]
+//
+// Otherwise, this calls string[start, length].
+func substring(start, length int, s string) string {
+	if start < 0 {
+		return s[:length]
+	}
+	if length < 0 {
+		return s[start:]
+	}
+	return s[start:length]
+}
diff --git a/backend/vendor/github.com/aokoli/goutils/.travis.yml b/backend/vendor/github.com/aokoli/goutils/.travis.yml
new file mode 100644
index 00000000..4025e01e
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/.travis.yml
@@ -0,0 +1,18 @@
+language: go
+
+go:
+  - 1.6
+  - 1.7
+  - 1.8
+  - tip
+
+script:
+  - go test -v
+
+notifications:
+  webhooks:
+    urls:
+      - https://webhooks.gitter.im/e/06e3328629952dabe3e0
+    on_success: change  # options: [always|never|change] default: always
+    on_failure: always  # options: [always|never|change] default: always
+    on_start: never     # options: [always|never|change] default: always
diff --git a/backend/vendor/github.com/aokoli/goutils/CHANGELOG.md b/backend/vendor/github.com/aokoli/goutils/CHANGELOG.md
new file mode 100644
index 00000000..d700ec47
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/CHANGELOG.md
@@ -0,0 +1,8 @@
+# 1.0.1 (2017-05-31)
+
+## Fixed
+- #21: Fix generation of alphanumeric strings (thanks @dbarranco)
+
+# 1.0.0 (2014-04-30)
+
+- Initial release.
diff --git a/backend/vendor/github.com/aokoli/goutils/LICENSE.txt b/backend/vendor/github.com/aokoli/goutils/LICENSE.txt
new file mode 100644
index 00000000..d6456956
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/LICENSE.txt
@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
diff --git a/backend/vendor/github.com/aokoli/goutils/README.md b/backend/vendor/github.com/aokoli/goutils/README.md
new file mode 100644
index 00000000..163ffe72
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/README.md
@@ -0,0 +1,70 @@
+GoUtils
+===========
+[![Stability: Maintenance](https://masterminds.github.io/stability/maintenance.svg)](https://masterminds.github.io/stability/maintenance.html)
+[![GoDoc](https://godoc.org/github.com/Masterminds/goutils?status.png)](https://godoc.org/github.com/Masterminds/goutils) [![Build Status](https://travis-ci.org/Masterminds/goutils.svg?branch=master)](https://travis-ci.org/Masterminds/goutils) [![Build status](https://ci.appveyor.com/api/projects/status/sc2b1ew0m7f0aiju?svg=true)](https://ci.appveyor.com/project/mattfarina/goutils)
+
+
+GoUtils provides users with utility functions to manipulate strings in various ways. It is a Go implementation of some
+string manipulation libraries of Java Apache Commons. GoUtils includes the following Java Apache Commons classes:
+* WordUtils    
+* RandomStringUtils  
+* StringUtils (partial implementation)
+
+## Installation
+If you have Go set up on your system, from the GOPATH directory within the command line/terminal, enter this:
+
+	go get github.com/Masterminds/goutils
+
+If you do not have Go set up on your system, please follow the [Go installation directions from the documenation](http://golang.org/doc/install), and then follow the instructions above to install GoUtils.
+
+
+## Documentation
+GoUtils doc is available here: [![GoDoc](https://godoc.org/github.com/Masterminds/goutils?status.png)](https://godoc.org/github.com/Masterminds/goutils)
+
+
+## Usage
+The code snippets below show examples of how to use GoUtils. Some functions return errors while others do not. The first instance below, which does not return an error, is the `Initials` function (located within the `wordutils.go` file).
+
+    package main
+
+    import (
+        "fmt"
+    	"github.com/Masterminds/goutils"
+    )
+
+    func main() {
+
+    	// EXAMPLE 1: A goutils function which returns no errors
+        fmt.Println (goutils.Initials("John Doe Foo")) // Prints out "JDF"
+
+    }
+Some functions return errors mainly due to illegal arguements used as parameters. The code example below illustrates how to deal with function that returns an error. In this instance, the function is the `Random` function (located within the `randomstringutils.go` file).
+
+    package main
+
+    import (
+        "fmt"
+        "github.com/Masterminds/goutils"
+    )
+
+    func main() {
+
+        // EXAMPLE 2: A goutils function which returns an error
+        rand1, err1 := goutils.Random (-1, 0, 0, true, true)  
+
+        if err1 != nil {
+			fmt.Println(err1) // Prints out error message because -1 was entered as the first parameter in goutils.Random(...)
+		} else {
+			fmt.Println(rand1)
+		}
+
+    }
+
+## License
+GoUtils is licensed under the Apache License, Version 2.0. Please check the LICENSE.txt file or visit http://www.apache.org/licenses/LICENSE-2.0 for a copy of the license.
+
+## Issue Reporting
+Make suggestions or report issues using the Git issue tracker: https://github.com/Masterminds/goutils/issues
+
+## Website
+* [GoUtils webpage](http://Masterminds.github.io/goutils/)
diff --git a/backend/vendor/github.com/aokoli/goutils/appveyor.yml b/backend/vendor/github.com/aokoli/goutils/appveyor.yml
new file mode 100644
index 00000000..657564a8
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/appveyor.yml
@@ -0,0 +1,21 @@
+version: build-{build}.{branch}
+
+clone_folder: C:\gopath\src\github.com\Masterminds\goutils
+shallow_clone: true
+
+environment:
+  GOPATH: C:\gopath
+
+platform:
+  - x64
+
+build: off
+
+install:
+  - go version
+  - go env
+
+test_script:
+  - go test -v
+
+deploy: off
diff --git a/backend/vendor/github.com/aokoli/goutils/randomstringutils.go b/backend/vendor/github.com/aokoli/goutils/randomstringutils.go
new file mode 100644
index 00000000..1364e0ca
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/randomstringutils.go
@@ -0,0 +1,268 @@
+/*
+Copyright 2014 Alexander Okoli
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package goutils
+
+import (
+	"fmt"
+	"math"
+	"math/rand"
+	"regexp"
+	"time"
+	"unicode"
+)
+
+// RANDOM provides the time-based seed used to generate random numbers
+var RANDOM = rand.New(rand.NewSource(time.Now().UnixNano()))
+
+/*
+RandomNonAlphaNumeric creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of all characters (ASCII/Unicode values between 0 to 2,147,483,647 (math.MaxInt32)).
+
+Parameter:
+	count - the length of random string to create
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomNonAlphaNumeric(count int) (string, error) {
+	return RandomAlphaNumericCustom(count, false, false)
+}
+
+/*
+RandomAscii creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of characters whose ASCII value is between 32 and 126 (inclusive).
+
+Parameter:
+	count - the length of random string to create
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomAscii(count int) (string, error) {
+	return Random(count, 32, 127, false, false)
+}
+
+/*
+RandomNumeric creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of numeric characters.
+
+Parameter:
+	count - the length of random string to create
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomNumeric(count int) (string, error) {
+	return Random(count, 0, 0, false, true)
+}
+
+/*
+RandomAlphabetic creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of alpha-numeric characters as indicated by the arguments.
+
+Parameters:
+	count - the length of random string to create
+	letters - if true, generated string may include alphabetic characters
+	numbers - if true, generated string may include numeric characters
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomAlphabetic(count int) (string, error) {
+	return Random(count, 0, 0, true, false)
+}
+
+/*
+RandomAlphaNumeric creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of alpha-numeric characters.
+
+Parameter:
+	count - the length of random string to create
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomAlphaNumeric(count int) (string, error) {
+	RandomString, err := Random(count, 0, 0, true, true)
+	if err != nil {
+		return "", fmt.Errorf("Error: %s", err)
+	}
+	match, err := regexp.MatchString("([0-9]+)", RandomString)
+	if err != nil {
+		panic(err)
+	}
+
+	if !match {
+		//Get the position between 0 and the length of the string-1  to insert a random number
+		position := rand.Intn(count)
+		//Insert a random number between [0-9] in the position
+		RandomString = RandomString[:position] + string('0'+rand.Intn(10)) + RandomString[position+1:]
+		return RandomString, err
+	}
+	return RandomString, err
+
+}
+
+/*
+RandomAlphaNumericCustom creates a random string whose length is the number of characters specified.
+Characters will be chosen from the set of alpha-numeric characters as indicated by the arguments.
+
+Parameters:
+	count - the length of random string to create
+	letters - if true, generated string may include alphabetic characters
+	numbers - if true, generated string may include numeric characters
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func RandomAlphaNumericCustom(count int, letters bool, numbers bool) (string, error) {
+	return Random(count, 0, 0, letters, numbers)
+}
+
+/*
+Random creates a random string based on a variety of options, using default source of randomness.
+This method has exactly the same semantics as RandomSeed(int, int, int, bool, bool, []char, *rand.Rand), but
+instead of using an externally supplied source of randomness, it uses the internal *rand.Rand instance.
+
+Parameters:
+	count - the length of random string to create
+	start - the position in set of chars (ASCII/Unicode int) to start at
+	end - the position in set of chars (ASCII/Unicode int) to end before
+	letters - if true, generated string may include alphabetic characters
+	numbers - if true, generated string may include numeric characters
+	chars - the set of chars to choose randoms from. If nil, then it will use the set of all chars.
+
+Returns:
+	string - the random string
+	error - an error stemming from an invalid parameter within underlying function, RandomSeed(...)
+*/
+func Random(count int, start int, end int, letters bool, numbers bool, chars ...rune) (string, error) {
+	return RandomSeed(count, start, end, letters, numbers, chars, RANDOM)
+}
+
+/*
+RandomSeed creates a random string based on a variety of options, using supplied source of randomness.
+If the parameters start and end are both 0, start and end are set to ' ' and 'z', the ASCII printable characters, will be used,
+unless letters and numbers are both false, in which case, start and end are set to 0 and math.MaxInt32, respectively.
+If chars is not nil, characters stored in chars that are between start and end are chosen.
+This method accepts a user-supplied *rand.Rand instance to use as a source of randomness. By seeding a single *rand.Rand instance
+with a fixed seed and using it for each call, the same random sequence of strings can be generated repeatedly and predictably.
+
+Parameters:
+	count - the length of random string to create
+	start - the position in set of chars (ASCII/Unicode decimals) to start at
+	end - the position in set of chars (ASCII/Unicode decimals) to end before
+	letters - if true, generated string may include alphabetic characters
+	numbers - if true, generated string may include numeric characters
+	chars - the set of chars to choose randoms from. If nil, then it will use the set of all chars.
+	random - a source of randomness.
+
+Returns:
+	string - the random string
+	error - an error stemming from invalid parameters: if count < 0; or the provided chars array is empty; or end <= start; or end > len(chars)
+*/
+func RandomSeed(count int, start int, end int, letters bool, numbers bool, chars []rune, random *rand.Rand) (string, error) {
+
+	if count == 0 {
+		return "", nil
+	} else if count < 0 {
+		err := fmt.Errorf("randomstringutils illegal argument: Requested random string length %v is less than 0.", count) // equiv to err := errors.New("...")
+		return "", err
+	}
+	if chars != nil && len(chars) == 0 {
+		err := fmt.Errorf("randomstringutils illegal argument: The chars array must not be empty")
+		return "", err
+	}
+
+	if start == 0 && end == 0 {
+		if chars != nil {
+			end = len(chars)
+		} else {
+			if !letters && !numbers {
+				end = math.MaxInt32
+			} else {
+				end = 'z' + 1
+				start = ' '
+			}
+		}
+	} else {
+		if end <= start {
+			err := fmt.Errorf("randomstringutils illegal argument: Parameter end (%v) must be greater than start (%v)", end, start)
+			return "", err
+		}
+
+		if chars != nil && end > len(chars) {
+			err := fmt.Errorf("randomstringutils illegal argument: Parameter end (%v) cannot be greater than len(chars) (%v)", end, len(chars))
+			return "", err
+		}
+	}
+
+	buffer := make([]rune, count)
+	gap := end - start
+
+	// high-surrogates range, (\uD800-\uDBFF) = 55296 - 56319
+	//  low-surrogates range, (\uDC00-\uDFFF) = 56320 - 57343
+
+	for count != 0 {
+		count--
+		var ch rune
+		if chars == nil {
+			ch = rune(random.Intn(gap) + start)
+		} else {
+			ch = chars[random.Intn(gap)+start]
+		}
+
+		if letters && unicode.IsLetter(ch) || numbers && unicode.IsDigit(ch) || !letters && !numbers {
+			if ch >= 56320 && ch <= 57343 { // low surrogate range
+				if count == 0 {
+					count++
+				} else {
+					// Insert low surrogate
+					buffer[count] = ch
+					count--
+					// Insert high surrogate
+					buffer[count] = rune(55296 + random.Intn(128))
+				}
+			} else if ch >= 55296 && ch <= 56191 { // High surrogates range (Partial)
+				if count == 0 {
+					count++
+				} else {
+					// Insert low surrogate
+					buffer[count] = rune(56320 + random.Intn(128))
+					count--
+					// Insert high surrogate
+					buffer[count] = ch
+				}
+			} else if ch >= 56192 && ch <= 56319 {
+				// private high surrogate, skip it
+				count++
+			} else {
+				// not one of the surrogates*
+				buffer[count] = ch
+			}
+		} else {
+			count++
+		}
+	}
+	return string(buffer), nil
+}
diff --git a/backend/vendor/github.com/aokoli/goutils/stringutils.go b/backend/vendor/github.com/aokoli/goutils/stringutils.go
new file mode 100644
index 00000000..5037c451
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/stringutils.go
@@ -0,0 +1,224 @@
+/*
+Copyright 2014 Alexander Okoli
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package goutils
+
+import (
+	"bytes"
+	"fmt"
+	"strings"
+	"unicode"
+)
+
+// Typically returned by functions where a searched item cannot be found
+const INDEX_NOT_FOUND = -1
+
+/*
+Abbreviate abbreviates a string using ellipses. This will turn  the string "Now is the time for all good men" into "Now is the time for..."
+
+Specifically, the algorithm is as follows:
+
+    - If str is less than maxWidth characters long, return it.
+    - Else abbreviate it to (str[0:maxWidth - 3] + "...").
+    - If maxWidth is less than 4, return an illegal argument error.
+    - In no case will it return a string of length greater than maxWidth.
+
+Parameters:
+    str -  the string to check
+    maxWidth - maximum length of result string, must be at least 4
+
+Returns:
+    string - abbreviated string
+    error - if the width is too small
+*/
+func Abbreviate(str string, maxWidth int) (string, error) {
+	return AbbreviateFull(str, 0, maxWidth)
+}
+
+/*
+AbbreviateFull abbreviates a string using ellipses. This will turn the string "Now is the time for all good men" into "...is the time for..."
+This function works like Abbreviate(string, int), but allows you to specify a "left edge" offset. Note that this left edge is not
+necessarily going to be the leftmost character in the result, or the first character following the ellipses, but it will appear
+somewhere in the result.
+In no case will it return a string of length greater than maxWidth.
+
+Parameters:
+    str - the string to check
+    offset - left edge of source string
+    maxWidth - maximum length of result string, must be at least 4
+
+Returns:
+    string - abbreviated string
+    error - if the width is too small
+*/
+func AbbreviateFull(str string, offset int, maxWidth int) (string, error) {
+	if str == "" {
+		return "", nil
+	}
+	if maxWidth < 4 {
+		err := fmt.Errorf("stringutils illegal argument: Minimum abbreviation width is 4")
+		return "", err
+	}
+	if len(str) <= maxWidth {
+		return str, nil
+	}
+	if offset > len(str) {
+		offset = len(str)
+	}
+	if len(str)-offset < (maxWidth - 3) { // 15 - 5 < 10 - 3 =  10 < 7
+		offset = len(str) - (maxWidth - 3)
+	}
+	abrevMarker := "..."
+	if offset <= 4 {
+		return str[0:maxWidth-3] + abrevMarker, nil // str.substring(0, maxWidth - 3) + abrevMarker;
+	}
+	if maxWidth < 7 {
+		err := fmt.Errorf("stringutils illegal argument: Minimum abbreviation width with offset is 7")
+		return "", err
+	}
+	if (offset + maxWidth - 3) < len(str) { // 5 + (10-3) < 15 = 12 < 15
+		abrevStr, _ := Abbreviate(str[offset:len(str)], (maxWidth - 3))
+		return abrevMarker + abrevStr, nil // abrevMarker + abbreviate(str.substring(offset), maxWidth - 3);
+	}
+	return abrevMarker + str[(len(str)-(maxWidth-3)):len(str)], nil // abrevMarker + str.substring(str.length() - (maxWidth - 3));
+}
+
+/*
+DeleteWhiteSpace deletes all whitespaces from a string as defined by unicode.IsSpace(rune).
+It returns the string without whitespaces.
+
+Parameter:
+    str - the string to delete whitespace from, may be nil
+
+Returns:
+    the string without whitespaces
+*/
+func DeleteWhiteSpace(str string) string {
+	if str == "" {
+		return str
+	}
+	sz := len(str)
+	var chs bytes.Buffer
+	count := 0
+	for i := 0; i < sz; i++ {
+		ch := rune(str[i])
+		if !unicode.IsSpace(ch) {
+			chs.WriteRune(ch)
+			count++
+		}
+	}
+	if count == sz {
+		return str
+	}
+	return chs.String()
+}
+
+/*
+IndexOfDifference compares two strings, and returns the index at which the strings begin to differ.
+
+Parameters:
+    str1 - the first string
+    str2 - the second string
+
+Returns:
+    the index where str1 and str2 begin to differ; -1 if they are equal
+*/
+func IndexOfDifference(str1 string, str2 string) int {
+	if str1 == str2 {
+		return INDEX_NOT_FOUND
+	}
+	if IsEmpty(str1) || IsEmpty(str2) {
+		return 0
+	}
+	var i int
+	for i = 0; i < len(str1) && i < len(str2); i++ {
+		if rune(str1[i]) != rune(str2[i]) {
+			break
+		}
+	}
+	if i < len(str2) || i < len(str1) {
+		return i
+	}
+	return INDEX_NOT_FOUND
+}
+
+/*
+IsBlank checks if a string is whitespace or empty (""). Observe the following behavior:
+
+    goutils.IsBlank("")        = true
+    goutils.IsBlank(" ")       = true
+    goutils.IsBlank("bob")     = false
+    goutils.IsBlank("  bob  ") = false
+
+Parameter:
+    str - the string to check
+
+Returns:
+    true - if the string is whitespace or empty ("")
+*/
+func IsBlank(str string) bool {
+	strLen := len(str)
+	if str == "" || strLen == 0 {
+		return true
+	}
+	for i := 0; i < strLen; i++ {
+		if unicode.IsSpace(rune(str[i])) == false {
+			return false
+		}
+	}
+	return true
+}
+
+/*
+IndexOf returns the index of the first instance of sub in str, with the search beginning from the
+index start point specified. -1 is returned if sub is not present in str.
+
+An empty string ("") will return -1 (INDEX_NOT_FOUND). A negative start position is treated as zero.
+A start position greater than the string length returns -1.
+
+Parameters:
+    str - the string to check
+    sub - the substring to find
+    start - the start position; negative treated as zero
+
+Returns:
+    the first index where the sub string was found  (always >= start)
+*/
+func IndexOf(str string, sub string, start int) int {
+
+	if start < 0 {
+		start = 0
+	}
+
+	if len(str) < start {
+		return INDEX_NOT_FOUND
+	}
+
+	if IsEmpty(str) || IsEmpty(sub) {
+		return INDEX_NOT_FOUND
+	}
+
+	partialIndex := strings.Index(str[start:len(str)], sub)
+	if partialIndex == -1 {
+		return INDEX_NOT_FOUND
+	}
+	return partialIndex + start
+}
+
+// IsEmpty checks if a string is empty (""). Returns true if empty, and false otherwise.
+func IsEmpty(str string) bool {
+	return len(str) == 0
+}
diff --git a/backend/vendor/github.com/aokoli/goutils/wordutils.go b/backend/vendor/github.com/aokoli/goutils/wordutils.go
new file mode 100644
index 00000000..e92dd399
--- /dev/null
+++ b/backend/vendor/github.com/aokoli/goutils/wordutils.go
@@ -0,0 +1,356 @@
+/*
+Copyright 2014 Alexander Okoli
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+/*
+Package goutils provides utility functions to manipulate strings in various ways.
+The code snippets below show examples of how to use goutils. Some functions return
+errors while others do not, so usage would vary as a result.
+
+Example:
+
+    package main
+
+    import (
+        "fmt"
+        "github.com/aokoli/goutils"
+    )
+
+    func main() {
+
+        // EXAMPLE 1: A goutils function which returns no errors
+        fmt.Println (goutils.Initials("John Doe Foo")) // Prints out "JDF"
+
+
+
+        // EXAMPLE 2: A goutils function which returns an error
+        rand1, err1 := goutils.Random (-1, 0, 0, true, true)
+
+        if err1 != nil {
+            fmt.Println(err1) // Prints out error message because -1 was entered as the first parameter in goutils.Random(...)
+        } else {
+            fmt.Println(rand1)
+        }
+    }
+*/
+package goutils
+
+import (
+	"bytes"
+	"strings"
+	"unicode"
+)
+
+// VERSION indicates the current version of goutils
+const VERSION = "1.0.0"
+
+/*
+Wrap wraps a single line of text, identifying words by ' '.
+New lines will be separated by '\n'. Very long words, such as URLs will not be wrapped.
+Leading spaces on a new line are stripped. Trailing spaces are not stripped.
+
+Parameters:
+    str - the string to be word wrapped
+    wrapLength - the column (a column can fit only one character) to wrap the words at, less than 1 is treated as 1
+
+Returns:
+    a line with newlines inserted
+*/
+func Wrap(str string, wrapLength int) string {
+	return WrapCustom(str, wrapLength, "", false)
+}
+
+/*
+WrapCustom wraps a single line of text, identifying words by ' '.
+Leading spaces on a new line are stripped. Trailing spaces are not stripped.
+
+Parameters:
+    str - the string to be word wrapped
+    wrapLength - the column number (a column can fit only one character) to wrap the words at, less than 1 is treated as 1
+    newLineStr - the string to insert for a new line, "" uses '\n'
+    wrapLongWords - true if long words (such as URLs) should be wrapped
+
+Returns:
+    a line with newlines inserted
+*/
+func WrapCustom(str string, wrapLength int, newLineStr string, wrapLongWords bool) string {
+
+	if str == "" {
+		return ""
+	}
+	if newLineStr == "" {
+		newLineStr = "\n" // TODO Assumes "\n" is seperator. Explore SystemUtils.LINE_SEPARATOR from Apache Commons
+	}
+	if wrapLength < 1 {
+		wrapLength = 1
+	}
+
+	inputLineLength := len(str)
+	offset := 0
+
+	var wrappedLine bytes.Buffer
+
+	for inputLineLength-offset > wrapLength {
+
+		if rune(str[offset]) == ' ' {
+			offset++
+			continue
+		}
+
+		end := wrapLength + offset + 1
+		spaceToWrapAt := strings.LastIndex(str[offset:end], " ") + offset
+
+		if spaceToWrapAt >= offset {
+			// normal word (not longer than wrapLength)
+			wrappedLine.WriteString(str[offset:spaceToWrapAt])
+			wrappedLine.WriteString(newLineStr)
+			offset = spaceToWrapAt + 1
+
+		} else {
+			// long word or URL
+			if wrapLongWords {
+				end := wrapLength + offset
+				// long words are wrapped one line at a time
+				wrappedLine.WriteString(str[offset:end])
+				wrappedLine.WriteString(newLineStr)
+				offset += wrapLength
+			} else {
+				// long words aren't wrapped, just extended beyond limit
+				end := wrapLength + offset
+				spaceToWrapAt = strings.IndexRune(str[end:len(str)], ' ') + end
+				if spaceToWrapAt >= 0 {
+					wrappedLine.WriteString(str[offset:spaceToWrapAt])
+					wrappedLine.WriteString(newLineStr)
+					offset = spaceToWrapAt + 1
+				} else {
+					wrappedLine.WriteString(str[offset:len(str)])
+					offset = inputLineLength
+				}
+			}
+		}
+	}
+
+	wrappedLine.WriteString(str[offset:len(str)])
+
+	return wrappedLine.String()
+
+}
+
+/*
+Capitalize capitalizes all the delimiter separated words in a string. Only the first letter of each word is changed.
+To convert the rest of each word to lowercase at the same time, use CapitalizeFully(str string, delimiters ...rune).
+The delimiters represent a set of characters understood to separate words. The first string character
+and the first non-delimiter character after a delimiter will be capitalized. A "" input string returns "".
+Capitalization uses the Unicode title case, normally equivalent to upper case.
+
+Parameters:
+    str - the string to capitalize
+    delimiters - set of characters to determine capitalization, exclusion of this parameter means whitespace would be delimeter
+
+Returns:
+    capitalized string
+*/
+func Capitalize(str string, delimiters ...rune) string {
+
+	var delimLen int
+
+	if delimiters == nil {
+		delimLen = -1
+	} else {
+		delimLen = len(delimiters)
+	}
+
+	if str == "" || delimLen == 0 {
+		return str
+	}
+
+	buffer := []rune(str)
+	capitalizeNext := true
+	for i := 0; i < len(buffer); i++ {
+		ch := buffer[i]
+		if isDelimiter(ch, delimiters...) {
+			capitalizeNext = true
+		} else if capitalizeNext {
+			buffer[i] = unicode.ToTitle(ch)
+			capitalizeNext = false
+		}
+	}
+	return string(buffer)
+
+}
+
+/*
+CapitalizeFully converts all the delimiter separated words in a string into capitalized words, that is each word is made up of a
+titlecase character and then a series of lowercase characters. The delimiters represent a set of characters understood
+to separate words. The first string character and the first non-delimiter character after a delimiter will be capitalized.
+Capitalization uses the Unicode title case, normally equivalent to upper case.
+
+Parameters:
+    str - the string to capitalize fully
+    delimiters - set of characters to determine capitalization, exclusion of this parameter means whitespace would be delimeter
+
+Returns:
+    capitalized string
+*/
+func CapitalizeFully(str string, delimiters ...rune) string {
+
+	var delimLen int
+
+	if delimiters == nil {
+		delimLen = -1
+	} else {
+		delimLen = len(delimiters)
+	}
+
+	if str == "" || delimLen == 0 {
+		return str
+	}
+	str = strings.ToLower(str)
+	return Capitalize(str, delimiters...)
+}
+
+/*
+Uncapitalize uncapitalizes all the whitespace separated words in a string. Only the first letter of each word is changed.
+The delimiters represent a set of characters understood to separate words. The first string character and the first non-delimiter
+character after a delimiter will be uncapitalized. Whitespace is defined by unicode.IsSpace(char).
+
+Parameters:
+    str - the string to uncapitalize fully
+    delimiters - set of characters to determine capitalization, exclusion of this parameter means whitespace would be delimeter
+
+Returns:
+    uncapitalized string
+*/
+func Uncapitalize(str string, delimiters ...rune) string {
+
+	var delimLen int
+
+	if delimiters == nil {
+		delimLen = -1
+	} else {
+		delimLen = len(delimiters)
+	}
+
+	if str == "" || delimLen == 0 {
+		return str
+	}
+
+	buffer := []rune(str)
+	uncapitalizeNext := true // TODO Always makes capitalize/un apply to first char.
+	for i := 0; i < len(buffer); i++ {
+		ch := buffer[i]
+		if isDelimiter(ch, delimiters...) {
+			uncapitalizeNext = true
+		} else if uncapitalizeNext {
+			buffer[i] = unicode.ToLower(ch)
+			uncapitalizeNext = false
+		}
+	}
+	return string(buffer)
+}
+
+/*
+SwapCase swaps the case of a string using a word based algorithm.
+
+Conversion algorithm:
+
+    Upper case character converts to Lower case
+    Title case character converts to Lower case
+    Lower case character after Whitespace or at start converts to Title case
+    Other Lower case character converts to Upper case
+    Whitespace is defined by unicode.IsSpace(char).
+
+Parameters:
+    str - the string to swap case
+
+Returns:
+    the changed string
+*/
+func SwapCase(str string) string {
+	if str == "" {
+		return str
+	}
+	buffer := []rune(str)
+
+	whitespace := true
+
+	for i := 0; i < len(buffer); i++ {
+		ch := buffer[i]
+		if unicode.IsUpper(ch) {
+			buffer[i] = unicode.ToLower(ch)
+			whitespace = false
+		} else if unicode.IsTitle(ch) {
+			buffer[i] = unicode.ToLower(ch)
+			whitespace = false
+		} else if unicode.IsLower(ch) {
+			if whitespace {
+				buffer[i] = unicode.ToTitle(ch)
+				whitespace = false
+			} else {
+				buffer[i] = unicode.ToUpper(ch)
+			}
+		} else {
+			whitespace = unicode.IsSpace(ch)
+		}
+	}
+	return string(buffer)
+}
+
+/*
+Initials extracts the initial letters from each word in the string. The first letter of the string and all first
+letters after the defined delimiters are returned as a new string. Their case is not changed. If the delimiters
+parameter is excluded, then Whitespace is used. Whitespace is defined by unicode.IsSpacea(char). An empty delimiter array returns an empty string.
+
+Parameters:
+    str - the string to get initials from
+    delimiters - set of characters to determine words, exclusion of this parameter means whitespace would be delimeter
+Returns:
+    string of initial letters
+*/
+func Initials(str string, delimiters ...rune) string {
+	if str == "" {
+		return str
+	}
+	if delimiters != nil && len(delimiters) == 0 {
+		return ""
+	}
+	strLen := len(str)
+	var buf bytes.Buffer
+	lastWasGap := true
+	for i := 0; i < strLen; i++ {
+		ch := rune(str[i])
+
+		if isDelimiter(ch, delimiters...) {
+			lastWasGap = true
+		} else if lastWasGap {
+			buf.WriteRune(ch)
+			lastWasGap = false
+		}
+	}
+	return buf.String()
+}
+
+// private function (lower case func name)
+func isDelimiter(ch rune, delimiters ...rune) bool {
+	if delimiters == nil {
+		return unicode.IsSpace(ch)
+	}
+	for _, delimiter := range delimiters {
+		if ch == delimiter {
+			return true
+		}
+	}
+	return false
+}
diff --git a/backend/vendor/github.com/google/uuid/.travis.yml b/backend/vendor/github.com/google/uuid/.travis.yml
new file mode 100644
index 00000000..d8156a60
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/.travis.yml
@@ -0,0 +1,9 @@
+language: go
+
+go:
+  - 1.4.3
+  - 1.5.3
+  - tip
+
+script:
+  - go test -v ./...
diff --git a/backend/vendor/github.com/google/uuid/CONTRIBUTING.md b/backend/vendor/github.com/google/uuid/CONTRIBUTING.md
new file mode 100644
index 00000000..04fdf09f
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/CONTRIBUTING.md
@@ -0,0 +1,10 @@
+# How to contribute
+
+We definitely welcome patches and contribution to this project!
+
+### Legal requirements
+
+In order to protect both you and ourselves, you will need to sign the
+[Contributor License Agreement](https://cla.developers.google.com/clas).
+
+You may have already signed it for other Google projects.
diff --git a/backend/vendor/github.com/google/uuid/CONTRIBUTORS b/backend/vendor/github.com/google/uuid/CONTRIBUTORS
new file mode 100644
index 00000000..b4bb97f6
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/CONTRIBUTORS
@@ -0,0 +1,9 @@
+Paul Borman <borman@google.com>
+bmatsuo
+shawnps
+theory
+jboverfelt
+dsymonds
+cd1
+wallclockbuilder
+dansouza
diff --git a/backend/vendor/github.com/google/uuid/LICENSE b/backend/vendor/github.com/google/uuid/LICENSE
new file mode 100644
index 00000000..5dc68268
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/LICENSE
@@ -0,0 +1,27 @@
+Copyright (c) 2009,2014 Google Inc. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+   * Redistributions in binary form must reproduce the above
+copyright notice, this list of conditions and the following disclaimer
+in the documentation and/or other materials provided with the
+distribution.
+   * Neither the name of Google Inc. nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/backend/vendor/github.com/google/uuid/README.md b/backend/vendor/github.com/google/uuid/README.md
new file mode 100644
index 00000000..9d92c11f
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/README.md
@@ -0,0 +1,19 @@
+# uuid ![build status](https://travis-ci.org/google/uuid.svg?branch=master)
+The uuid package generates and inspects UUIDs based on
+[RFC 4122](http://tools.ietf.org/html/rfc4122)
+and DCE 1.1: Authentication and Security Services. 
+
+This package is based on the github.com/pborman/uuid package (previously named
+code.google.com/p/go-uuid).  It differs from these earlier packages in that
+a UUID is a 16 byte array rather than a byte slice.  One loss due to this
+change is the ability to represent an invalid UUID (vs a NIL UUID).
+
+###### Install
+`go get github.com/google/uuid`
+
+###### Documentation 
+[![GoDoc](https://godoc.org/github.com/google/uuid?status.svg)](http://godoc.org/github.com/google/uuid)
+
+Full `go doc` style documentation for the package can be viewed online without
+installing this package by using the GoDoc site here: 
+http://godoc.org/github.com/google/uuid
diff --git a/backend/vendor/github.com/google/uuid/dce.go b/backend/vendor/github.com/google/uuid/dce.go
new file mode 100644
index 00000000..fa820b9d
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/dce.go
@@ -0,0 +1,80 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"encoding/binary"
+	"fmt"
+	"os"
+)
+
+// A Domain represents a Version 2 domain
+type Domain byte
+
+// Domain constants for DCE Security (Version 2) UUIDs.
+const (
+	Person = Domain(0)
+	Group  = Domain(1)
+	Org    = Domain(2)
+)
+
+// NewDCESecurity returns a DCE Security (Version 2) UUID.
+//
+// The domain should be one of Person, Group or Org.
+// On a POSIX system the id should be the users UID for the Person
+// domain and the users GID for the Group.  The meaning of id for
+// the domain Org or on non-POSIX systems is site defined.
+//
+// For a given domain/id pair the same token may be returned for up to
+// 7 minutes and 10 seconds.
+func NewDCESecurity(domain Domain, id uint32) (UUID, error) {
+	uuid, err := NewUUID()
+	if err == nil {
+		uuid[6] = (uuid[6] & 0x0f) | 0x20 // Version 2
+		uuid[9] = byte(domain)
+		binary.BigEndian.PutUint32(uuid[0:], id)
+	}
+	return uuid, err
+}
+
+// NewDCEPerson returns a DCE Security (Version 2) UUID in the person
+// domain with the id returned by os.Getuid.
+//
+//  NewDCESecurity(Person, uint32(os.Getuid()))
+func NewDCEPerson() (UUID, error) {
+	return NewDCESecurity(Person, uint32(os.Getuid()))
+}
+
+// NewDCEGroup returns a DCE Security (Version 2) UUID in the group
+// domain with the id returned by os.Getgid.
+//
+//  NewDCESecurity(Group, uint32(os.Getgid()))
+func NewDCEGroup() (UUID, error) {
+	return NewDCESecurity(Group, uint32(os.Getgid()))
+}
+
+// Domain returns the domain for a Version 2 UUID.  Domains are only defined
+// for Version 2 UUIDs.
+func (uuid UUID) Domain() Domain {
+	return Domain(uuid[9])
+}
+
+// ID returns the id for a Version 2 UUID. IDs are only defined for Version 2
+// UUIDs.
+func (uuid UUID) ID() uint32 {
+	return binary.BigEndian.Uint32(uuid[0:4])
+}
+
+func (d Domain) String() string {
+	switch d {
+	case Person:
+		return "Person"
+	case Group:
+		return "Group"
+	case Org:
+		return "Org"
+	}
+	return fmt.Sprintf("Domain%d", int(d))
+}
diff --git a/backend/vendor/github.com/google/uuid/doc.go b/backend/vendor/github.com/google/uuid/doc.go
new file mode 100644
index 00000000..5b8a4b9a
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/doc.go
@@ -0,0 +1,12 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package uuid generates and inspects UUIDs.
+//
+// UUIDs are based on RFC 4122 and DCE 1.1: Authentication and Security
+// Services.
+//
+// A UUID is a 16 byte (128 bit) array.  UUIDs may be used as keys to
+// maps or compared directly.
+package uuid
diff --git a/backend/vendor/github.com/google/uuid/go.mod b/backend/vendor/github.com/google/uuid/go.mod
new file mode 100644
index 00000000..fc84cd79
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/go.mod
@@ -0,0 +1 @@
+module github.com/google/uuid
diff --git a/backend/vendor/github.com/google/uuid/hash.go b/backend/vendor/github.com/google/uuid/hash.go
new file mode 100644
index 00000000..b1746163
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/hash.go
@@ -0,0 +1,53 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"crypto/md5"
+	"crypto/sha1"
+	"hash"
+)
+
+// Well known namespace IDs and UUIDs
+var (
+	NameSpaceDNS  = Must(Parse("6ba7b810-9dad-11d1-80b4-00c04fd430c8"))
+	NameSpaceURL  = Must(Parse("6ba7b811-9dad-11d1-80b4-00c04fd430c8"))
+	NameSpaceOID  = Must(Parse("6ba7b812-9dad-11d1-80b4-00c04fd430c8"))
+	NameSpaceX500 = Must(Parse("6ba7b814-9dad-11d1-80b4-00c04fd430c8"))
+	Nil           UUID // empty UUID, all zeros
+)
+
+// NewHash returns a new UUID derived from the hash of space concatenated with
+// data generated by h.  The hash should be at least 16 byte in length.  The
+// first 16 bytes of the hash are used to form the UUID.  The version of the
+// UUID will be the lower 4 bits of version.  NewHash is used to implement
+// NewMD5 and NewSHA1.
+func NewHash(h hash.Hash, space UUID, data []byte, version int) UUID {
+	h.Reset()
+	h.Write(space[:])
+	h.Write(data)
+	s := h.Sum(nil)
+	var uuid UUID
+	copy(uuid[:], s)
+	uuid[6] = (uuid[6] & 0x0f) | uint8((version&0xf)<<4)
+	uuid[8] = (uuid[8] & 0x3f) | 0x80 // RFC 4122 variant
+	return uuid
+}
+
+// NewMD5 returns a new MD5 (Version 3) UUID based on the
+// supplied name space and data.  It is the same as calling:
+//
+//  NewHash(md5.New(), space, data, 3)
+func NewMD5(space UUID, data []byte) UUID {
+	return NewHash(md5.New(), space, data, 3)
+}
+
+// NewSHA1 returns a new SHA1 (Version 5) UUID based on the
+// supplied name space and data.  It is the same as calling:
+//
+//  NewHash(sha1.New(), space, data, 5)
+func NewSHA1(space UUID, data []byte) UUID {
+	return NewHash(sha1.New(), space, data, 5)
+}
diff --git a/backend/vendor/github.com/google/uuid/marshal.go b/backend/vendor/github.com/google/uuid/marshal.go
new file mode 100644
index 00000000..7f9e0c6c
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/marshal.go
@@ -0,0 +1,37 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import "fmt"
+
+// MarshalText implements encoding.TextMarshaler.
+func (uuid UUID) MarshalText() ([]byte, error) {
+	var js [36]byte
+	encodeHex(js[:], uuid)
+	return js[:], nil
+}
+
+// UnmarshalText implements encoding.TextUnmarshaler.
+func (uuid *UUID) UnmarshalText(data []byte) error {
+	id, err := ParseBytes(data)
+	if err == nil {
+		*uuid = id
+	}
+	return err
+}
+
+// MarshalBinary implements encoding.BinaryMarshaler.
+func (uuid UUID) MarshalBinary() ([]byte, error) {
+	return uuid[:], nil
+}
+
+// UnmarshalBinary implements encoding.BinaryUnmarshaler.
+func (uuid *UUID) UnmarshalBinary(data []byte) error {
+	if len(data) != 16 {
+		return fmt.Errorf("invalid UUID (got %d bytes)", len(data))
+	}
+	copy(uuid[:], data)
+	return nil
+}
diff --git a/backend/vendor/github.com/google/uuid/node.go b/backend/vendor/github.com/google/uuid/node.go
new file mode 100644
index 00000000..d651a2b0
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/node.go
@@ -0,0 +1,90 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"sync"
+)
+
+var (
+	nodeMu sync.Mutex
+	ifname string  // name of interface being used
+	nodeID [6]byte // hardware for version 1 UUIDs
+	zeroID [6]byte // nodeID with only 0's
+)
+
+// NodeInterface returns the name of the interface from which the NodeID was
+// derived.  The interface "user" is returned if the NodeID was set by
+// SetNodeID.
+func NodeInterface() string {
+	defer nodeMu.Unlock()
+	nodeMu.Lock()
+	return ifname
+}
+
+// SetNodeInterface selects the hardware address to be used for Version 1 UUIDs.
+// If name is "" then the first usable interface found will be used or a random
+// Node ID will be generated.  If a named interface cannot be found then false
+// is returned.
+//
+// SetNodeInterface never fails when name is "".
+func SetNodeInterface(name string) bool {
+	defer nodeMu.Unlock()
+	nodeMu.Lock()
+	return setNodeInterface(name)
+}
+
+func setNodeInterface(name string) bool {
+	iname, addr := getHardwareInterface(name) // null implementation for js
+	if iname != "" && addr != nil {
+		ifname = iname
+		copy(nodeID[:], addr)
+		return true
+	}
+
+	// We found no interfaces with a valid hardware address.  If name
+	// does not specify a specific interface generate a random Node ID
+	// (section 4.1.6)
+	if name == "" {
+		ifname = "random"
+		randomBits(nodeID[:])
+		return true
+	}
+	return false
+}
+
+// NodeID returns a slice of a copy of the current Node ID, setting the Node ID
+// if not already set.
+func NodeID() []byte {
+	defer nodeMu.Unlock()
+	nodeMu.Lock()
+	if nodeID == zeroID {
+		setNodeInterface("")
+	}
+	nid := nodeID
+	return nid[:]
+}
+
+// SetNodeID sets the Node ID to be used for Version 1 UUIDs.  The first 6 bytes
+// of id are used.  If id is less than 6 bytes then false is returned and the
+// Node ID is not set.
+func SetNodeID(id []byte) bool {
+	if len(id) < 6 {
+		return false
+	}
+	defer nodeMu.Unlock()
+	nodeMu.Lock()
+	copy(nodeID[:], id)
+	ifname = "user"
+	return true
+}
+
+// NodeID returns the 6 byte node id encoded in uuid.  It returns nil if uuid is
+// not valid.  The NodeID is only well defined for version 1 and 2 UUIDs.
+func (uuid UUID) NodeID() []byte {
+	var node [6]byte
+	copy(node[:], uuid[10:])
+	return node[:]
+}
diff --git a/backend/vendor/github.com/google/uuid/node_js.go b/backend/vendor/github.com/google/uuid/node_js.go
new file mode 100644
index 00000000..24b78edc
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/node_js.go
@@ -0,0 +1,12 @@
+// Copyright 2017 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// +build js
+
+package uuid
+
+// getHardwareInterface returns nil values for the JS version of the code.
+// This remvoves the "net" dependency, because it is not used in the browser.
+// Using the "net" library inflates the size of the transpiled JS code by 673k bytes.
+func getHardwareInterface(name string) (string, []byte) { return "", nil }
diff --git a/backend/vendor/github.com/google/uuid/node_net.go b/backend/vendor/github.com/google/uuid/node_net.go
new file mode 100644
index 00000000..0cbbcddb
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/node_net.go
@@ -0,0 +1,33 @@
+// Copyright 2017 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// +build !js
+
+package uuid
+
+import "net"
+
+var interfaces []net.Interface // cached list of interfaces
+
+// getHardwareInterface returns the name and hardware address of interface name.
+// If name is "" then the name and hardware address of one of the system's
+// interfaces is returned.  If no interfaces are found (name does not exist or
+// there are no interfaces) then "", nil is returned.
+//
+// Only addresses of at least 6 bytes are returned.
+func getHardwareInterface(name string) (string, []byte) {
+	if interfaces == nil {
+		var err error
+		interfaces, err = net.Interfaces()
+		if err != nil {
+			return "", nil
+		}
+	}
+	for _, ifs := range interfaces {
+		if len(ifs.HardwareAddr) >= 6 && (name == "" || name == ifs.Name) {
+			return ifs.Name, ifs.HardwareAddr
+		}
+	}
+	return "", nil
+}
diff --git a/backend/vendor/github.com/google/uuid/sql.go b/backend/vendor/github.com/google/uuid/sql.go
new file mode 100644
index 00000000..f326b54d
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/sql.go
@@ -0,0 +1,59 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"database/sql/driver"
+	"fmt"
+)
+
+// Scan implements sql.Scanner so UUIDs can be read from databases transparently
+// Currently, database types that map to string and []byte are supported. Please
+// consult database-specific driver documentation for matching types.
+func (uuid *UUID) Scan(src interface{}) error {
+	switch src := src.(type) {
+	case nil:
+		return nil
+
+	case string:
+		// if an empty UUID comes from a table, we return a null UUID
+		if src == "" {
+			return nil
+		}
+
+		// see Parse for required string format
+		u, err := Parse(src)
+		if err != nil {
+			return fmt.Errorf("Scan: %v", err)
+		}
+
+		*uuid = u
+
+	case []byte:
+		// if an empty UUID comes from a table, we return a null UUID
+		if len(src) == 0 {
+			return nil
+		}
+
+		// assumes a simple slice of bytes if 16 bytes
+		// otherwise attempts to parse
+		if len(src) != 16 {
+			return uuid.Scan(string(src))
+		}
+		copy((*uuid)[:], src)
+
+	default:
+		return fmt.Errorf("Scan: unable to scan type %T into UUID", src)
+	}
+
+	return nil
+}
+
+// Value implements sql.Valuer so that UUIDs can be written to databases
+// transparently. Currently, UUIDs map to strings. Please consult
+// database-specific driver documentation for matching types.
+func (uuid UUID) Value() (driver.Value, error) {
+	return uuid.String(), nil
+}
diff --git a/backend/vendor/github.com/google/uuid/time.go b/backend/vendor/github.com/google/uuid/time.go
new file mode 100644
index 00000000..e6ef06cd
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/time.go
@@ -0,0 +1,123 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"encoding/binary"
+	"sync"
+	"time"
+)
+
+// A Time represents a time as the number of 100's of nanoseconds since 15 Oct
+// 1582.
+type Time int64
+
+const (
+	lillian    = 2299160          // Julian day of 15 Oct 1582
+	unix       = 2440587          // Julian day of 1 Jan 1970
+	epoch      = unix - lillian   // Days between epochs
+	g1582      = epoch * 86400    // seconds between epochs
+	g1582ns100 = g1582 * 10000000 // 100s of a nanoseconds between epochs
+)
+
+var (
+	timeMu   sync.Mutex
+	lasttime uint64 // last time we returned
+	clockSeq uint16 // clock sequence for this run
+
+	timeNow = time.Now // for testing
+)
+
+// UnixTime converts t the number of seconds and nanoseconds using the Unix
+// epoch of 1 Jan 1970.
+func (t Time) UnixTime() (sec, nsec int64) {
+	sec = int64(t - g1582ns100)
+	nsec = (sec % 10000000) * 100
+	sec /= 10000000
+	return sec, nsec
+}
+
+// GetTime returns the current Time (100s of nanoseconds since 15 Oct 1582) and
+// clock sequence as well as adjusting the clock sequence as needed.  An error
+// is returned if the current time cannot be determined.
+func GetTime() (Time, uint16, error) {
+	defer timeMu.Unlock()
+	timeMu.Lock()
+	return getTime()
+}
+
+func getTime() (Time, uint16, error) {
+	t := timeNow()
+
+	// If we don't have a clock sequence already, set one.
+	if clockSeq == 0 {
+		setClockSequence(-1)
+	}
+	now := uint64(t.UnixNano()/100) + g1582ns100
+
+	// If time has gone backwards with this clock sequence then we
+	// increment the clock sequence
+	if now <= lasttime {
+		clockSeq = ((clockSeq + 1) & 0x3fff) | 0x8000
+	}
+	lasttime = now
+	return Time(now), clockSeq, nil
+}
+
+// ClockSequence returns the current clock sequence, generating one if not
+// already set.  The clock sequence is only used for Version 1 UUIDs.
+//
+// The uuid package does not use global static storage for the clock sequence or
+// the last time a UUID was generated.  Unless SetClockSequence is used, a new
+// random clock sequence is generated the first time a clock sequence is
+// requested by ClockSequence, GetTime, or NewUUID.  (section 4.2.1.1)
+func ClockSequence() int {
+	defer timeMu.Unlock()
+	timeMu.Lock()
+	return clockSequence()
+}
+
+func clockSequence() int {
+	if clockSeq == 0 {
+		setClockSequence(-1)
+	}
+	return int(clockSeq & 0x3fff)
+}
+
+// SetClockSequence sets the clock sequence to the lower 14 bits of seq.  Setting to
+// -1 causes a new sequence to be generated.
+func SetClockSequence(seq int) {
+	defer timeMu.Unlock()
+	timeMu.Lock()
+	setClockSequence(seq)
+}
+
+func setClockSequence(seq int) {
+	if seq == -1 {
+		var b [2]byte
+		randomBits(b[:]) // clock sequence
+		seq = int(b[0])<<8 | int(b[1])
+	}
+	oldSeq := clockSeq
+	clockSeq = uint16(seq&0x3fff) | 0x8000 // Set our variant
+	if oldSeq != clockSeq {
+		lasttime = 0
+	}
+}
+
+// Time returns the time in 100s of nanoseconds since 15 Oct 1582 encoded in
+// uuid.  The time is only defined for version 1 and 2 UUIDs.
+func (uuid UUID) Time() Time {
+	time := int64(binary.BigEndian.Uint32(uuid[0:4]))
+	time |= int64(binary.BigEndian.Uint16(uuid[4:6])) << 32
+	time |= int64(binary.BigEndian.Uint16(uuid[6:8])&0xfff) << 48
+	return Time(time)
+}
+
+// ClockSequence returns the clock sequence encoded in uuid.
+// The clock sequence is only well defined for version 1 and 2 UUIDs.
+func (uuid UUID) ClockSequence() int {
+	return int(binary.BigEndian.Uint16(uuid[8:10])) & 0x3fff
+}
diff --git a/backend/vendor/github.com/google/uuid/util.go b/backend/vendor/github.com/google/uuid/util.go
new file mode 100644
index 00000000..5ea6c737
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/util.go
@@ -0,0 +1,43 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"io"
+)
+
+// randomBits completely fills slice b with random data.
+func randomBits(b []byte) {
+	if _, err := io.ReadFull(rander, b); err != nil {
+		panic(err.Error()) // rand should never fail
+	}
+}
+
+// xvalues returns the value of a byte as a hexadecimal digit or 255.
+var xvalues = [256]byte{
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 255, 255, 255, 255, 255, 255,
+	255, 10, 11, 12, 13, 14, 15, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 10, 11, 12, 13, 14, 15, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+	255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+}
+
+// xtob converts hex characters x1 and x2 into a byte.
+func xtob(x1, x2 byte) (byte, bool) {
+	b1 := xvalues[x1]
+	b2 := xvalues[x2]
+	return (b1 << 4) | b2, b1 != 255 && b2 != 255
+}
diff --git a/backend/vendor/github.com/google/uuid/uuid.go b/backend/vendor/github.com/google/uuid/uuid.go
new file mode 100644
index 00000000..524404cc
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/uuid.go
@@ -0,0 +1,245 @@
+// Copyright 2018 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"bytes"
+	"crypto/rand"
+	"encoding/hex"
+	"errors"
+	"fmt"
+	"io"
+	"strings"
+)
+
+// A UUID is a 128 bit (16 byte) Universal Unique IDentifier as defined in RFC
+// 4122.
+type UUID [16]byte
+
+// A Version represents a UUID's version.
+type Version byte
+
+// A Variant represents a UUID's variant.
+type Variant byte
+
+// Constants returned by Variant.
+const (
+	Invalid   = Variant(iota) // Invalid UUID
+	RFC4122                   // The variant specified in RFC4122
+	Reserved                  // Reserved, NCS backward compatibility.
+	Microsoft                 // Reserved, Microsoft Corporation backward compatibility.
+	Future                    // Reserved for future definition.
+)
+
+var rander = rand.Reader // random function
+
+// Parse decodes s into a UUID or returns an error.  Both the standard UUID
+// forms of xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx and
+// urn:uuid:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx are decoded as well as the
+// Microsoft encoding {xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx} and the raw hex
+// encoding: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.
+func Parse(s string) (UUID, error) {
+	var uuid UUID
+	switch len(s) {
+	// xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+	case 36:
+
+	// urn:uuid:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+	case 36 + 9:
+		if strings.ToLower(s[:9]) != "urn:uuid:" {
+			return uuid, fmt.Errorf("invalid urn prefix: %q", s[:9])
+		}
+		s = s[9:]
+
+	// {xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx}
+	case 36 + 2:
+		s = s[1:]
+
+	// xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+	case 32:
+		var ok bool
+		for i := range uuid {
+			uuid[i], ok = xtob(s[i*2], s[i*2+1])
+			if !ok {
+				return uuid, errors.New("invalid UUID format")
+			}
+		}
+		return uuid, nil
+	default:
+		return uuid, fmt.Errorf("invalid UUID length: %d", len(s))
+	}
+	// s is now at least 36 bytes long
+	// it must be of the form  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+	if s[8] != '-' || s[13] != '-' || s[18] != '-' || s[23] != '-' {
+		return uuid, errors.New("invalid UUID format")
+	}
+	for i, x := range [16]int{
+		0, 2, 4, 6,
+		9, 11,
+		14, 16,
+		19, 21,
+		24, 26, 28, 30, 32, 34} {
+		v, ok := xtob(s[x], s[x+1])
+		if !ok {
+			return uuid, errors.New("invalid UUID format")
+		}
+		uuid[i] = v
+	}
+	return uuid, nil
+}
+
+// ParseBytes is like Parse, except it parses a byte slice instead of a string.
+func ParseBytes(b []byte) (UUID, error) {
+	var uuid UUID
+	switch len(b) {
+	case 36: // xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+	case 36 + 9: // urn:uuid:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+		if !bytes.Equal(bytes.ToLower(b[:9]), []byte("urn:uuid:")) {
+			return uuid, fmt.Errorf("invalid urn prefix: %q", b[:9])
+		}
+		b = b[9:]
+	case 36 + 2: // {xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx}
+		b = b[1:]
+	case 32: // xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+		var ok bool
+		for i := 0; i < 32; i += 2 {
+			uuid[i/2], ok = xtob(b[i], b[i+1])
+			if !ok {
+				return uuid, errors.New("invalid UUID format")
+			}
+		}
+		return uuid, nil
+	default:
+		return uuid, fmt.Errorf("invalid UUID length: %d", len(b))
+	}
+	// s is now at least 36 bytes long
+	// it must be of the form  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+	if b[8] != '-' || b[13] != '-' || b[18] != '-' || b[23] != '-' {
+		return uuid, errors.New("invalid UUID format")
+	}
+	for i, x := range [16]int{
+		0, 2, 4, 6,
+		9, 11,
+		14, 16,
+		19, 21,
+		24, 26, 28, 30, 32, 34} {
+		v, ok := xtob(b[x], b[x+1])
+		if !ok {
+			return uuid, errors.New("invalid UUID format")
+		}
+		uuid[i] = v
+	}
+	return uuid, nil
+}
+
+// MustParse is like Parse but panics if the string cannot be parsed.
+// It simplifies safe initialization of global variables holding compiled UUIDs.
+func MustParse(s string) UUID {
+	uuid, err := Parse(s)
+	if err != nil {
+		panic(`uuid: Parse(` + s + `): ` + err.Error())
+	}
+	return uuid
+}
+
+// FromBytes creates a new UUID from a byte slice. Returns an error if the slice
+// does not have a length of 16. The bytes are copied from the slice.
+func FromBytes(b []byte) (uuid UUID, err error) {
+	err = uuid.UnmarshalBinary(b)
+	return uuid, err
+}
+
+// Must returns uuid if err is nil and panics otherwise.
+func Must(uuid UUID, err error) UUID {
+	if err != nil {
+		panic(err)
+	}
+	return uuid
+}
+
+// String returns the string form of uuid, xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
+// , or "" if uuid is invalid.
+func (uuid UUID) String() string {
+	var buf [36]byte
+	encodeHex(buf[:], uuid)
+	return string(buf[:])
+}
+
+// URN returns the RFC 2141 URN form of uuid,
+// urn:uuid:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx,  or "" if uuid is invalid.
+func (uuid UUID) URN() string {
+	var buf [36 + 9]byte
+	copy(buf[:], "urn:uuid:")
+	encodeHex(buf[9:], uuid)
+	return string(buf[:])
+}
+
+func encodeHex(dst []byte, uuid UUID) {
+	hex.Encode(dst, uuid[:4])
+	dst[8] = '-'
+	hex.Encode(dst[9:13], uuid[4:6])
+	dst[13] = '-'
+	hex.Encode(dst[14:18], uuid[6:8])
+	dst[18] = '-'
+	hex.Encode(dst[19:23], uuid[8:10])
+	dst[23] = '-'
+	hex.Encode(dst[24:], uuid[10:])
+}
+
+// Variant returns the variant encoded in uuid.
+func (uuid UUID) Variant() Variant {
+	switch {
+	case (uuid[8] & 0xc0) == 0x80:
+		return RFC4122
+	case (uuid[8] & 0xe0) == 0xc0:
+		return Microsoft
+	case (uuid[8] & 0xe0) == 0xe0:
+		return Future
+	default:
+		return Reserved
+	}
+}
+
+// Version returns the version of uuid.
+func (uuid UUID) Version() Version {
+	return Version(uuid[6] >> 4)
+}
+
+func (v Version) String() string {
+	if v > 15 {
+		return fmt.Sprintf("BAD_VERSION_%d", v)
+	}
+	return fmt.Sprintf("VERSION_%d", v)
+}
+
+func (v Variant) String() string {
+	switch v {
+	case RFC4122:
+		return "RFC4122"
+	case Reserved:
+		return "Reserved"
+	case Microsoft:
+		return "Microsoft"
+	case Future:
+		return "Future"
+	case Invalid:
+		return "Invalid"
+	}
+	return fmt.Sprintf("BadVariant%d", int(v))
+}
+
+// SetRand sets the random number generator to r, which implements io.Reader.
+// If r.Read returns an error when the package requests random data then
+// a panic will be issued.
+//
+// Calling SetRand with nil sets the random number generator to the default
+// generator.
+func SetRand(r io.Reader) {
+	if r == nil {
+		rander = rand.Reader
+		return
+	}
+	rander = r
+}
diff --git a/backend/vendor/github.com/google/uuid/version1.go b/backend/vendor/github.com/google/uuid/version1.go
new file mode 100644
index 00000000..199a1ac6
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/version1.go
@@ -0,0 +1,44 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import (
+	"encoding/binary"
+)
+
+// NewUUID returns a Version 1 UUID based on the current NodeID and clock
+// sequence, and the current time.  If the NodeID has not been set by SetNodeID
+// or SetNodeInterface then it will be set automatically.  If the NodeID cannot
+// be set NewUUID returns nil.  If clock sequence has not been set by
+// SetClockSequence then it will be set automatically.  If GetTime fails to
+// return the current NewUUID returns nil and an error.
+//
+// In most cases, New should be used.
+func NewUUID() (UUID, error) {
+	nodeMu.Lock()
+	if nodeID == zeroID {
+		setNodeInterface("")
+	}
+	nodeMu.Unlock()
+
+	var uuid UUID
+	now, seq, err := GetTime()
+	if err != nil {
+		return uuid, err
+	}
+
+	timeLow := uint32(now & 0xffffffff)
+	timeMid := uint16((now >> 32) & 0xffff)
+	timeHi := uint16((now >> 48) & 0x0fff)
+	timeHi |= 0x1000 // Version 1
+
+	binary.BigEndian.PutUint32(uuid[0:], timeLow)
+	binary.BigEndian.PutUint16(uuid[4:], timeMid)
+	binary.BigEndian.PutUint16(uuid[6:], timeHi)
+	binary.BigEndian.PutUint16(uuid[8:], seq)
+	copy(uuid[10:], nodeID[:])
+
+	return uuid, nil
+}
diff --git a/backend/vendor/github.com/google/uuid/version4.go b/backend/vendor/github.com/google/uuid/version4.go
new file mode 100644
index 00000000..84af91c9
--- /dev/null
+++ b/backend/vendor/github.com/google/uuid/version4.go
@@ -0,0 +1,38 @@
+// Copyright 2016 Google Inc.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package uuid
+
+import "io"
+
+// New creates a new random UUID or panics.  New is equivalent to
+// the expression
+//
+//    uuid.Must(uuid.NewRandom())
+func New() UUID {
+	return Must(NewRandom())
+}
+
+// NewRandom returns a Random (Version 4) UUID.
+//
+// The strength of the UUIDs is based on the strength of the crypto/rand
+// package.
+//
+// A note about uniqueness derived from the UUID Wikipedia entry:
+//
+//  Randomly generated UUIDs have 122 random bits.  One's annual risk of being
+//  hit by a meteorite is estimated to be one chance in 17 billion, that
+//  means the probability is about 0.00000000006 (6 × 10−11),
+//  equivalent to the odds of creating a few tens of trillions of UUIDs in a
+//  year and having one duplicate.
+func NewRandom() (UUID, error) {
+	var uuid UUID
+	_, err := io.ReadFull(rander, uuid[:])
+	if err != nil {
+		return Nil, err
+	}
+	uuid[6] = (uuid[6] & 0x0f) | 0x40 // Version 4
+	uuid[8] = (uuid[8] & 0x3f) | 0x80 // Variant is 10
+	return uuid, nil
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/.gitignore b/backend/vendor/github.com/huandu/xstrings/.gitignore
new file mode 100644
index 00000000..daf913b1
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/.gitignore
@@ -0,0 +1,24 @@
+# Compiled Object files, Static and Dynamic libs (Shared Objects)
+*.o
+*.a
+*.so
+
+# Folders
+_obj
+_test
+
+# Architecture specific extensions/prefixes
+*.[568vq]
+[568vq].out
+
+*.cgo1.go
+*.cgo2.c
+_cgo_defun.c
+_cgo_gotypes.go
+_cgo_export.*
+
+_testmain.go
+
+*.exe
+*.test
+*.prof
diff --git a/backend/vendor/github.com/huandu/xstrings/.travis.yml b/backend/vendor/github.com/huandu/xstrings/.travis.yml
new file mode 100644
index 00000000..4f2ee4d9
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/.travis.yml
@@ -0,0 +1 @@
+language: go
diff --git a/backend/vendor/github.com/huandu/xstrings/CONTRIBUTING.md b/backend/vendor/github.com/huandu/xstrings/CONTRIBUTING.md
new file mode 100644
index 00000000..d7b4b8d5
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/CONTRIBUTING.md
@@ -0,0 +1,23 @@
+# Contributing #
+
+Thanks for your contribution in advance. No matter what you will contribute to this project, pull request or bug report or feature discussion, it's always highly appreciated.
+
+## New API or feature ##
+
+I want to speak more about how to add new functions to this package.
+
+Package `xstring` is a collection of useful string functions which should be implemented in Go. It's a bit subject to say which function should be included and which should not. I set up following rules in order to make it clear and as objective as possible.
+
+* Rule 1: Only string algorithm, which takes string as input, can be included.
+* Rule 2: If a function has been implemented in package `string`, it must not be included.
+* Rule 3: If a function is not language neutral, it must not be included.
+* Rule 4: If a function is a part of standard library in other languages, it can be included.
+* Rule 5: If a function is quite useful in some famous framework or library, it can be included.
+
+New function must be discussed in project issues before submitting any code. If a pull request with new functions is sent without any ref issue, it will be rejected.
+
+## Pull request ##
+
+Pull request is always welcome. Just make sure you have run `go fmt` and all test cases passed before submit.
+
+If the pull request is to add a new API or feature, don't forget to update README.md and add new API in function list.
diff --git a/backend/vendor/github.com/huandu/xstrings/LICENSE b/backend/vendor/github.com/huandu/xstrings/LICENSE
new file mode 100644
index 00000000..27017725
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/LICENSE
@@ -0,0 +1,22 @@
+The MIT License (MIT)
+
+Copyright (c) 2015 Huan Du
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
diff --git a/backend/vendor/github.com/huandu/xstrings/README.md b/backend/vendor/github.com/huandu/xstrings/README.md
new file mode 100644
index 00000000..a4a36f45
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/README.md
@@ -0,0 +1,115 @@
+# xstrings #
+
+[![Build Status](https://travis-ci.org/huandu/xstrings.svg?branch=master)](https://travis-ci.org/huandu/xstrings)
+[![GoDoc](https://godoc.org/github.com/huandu/xstrings?status.svg)](https://godoc.org/github.com/huandu/xstrings)
+
+Go package [xstrings](https://godoc.org/github.com/huandu/xstrings) is a collection of string functions, which are widely used in other languages but absent in Go package [strings](http://golang.org/pkg/strings).
+
+All functions are well tested and carefully tuned for performance.
+
+## Propose a new function ##
+
+Please review [contributing guideline](CONTRIBUTING.md) and [create new issue](https://github.com/huandu/xstrings/issues) to state why it should be included.
+
+## Install ##
+
+Use `go get` to install this library.
+
+	go get github.com/huandu/xstrings
+
+## API document ##
+
+See [GoDoc](https://godoc.org/github.com/huandu/xstrings) for full document.
+
+## Function list ##
+
+Go functions have a unique naming style. One, who has experience in other language but new in Go, may have difficulties to find out right string function to use.
+
+Here is a list of functions in [strings](http://golang.org/pkg/strings) and [xstrings](https://godoc.org/github.com/huandu/xstrings) with enough extra information about how to map these functions to their friends in other languages. Hope this list could be helpful for fresh gophers.
+
+### Package `xstrings` functions ###
+
+*Keep this table sorted by Function in ascending order.*
+
+| Function | Friends | # |
+| -------- | ------- | --- |
+| [Center](https://godoc.org/github.com/huandu/xstrings#Center) | `str.center` in Python; `String#center` in Ruby | [#30](https://github.com/huandu/xstrings/issues/30) |
+| [Count](https://godoc.org/github.com/huandu/xstrings#Count) | `String#count` in Ruby | [#16](https://github.com/huandu/xstrings/issues/16) |
+| [Delete](https://godoc.org/github.com/huandu/xstrings#Delete) | `String#delete` in Ruby | [#17](https://github.com/huandu/xstrings/issues/17) |
+| [ExpandTabs](https://godoc.org/github.com/huandu/xstrings#ExpandTabs) | `str.expandtabs` in Python | [#27](https://github.com/huandu/xstrings/issues/27) |
+| [FirstRuneToLower](https://godoc.org/github.com/huandu/xstrings#FirstRuneToLower) | `lcfirst` in PHP or Perl | [#15](https://github.com/huandu/xstrings/issues/15) |
+| [FirstRuneToUpper](https://godoc.org/github.com/huandu/xstrings#FirstRuneToUpper) | `String#capitalize` in Ruby; `ucfirst` in PHP or Perl | [#15](https://github.com/huandu/xstrings/issues/15) |
+| [Insert](https://godoc.org/github.com/huandu/xstrings#Insert) | `String#insert` in Ruby | [#18](https://github.com/huandu/xstrings/issues/18) |
+| [LastPartition](https://godoc.org/github.com/huandu/xstrings#LastPartition) | `str.rpartition` in Python; `String#rpartition` in Ruby | [#19](https://github.com/huandu/xstrings/issues/19) |
+| [LeftJustify](https://godoc.org/github.com/huandu/xstrings#LeftJustify) | `str.ljust` in Python; `String#ljust` in Ruby | [#28](https://github.com/huandu/xstrings/issues/28) |
+| [Len](https://godoc.org/github.com/huandu/xstrings#Len) | `mb_strlen` in PHP | [#23](https://github.com/huandu/xstrings/issues/23) |
+| [Partition](https://godoc.org/github.com/huandu/xstrings#Partition) | `str.partition` in Python; `String#partition` in Ruby | [#10](https://github.com/huandu/xstrings/issues/10) |
+| [Reverse](https://godoc.org/github.com/huandu/xstrings#Reverse) | `String#reverse` in Ruby; `strrev` in PHP; `reverse` in Perl | [#7](https://github.com/huandu/xstrings/issues/7) |
+| [RightJustify](https://godoc.org/github.com/huandu/xstrings#RightJustify) | `str.rjust` in Python; `String#rjust` in Ruby | [#29](https://github.com/huandu/xstrings/issues/29) |
+| [RuneWidth](https://godoc.org/github.com/huandu/xstrings#RuneWidth) | - | [#27](https://github.com/huandu/xstrings/issues/27) |
+| [Scrub](https://godoc.org/github.com/huandu/xstrings#Scrub) | `String#scrub` in Ruby | [#20](https://github.com/huandu/xstrings/issues/20) |
+| [Shuffle](https://godoc.org/github.com/huandu/xstrings#Shuffle) | `str_shuffle` in PHP | [#13](https://github.com/huandu/xstrings/issues/13) |
+| [ShuffleSource](https://godoc.org/github.com/huandu/xstrings#ShuffleSource) | `str_shuffle` in PHP | [#13](https://github.com/huandu/xstrings/issues/13) |
+| [Slice](https://godoc.org/github.com/huandu/xstrings#Slice) | `mb_substr` in PHP | [#9](https://github.com/huandu/xstrings/issues/9) |
+| [Squeeze](https://godoc.org/github.com/huandu/xstrings#Squeeze) | `String#squeeze` in Ruby | [#11](https://github.com/huandu/xstrings/issues/11) |
+| [Successor](https://godoc.org/github.com/huandu/xstrings#Successor) | `String#succ` or `String#next` in Ruby | [#22](https://github.com/huandu/xstrings/issues/22) |
+| [SwapCase](https://godoc.org/github.com/huandu/xstrings#SwapCase) | `str.swapcase` in Python; `String#swapcase` in Ruby | [#12](https://github.com/huandu/xstrings/issues/12) |
+| [ToCamelCase](https://godoc.org/github.com/huandu/xstrings#ToCamelCase) | `String#camelize` in RoR | [#1](https://github.com/huandu/xstrings/issues/1) |
+| [ToKebab](https://godoc.org/github.com/huandu/xstrings#ToKebabCase) | - | [#41](https://github.com/huandu/xstrings/issues/41) |
+| [ToSnakeCase](https://godoc.org/github.com/huandu/xstrings#ToSnakeCase) | `String#underscore` in RoR | [#1](https://github.com/huandu/xstrings/issues/1) |
+| [Translate](https://godoc.org/github.com/huandu/xstrings#Translate) | `str.translate` in Python; `String#tr` in Ruby; `strtr` in PHP; `tr///` in Perl | [#21](https://github.com/huandu/xstrings/issues/21) |
+| [Width](https://godoc.org/github.com/huandu/xstrings#Width) | `mb_strwidth` in PHP | [#26](https://github.com/huandu/xstrings/issues/26) |
+| [WordCount](https://godoc.org/github.com/huandu/xstrings#WordCount) | `str_word_count` in PHP | [#14](https://github.com/huandu/xstrings/issues/14) |
+| [WordSplit](https://godoc.org/github.com/huandu/xstrings#WordSplit) | - | [#14](https://github.com/huandu/xstrings/issues/14) |
+
+### Package `strings` functions ###
+
+*Keep this table sorted by Function in ascending order.*
+
+| Function | Friends |
+| -------- | ------- |
+| [Contains](http://golang.org/pkg/strings/#Contains) | `String#include?` in Ruby |
+| [ContainsAny](http://golang.org/pkg/strings/#ContainsAny) | - |
+| [ContainsRune](http://golang.org/pkg/strings/#ContainsRune) | - |
+| [Count](http://golang.org/pkg/strings/#Count) | `str.count` in Python; `substr_count` in PHP |
+| [EqualFold](http://golang.org/pkg/strings/#EqualFold) | `stricmp` in PHP; `String#casecmp` in Ruby |
+| [Fields](http://golang.org/pkg/strings/#Fields) | `str.split` in Python; `split` in Perl; `String#split` in Ruby |
+| [FieldsFunc](http://golang.org/pkg/strings/#FieldsFunc) | - |
+| [HasPrefix](http://golang.org/pkg/strings/#HasPrefix) | `str.startswith` in Python; `String#start_with?` in Ruby |
+| [HasSuffix](http://golang.org/pkg/strings/#HasSuffix) | `str.endswith` in Python; `String#end_with?` in Ruby |
+| [Index](http://golang.org/pkg/strings/#Index) | `str.index` in Python; `String#index` in Ruby; `strpos` in PHP; `index` in Perl |
+| [IndexAny](http://golang.org/pkg/strings/#IndexAny) | - |
+| [IndexByte](http://golang.org/pkg/strings/#IndexByte) | - |
+| [IndexFunc](http://golang.org/pkg/strings/#IndexFunc) | - |
+| [IndexRune](http://golang.org/pkg/strings/#IndexRune) | - |
+| [Join](http://golang.org/pkg/strings/#Join) | `str.join` in Python; `Array#join` in Ruby; `implode` in PHP; `join` in Perl |
+| [LastIndex](http://golang.org/pkg/strings/#LastIndex) | `str.rindex` in Python; `String#rindex`; `strrpos` in PHP; `rindex` in Perl |
+| [LastIndexAny](http://golang.org/pkg/strings/#LastIndexAny) | - |
+| [LastIndexFunc](http://golang.org/pkg/strings/#LastIndexFunc) | - |
+| [Map](http://golang.org/pkg/strings/#Map) | `String#each_codepoint` in Ruby |
+| [Repeat](http://golang.org/pkg/strings/#Repeat) | operator `*` in Python and Ruby; `str_repeat` in PHP |
+| [Replace](http://golang.org/pkg/strings/#Replace) | `str.replace` in Python; `String#sub` in Ruby; `str_replace` in PHP |
+| [Split](http://golang.org/pkg/strings/#Split) | `str.split` in Python; `String#split` in Ruby; `explode` in PHP; `split` in Perl |
+| [SplitAfter](http://golang.org/pkg/strings/#SplitAfter) | - |
+| [SplitAfterN](http://golang.org/pkg/strings/#SplitAfterN) | - |
+| [SplitN](http://golang.org/pkg/strings/#SplitN) | `str.split` in Python; `String#split` in Ruby; `explode` in PHP; `split` in Perl |
+| [Title](http://golang.org/pkg/strings/#Title) | `str.title` in Python |
+| [ToLower](http://golang.org/pkg/strings/#ToLower) | `str.lower` in Python; `String#downcase` in Ruby; `strtolower` in PHP; `lc` in Perl |
+| [ToLowerSpecial](http://golang.org/pkg/strings/#ToLowerSpecial) | - |
+| [ToTitle](http://golang.org/pkg/strings/#ToTitle) | - |
+| [ToTitleSpecial](http://golang.org/pkg/strings/#ToTitleSpecial) | - |
+| [ToUpper](http://golang.org/pkg/strings/#ToUpper) | `str.upper` in Python; `String#upcase` in Ruby; `strtoupper` in PHP; `uc` in Perl |
+| [ToUpperSpecial](http://golang.org/pkg/strings/#ToUpperSpecial) | - |
+| [Trim](http://golang.org/pkg/strings/#Trim) | `str.strip` in Python; `String#strip` in Ruby; `trim` in PHP |
+| [TrimFunc](http://golang.org/pkg/strings/#TrimFunc) | - |
+| [TrimLeft](http://golang.org/pkg/strings/#TrimLeft) | `str.lstrip` in Python; `String#lstrip` in Ruby; `ltrim` in PHP |
+| [TrimLeftFunc](http://golang.org/pkg/strings/#TrimLeftFunc) | - |
+| [TrimPrefix](http://golang.org/pkg/strings/#TrimPrefix) | - |
+| [TrimRight](http://golang.org/pkg/strings/#TrimRight) | `str.rstrip` in Python; `String#rstrip` in Ruby; `rtrim` in PHP |
+| [TrimRightFunc](http://golang.org/pkg/strings/#TrimRightFunc) | - |
+| [TrimSpace](http://golang.org/pkg/strings/#TrimSpace) | `str.strip` in Python; `String#strip` in Ruby; `trim` in PHP |
+| [TrimSuffix](http://golang.org/pkg/strings/#TrimSuffix) | `String#chomp` in Ruby; `chomp` in Perl |
+
+## License ##
+
+This library is licensed under MIT license. See LICENSE for details.
diff --git a/backend/vendor/github.com/huandu/xstrings/common.go b/backend/vendor/github.com/huandu/xstrings/common.go
new file mode 100644
index 00000000..2aff57aa
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/common.go
@@ -0,0 +1,25 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"bytes"
+)
+
+const bufferMaxInitGrowSize = 2048
+
+// Lazy initialize a buffer.
+func allocBuffer(orig, cur string) *bytes.Buffer {
+	output := &bytes.Buffer{}
+	maxSize := len(orig) * 4
+
+	// Avoid to reserve too much memory at once.
+	if maxSize > bufferMaxInitGrowSize {
+		maxSize = bufferMaxInitGrowSize
+	}
+
+	output.Grow(maxSize)
+	output.WriteString(orig[:len(orig)-len(cur)])
+	return output
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/convert.go b/backend/vendor/github.com/huandu/xstrings/convert.go
new file mode 100644
index 00000000..8253fa9c
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/convert.go
@@ -0,0 +1,400 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"bytes"
+	"math/rand"
+	"unicode"
+	"unicode/utf8"
+)
+
+// ToCamelCase can convert all lower case characters behind underscores
+// to upper case character.
+// Underscore character will be removed in result except following cases.
+//     * More than 1 underscore.
+//           "a__b" => "A_B"
+//     * At the beginning of string.
+//           "_a" => "_A"
+//     * At the end of string.
+//           "ab_" => "Ab_"
+func ToCamelCase(str string) string {
+	if len(str) == 0 {
+		return ""
+	}
+
+	buf := &bytes.Buffer{}
+	var r0, r1 rune
+	var size int
+
+	// leading '_' will appear in output.
+	for len(str) > 0 {
+		r0, size = utf8.DecodeRuneInString(str)
+		str = str[size:]
+
+		if r0 != '_' {
+			break
+		}
+
+		buf.WriteRune(r0)
+	}
+
+	if len(str) == 0 {
+		return buf.String()
+	}
+
+	r0 = unicode.ToUpper(r0)
+
+	for len(str) > 0 {
+		r1 = r0
+		r0, size = utf8.DecodeRuneInString(str)
+		str = str[size:]
+
+		if r1 == '_' && r0 == '_' {
+			buf.WriteRune(r1)
+			continue
+		}
+
+		if r1 == '_' {
+			r0 = unicode.ToUpper(r0)
+		} else {
+			r0 = unicode.ToLower(r0)
+		}
+
+		if r1 != '_' {
+			buf.WriteRune(r1)
+		}
+	}
+
+	buf.WriteRune(r0)
+	return buf.String()
+}
+
+// ToSnakeCase can convert all upper case characters in a string to
+// snake case format.
+//
+// Some samples.
+//     "FirstName"  => "first_name"
+//     "HTTPServer" => "http_server"
+//     "NoHTTPS"    => "no_https"
+//     "GO_PATH"    => "go_path"
+//     "GO PATH"    => "go_path"      // space is converted to underscore.
+//     "GO-PATH"    => "go_path"      // hyphen is converted to underscore.
+//     "HTTP2XX"    => "http_2xx"     // insert an underscore before a number and after an alphabet.
+//     "http2xx"    => "http_2xx"
+//     "HTTP20xOK"  => "http_20x_ok"
+func ToSnakeCase(str string) string {
+	return camelCaseToLowerCase(str, '_')
+}
+
+// ToKebabCase can convert all upper case characters in a string to
+// kebab case format.
+//
+// Some samples.
+//     "FirstName"  => "first-name"
+//     "HTTPServer" => "http-server"
+//     "NoHTTPS"    => "no-https"
+//     "GO_PATH"    => "go-path"
+//     "GO PATH"    => "go-path"      // space is converted to '-'.
+//     "GO-PATH"    => "go-path"      // hyphen is converted to '-'.
+//     "HTTP2XX"    => "http-2xx"     // insert a '-' before a number and after an alphabet.
+//     "http2xx"    => "http-2xx"
+//     "HTTP20xOK"  => "http-20x-ok"
+func ToKebabCase(str string) string {
+	return camelCaseToLowerCase(str, '-')
+}
+
+func camelCaseToLowerCase(str string, connector rune) string {
+	if len(str) == 0 {
+		return ""
+	}
+
+	buf := &bytes.Buffer{}
+	var prev, r0, r1 rune
+	var size int
+
+	r0 = connector
+
+	for len(str) > 0 {
+		prev = r0
+		r0, size = utf8.DecodeRuneInString(str)
+		str = str[size:]
+
+		switch {
+		case r0 == utf8.RuneError:
+			buf.WriteRune(r0)
+
+		case unicode.IsUpper(r0):
+			if prev != connector && !unicode.IsNumber(prev) {
+				buf.WriteRune(connector)
+			}
+
+			buf.WriteRune(unicode.ToLower(r0))
+
+			if len(str) == 0 {
+				break
+			}
+
+			r0, size = utf8.DecodeRuneInString(str)
+			str = str[size:]
+
+			if !unicode.IsUpper(r0) {
+				buf.WriteRune(r0)
+				break
+			}
+
+			// find next non-upper-case character and insert connector properly.
+			// it's designed to convert `HTTPServer` to `http_server`.
+			// if there are more than 2 adjacent upper case characters in a word,
+			// treat them as an abbreviation plus a normal word.
+			for len(str) > 0 {
+				r1 = r0
+				r0, size = utf8.DecodeRuneInString(str)
+				str = str[size:]
+
+				if r0 == utf8.RuneError {
+					buf.WriteRune(unicode.ToLower(r1))
+					buf.WriteRune(r0)
+					break
+				}
+
+				if !unicode.IsUpper(r0) {
+					if r0 == '_' || r0 == ' ' || r0 == '-' {
+						r0 = connector
+
+						buf.WriteRune(unicode.ToLower(r1))
+					} else if unicode.IsNumber(r0) {
+						// treat a number as an upper case rune
+						// so that both `http2xx` and `HTTP2XX` can be converted to `http_2xx`.
+						buf.WriteRune(unicode.ToLower(r1))
+						buf.WriteRune(connector)
+						buf.WriteRune(r0)
+					} else {
+						buf.WriteRune(connector)
+						buf.WriteRune(unicode.ToLower(r1))
+						buf.WriteRune(r0)
+					}
+
+					break
+				}
+
+				buf.WriteRune(unicode.ToLower(r1))
+			}
+
+			if len(str) == 0 || r0 == connector {
+				buf.WriteRune(unicode.ToLower(r0))
+			}
+
+		case unicode.IsNumber(r0):
+			if prev != connector && !unicode.IsNumber(prev) {
+				buf.WriteRune(connector)
+			}
+
+			buf.WriteRune(r0)
+
+		default:
+			if r0 == ' ' || r0 == '-' || r0 == '_' {
+				r0 = connector
+			}
+
+			buf.WriteRune(r0)
+		}
+	}
+
+	return buf.String()
+}
+
+// SwapCase will swap characters case from upper to lower or lower to upper.
+func SwapCase(str string) string {
+	var r rune
+	var size int
+
+	buf := &bytes.Buffer{}
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		switch {
+		case unicode.IsUpper(r):
+			buf.WriteRune(unicode.ToLower(r))
+
+		case unicode.IsLower(r):
+			buf.WriteRune(unicode.ToUpper(r))
+
+		default:
+			buf.WriteRune(r)
+		}
+
+		str = str[size:]
+	}
+
+	return buf.String()
+}
+
+// FirstRuneToUpper converts first rune to upper case if necessary.
+func FirstRuneToUpper(str string) string {
+	if str == "" {
+		return str
+	}
+
+	r, size := utf8.DecodeRuneInString(str)
+
+	if !unicode.IsLower(r) {
+		return str
+	}
+
+	buf := &bytes.Buffer{}
+	buf.WriteRune(unicode.ToUpper(r))
+	buf.WriteString(str[size:])
+	return buf.String()
+}
+
+// FirstRuneToLower converts first rune to lower case if necessary.
+func FirstRuneToLower(str string) string {
+	if str == "" {
+		return str
+	}
+
+	r, size := utf8.DecodeRuneInString(str)
+
+	if !unicode.IsUpper(r) {
+		return str
+	}
+
+	buf := &bytes.Buffer{}
+	buf.WriteRune(unicode.ToLower(r))
+	buf.WriteString(str[size:])
+	return buf.String()
+}
+
+// Shuffle randomizes runes in a string and returns the result.
+// It uses default random source in `math/rand`.
+func Shuffle(str string) string {
+	if str == "" {
+		return str
+	}
+
+	runes := []rune(str)
+	index := 0
+
+	for i := len(runes) - 1; i > 0; i-- {
+		index = rand.Intn(i + 1)
+
+		if i != index {
+			runes[i], runes[index] = runes[index], runes[i]
+		}
+	}
+
+	return string(runes)
+}
+
+// ShuffleSource randomizes runes in a string with given random source.
+func ShuffleSource(str string, src rand.Source) string {
+	if str == "" {
+		return str
+	}
+
+	runes := []rune(str)
+	index := 0
+	r := rand.New(src)
+
+	for i := len(runes) - 1; i > 0; i-- {
+		index = r.Intn(i + 1)
+
+		if i != index {
+			runes[i], runes[index] = runes[index], runes[i]
+		}
+	}
+
+	return string(runes)
+}
+
+// Successor returns the successor to string.
+//
+// If there is one alphanumeric rune is found in string, increase the rune by 1.
+// If increment generates a "carry", the rune to the left of it is incremented.
+// This process repeats until there is no carry, adding an additional rune if necessary.
+//
+// If there is no alphanumeric rune, the rightmost rune will be increased by 1
+// regardless whether the result is a valid rune or not.
+//
+// Only following characters are alphanumeric.
+//     * a - z
+//     * A - Z
+//     * 0 - 9
+//
+// Samples (borrowed from ruby's String#succ document):
+//     "abcd"      => "abce"
+//     "THX1138"   => "THX1139"
+//     "<<koala>>" => "<<koalb>>"
+//     "1999zzz"   => "2000aaa"
+//     "ZZZ9999"   => "AAAA0000"
+//     "***"       => "**+"
+func Successor(str string) string {
+	if str == "" {
+		return str
+	}
+
+	var r rune
+	var i int
+	carry := ' '
+	runes := []rune(str)
+	l := len(runes)
+	lastAlphanumeric := l
+
+	for i = l - 1; i >= 0; i-- {
+		r = runes[i]
+
+		if ('a' <= r && r <= 'y') ||
+			('A' <= r && r <= 'Y') ||
+			('0' <= r && r <= '8') {
+			runes[i]++
+			carry = ' '
+			lastAlphanumeric = i
+			break
+		}
+
+		switch r {
+		case 'z':
+			runes[i] = 'a'
+			carry = 'a'
+			lastAlphanumeric = i
+
+		case 'Z':
+			runes[i] = 'A'
+			carry = 'A'
+			lastAlphanumeric = i
+
+		case '9':
+			runes[i] = '0'
+			carry = '0'
+			lastAlphanumeric = i
+		}
+	}
+
+	// Needs to add one character for carry.
+	if i < 0 && carry != ' ' {
+		buf := &bytes.Buffer{}
+		buf.Grow(l + 4) // Reserve enough space for write.
+
+		if lastAlphanumeric != 0 {
+			buf.WriteString(str[:lastAlphanumeric])
+		}
+
+		buf.WriteRune(carry)
+
+		for _, r = range runes[lastAlphanumeric:] {
+			buf.WriteRune(r)
+		}
+
+		return buf.String()
+	}
+
+	// No alphanumeric character. Simply increase last rune's value.
+	if lastAlphanumeric == l {
+		runes[l-1]++
+	}
+
+	return string(runes)
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/count.go b/backend/vendor/github.com/huandu/xstrings/count.go
new file mode 100644
index 00000000..f96e3870
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/count.go
@@ -0,0 +1,120 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"unicode"
+	"unicode/utf8"
+)
+
+// Len returns str's utf8 rune length.
+func Len(str string) int {
+	return utf8.RuneCountInString(str)
+}
+
+// WordCount returns number of words in a string.
+//
+// Word is defined as a locale dependent string containing alphabetic characters,
+// which may also contain but not start with `'` and `-` characters.
+func WordCount(str string) int {
+	var r rune
+	var size, n int
+
+	inWord := false
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		switch {
+		case isAlphabet(r):
+			if !inWord {
+				inWord = true
+				n++
+			}
+
+		case inWord && (r == '\'' || r == '-'):
+			// Still in word.
+
+		default:
+			inWord = false
+		}
+
+		str = str[size:]
+	}
+
+	return n
+}
+
+const minCJKCharacter = '\u3400'
+
+// Checks r is a letter but not CJK character.
+func isAlphabet(r rune) bool {
+	if !unicode.IsLetter(r) {
+		return false
+	}
+
+	switch {
+	// Quick check for non-CJK character.
+	case r < minCJKCharacter:
+		return true
+
+	// Common CJK characters.
+	case r >= '\u4E00' && r <= '\u9FCC':
+		return false
+
+	// Rare CJK characters.
+	case r >= '\u3400' && r <= '\u4D85':
+		return false
+
+	// Rare and historic CJK characters.
+	case r >= '\U00020000' && r <= '\U0002B81D':
+		return false
+	}
+
+	return true
+}
+
+// Width returns string width in monotype font.
+// Multi-byte characters are usually twice the width of single byte characters.
+//
+// Algorithm comes from `mb_strwidth` in PHP.
+// http://php.net/manual/en/function.mb-strwidth.php
+func Width(str string) int {
+	var r rune
+	var size, n int
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+		n += RuneWidth(r)
+		str = str[size:]
+	}
+
+	return n
+}
+
+// RuneWidth returns character width in monotype font.
+// Multi-byte characters are usually twice the width of single byte characters.
+//
+// Algorithm comes from `mb_strwidth` in PHP.
+// http://php.net/manual/en/function.mb-strwidth.php
+func RuneWidth(r rune) int {
+	switch {
+	case r == utf8.RuneError || r < '\x20':
+		return 0
+
+	case '\x20' <= r && r < '\u2000':
+		return 1
+
+	case '\u2000' <= r && r < '\uFF61':
+		return 2
+
+	case '\uFF61' <= r && r < '\uFFA0':
+		return 1
+
+	case '\uFFA0' <= r:
+		return 2
+	}
+
+	return 0
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/doc.go b/backend/vendor/github.com/huandu/xstrings/doc.go
new file mode 100644
index 00000000..1a6ef069
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/doc.go
@@ -0,0 +1,8 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+// Package xstrings is to provide string algorithms which are useful but not included in `strings` package.
+// See project home page for details. https://github.com/huandu/xstrings
+//
+// Package xstrings assumes all strings are encoded in utf8.
+package xstrings
diff --git a/backend/vendor/github.com/huandu/xstrings/format.go b/backend/vendor/github.com/huandu/xstrings/format.go
new file mode 100644
index 00000000..2d02df1c
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/format.go
@@ -0,0 +1,170 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"bytes"
+	"unicode/utf8"
+)
+
+// ExpandTabs can expand tabs ('\t') rune in str to one or more spaces dpending on
+// current column and tabSize.
+// The column number is reset to zero after each newline ('\n') occurring in the str.
+//
+// ExpandTabs uses RuneWidth to decide rune's width.
+// For example, CJK characters will be treated as two characters.
+//
+// If tabSize <= 0, ExpandTabs panics with error.
+//
+// Samples:
+//     ExpandTabs("a\tbc\tdef\tghij\tk", 4) => "a   bc  def ghij    k"
+//     ExpandTabs("abcdefg\thij\nk\tl", 4)  => "abcdefg hij\nk   l"
+//     ExpandTabs("z中\t文\tw", 4)           => "z中 文  w"
+func ExpandTabs(str string, tabSize int) string {
+	if tabSize <= 0 {
+		panic("tab size must be positive")
+	}
+
+	var r rune
+	var i, size, column, expand int
+	var output *bytes.Buffer
+
+	orig := str
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		if r == '\t' {
+			expand = tabSize - column%tabSize
+
+			if output == nil {
+				output = allocBuffer(orig, str)
+			}
+
+			for i = 0; i < expand; i++ {
+				output.WriteByte(byte(' '))
+			}
+
+			column += expand
+		} else {
+			if r == '\n' {
+				column = 0
+			} else {
+				column += RuneWidth(r)
+			}
+
+			if output != nil {
+				output.WriteRune(r)
+			}
+		}
+
+		str = str[size:]
+	}
+
+	if output == nil {
+		return orig
+	}
+
+	return output.String()
+}
+
+// LeftJustify returns a string with pad string at right side if str's rune length is smaller than length.
+// If str's rune length is larger than length, str itself will be returned.
+//
+// If pad is an empty string, str will be returned.
+//
+// Samples:
+//     LeftJustify("hello", 4, " ")    => "hello"
+//     LeftJustify("hello", 10, " ")   => "hello     "
+//     LeftJustify("hello", 10, "123") => "hello12312"
+func LeftJustify(str string, length int, pad string) string {
+	l := Len(str)
+
+	if l >= length || pad == "" {
+		return str
+	}
+
+	remains := length - l
+	padLen := Len(pad)
+
+	output := &bytes.Buffer{}
+	output.Grow(len(str) + (remains/padLen+1)*len(pad))
+	output.WriteString(str)
+	writePadString(output, pad, padLen, remains)
+	return output.String()
+}
+
+// RightJustify returns a string with pad string at left side if str's rune length is smaller than length.
+// If str's rune length is larger than length, str itself will be returned.
+//
+// If pad is an empty string, str will be returned.
+//
+// Samples:
+//     RightJustify("hello", 4, " ")    => "hello"
+//     RightJustify("hello", 10, " ")   => "     hello"
+//     RightJustify("hello", 10, "123") => "12312hello"
+func RightJustify(str string, length int, pad string) string {
+	l := Len(str)
+
+	if l >= length || pad == "" {
+		return str
+	}
+
+	remains := length - l
+	padLen := Len(pad)
+
+	output := &bytes.Buffer{}
+	output.Grow(len(str) + (remains/padLen+1)*len(pad))
+	writePadString(output, pad, padLen, remains)
+	output.WriteString(str)
+	return output.String()
+}
+
+// Center returns a string with pad string at both side if str's rune length is smaller than length.
+// If str's rune length is larger than length, str itself will be returned.
+//
+// If pad is an empty string, str will be returned.
+//
+// Samples:
+//     Center("hello", 4, " ")    => "hello"
+//     Center("hello", 10, " ")   => "  hello   "
+//     Center("hello", 10, "123") => "12hello123"
+func Center(str string, length int, pad string) string {
+	l := Len(str)
+
+	if l >= length || pad == "" {
+		return str
+	}
+
+	remains := length - l
+	padLen := Len(pad)
+
+	output := &bytes.Buffer{}
+	output.Grow(len(str) + (remains/padLen+1)*len(pad))
+	writePadString(output, pad, padLen, remains/2)
+	output.WriteString(str)
+	writePadString(output, pad, padLen, (remains+1)/2)
+	return output.String()
+}
+
+func writePadString(output *bytes.Buffer, pad string, padLen, remains int) {
+	var r rune
+	var size int
+
+	repeats := remains / padLen
+
+	for i := 0; i < repeats; i++ {
+		output.WriteString(pad)
+	}
+
+	remains = remains % padLen
+
+	if remains != 0 {
+		for i := 0; i < remains; i++ {
+			r, size = utf8.DecodeRuneInString(pad)
+			output.WriteRune(r)
+			pad = pad[size:]
+		}
+	}
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/go.mod b/backend/vendor/github.com/huandu/xstrings/go.mod
new file mode 100644
index 00000000..5866b3a8
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/go.mod
@@ -0,0 +1 @@
+module github.com/huandu/xstrings
\ No newline at end of file
diff --git a/backend/vendor/github.com/huandu/xstrings/manipulate.go b/backend/vendor/github.com/huandu/xstrings/manipulate.go
new file mode 100644
index 00000000..0eefb43e
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/manipulate.go
@@ -0,0 +1,217 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"bytes"
+	"strings"
+	"unicode/utf8"
+)
+
+// Reverse a utf8 encoded string.
+func Reverse(str string) string {
+	var size int
+
+	tail := len(str)
+	buf := make([]byte, tail)
+	s := buf
+
+	for len(str) > 0 {
+		_, size = utf8.DecodeRuneInString(str)
+		tail -= size
+		s = append(s[:tail], []byte(str[:size])...)
+		str = str[size:]
+	}
+
+	return string(buf)
+}
+
+// Slice a string by rune.
+//
+// Start must satisfy 0 <= start <= rune length.
+//
+// End can be positive, zero or negative.
+// If end >= 0, start and end must satisfy start <= end <= rune length.
+// If end < 0, it means slice to the end of string.
+//
+// Otherwise, Slice will panic as out of range.
+func Slice(str string, start, end int) string {
+	var size, startPos, endPos int
+
+	origin := str
+
+	if start < 0 || end > len(str) || (end >= 0 && start > end) {
+		panic("out of range")
+	}
+
+	if end >= 0 {
+		end -= start
+	}
+
+	for start > 0 && len(str) > 0 {
+		_, size = utf8.DecodeRuneInString(str)
+		start--
+		startPos += size
+		str = str[size:]
+	}
+
+	if end < 0 {
+		return origin[startPos:]
+	}
+
+	endPos = startPos
+
+	for end > 0 && len(str) > 0 {
+		_, size = utf8.DecodeRuneInString(str)
+		end--
+		endPos += size
+		str = str[size:]
+	}
+
+	if len(str) == 0 && (start > 0 || end > 0) {
+		panic("out of range")
+	}
+
+	return origin[startPos:endPos]
+}
+
+// Partition splits a string by sep into three parts.
+// The return value is a slice of strings with head, match and tail.
+//
+// If str contains sep, for example "hello" and "l", Partition returns
+//     "he", "l", "lo"
+//
+// If str doesn't contain sep, for example "hello" and "x", Partition returns
+//     "hello", "", ""
+func Partition(str, sep string) (head, match, tail string) {
+	index := strings.Index(str, sep)
+
+	if index == -1 {
+		head = str
+		return
+	}
+
+	head = str[:index]
+	match = str[index : index+len(sep)]
+	tail = str[index+len(sep):]
+	return
+}
+
+// LastPartition splits a string by last instance of sep into three parts.
+// The return value is a slice of strings with head, match and tail.
+//
+// If str contains sep, for example "hello" and "l", LastPartition returns
+//     "hel", "l", "o"
+//
+// If str doesn't contain sep, for example "hello" and "x", LastPartition returns
+//     "", "", "hello"
+func LastPartition(str, sep string) (head, match, tail string) {
+	index := strings.LastIndex(str, sep)
+
+	if index == -1 {
+		tail = str
+		return
+	}
+
+	head = str[:index]
+	match = str[index : index+len(sep)]
+	tail = str[index+len(sep):]
+	return
+}
+
+// Insert src into dst at given rune index.
+// Index is counted by runes instead of bytes.
+//
+// If index is out of range of dst, panic with out of range.
+func Insert(dst, src string, index int) string {
+	return Slice(dst, 0, index) + src + Slice(dst, index, -1)
+}
+
+// Scrub scrubs invalid utf8 bytes with repl string.
+// Adjacent invalid bytes are replaced only once.
+func Scrub(str, repl string) string {
+	var buf *bytes.Buffer
+	var r rune
+	var size, pos int
+	var hasError bool
+
+	origin := str
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		if r == utf8.RuneError {
+			if !hasError {
+				if buf == nil {
+					buf = &bytes.Buffer{}
+				}
+
+				buf.WriteString(origin[:pos])
+				hasError = true
+			}
+		} else if hasError {
+			hasError = false
+			buf.WriteString(repl)
+
+			origin = origin[pos:]
+			pos = 0
+		}
+
+		pos += size
+		str = str[size:]
+	}
+
+	if buf != nil {
+		buf.WriteString(origin)
+		return buf.String()
+	}
+
+	// No invalid byte.
+	return origin
+}
+
+// WordSplit splits a string into words. Returns a slice of words.
+// If there is no word in a string, return nil.
+//
+// Word is defined as a locale dependent string containing alphabetic characters,
+// which may also contain but not start with `'` and `-` characters.
+func WordSplit(str string) []string {
+	var word string
+	var words []string
+	var r rune
+	var size, pos int
+
+	inWord := false
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		switch {
+		case isAlphabet(r):
+			if !inWord {
+				inWord = true
+				word = str
+				pos = 0
+			}
+
+		case inWord && (r == '\'' || r == '-'):
+			// Still in word.
+
+		default:
+			if inWord {
+				inWord = false
+				words = append(words, word[:pos])
+			}
+		}
+
+		pos += size
+		str = str[size:]
+	}
+
+	if inWord {
+		words = append(words, word[:pos])
+	}
+
+	return words
+}
diff --git a/backend/vendor/github.com/huandu/xstrings/translate.go b/backend/vendor/github.com/huandu/xstrings/translate.go
new file mode 100644
index 00000000..d86a4cbb
--- /dev/null
+++ b/backend/vendor/github.com/huandu/xstrings/translate.go
@@ -0,0 +1,547 @@
+// Copyright 2015 Huan Du. All rights reserved.
+// Licensed under the MIT license that can be found in the LICENSE file.
+
+package xstrings
+
+import (
+	"bytes"
+	"unicode"
+	"unicode/utf8"
+)
+
+type runeRangeMap struct {
+	FromLo rune // Lower bound of range map.
+	FromHi rune // An inclusive higher bound of range map.
+	ToLo   rune
+	ToHi   rune
+}
+
+type runeDict struct {
+	Dict [unicode.MaxASCII + 1]rune
+}
+
+type runeMap map[rune]rune
+
+// Translator can translate string with pre-compiled from and to patterns.
+// If a from/to pattern pair needs to be used more than once, it's recommended
+// to create a Translator and reuse it.
+type Translator struct {
+	quickDict  *runeDict       // A quick dictionary to look up rune by index. Only availabe for latin runes.
+	runeMap    runeMap         // Rune map for translation.
+	ranges     []*runeRangeMap // Ranges of runes.
+	mappedRune rune            // If mappedRune >= 0, all matched runes are translated to the mappedRune.
+	reverted   bool            // If to pattern is empty, all matched characters will be deleted.
+	hasPattern bool
+}
+
+// NewTranslator creates new Translator through a from/to pattern pair.
+func NewTranslator(from, to string) *Translator {
+	tr := &Translator{}
+
+	if from == "" {
+		return tr
+	}
+
+	reverted := from[0] == '^'
+	deletion := len(to) == 0
+
+	if reverted {
+		from = from[1:]
+	}
+
+	var fromStart, fromEnd, fromRangeStep rune
+	var toStart, toEnd, toRangeStep rune
+	var fromRangeSize, toRangeSize rune
+	var singleRunes []rune
+
+	// Update the to rune range.
+	updateRange := func() {
+		// No more rune to read in the to rune pattern.
+		if toEnd == utf8.RuneError {
+			return
+		}
+
+		if toRangeStep == 0 {
+			to, toStart, toEnd, toRangeStep = nextRuneRange(to, toEnd)
+			return
+		}
+
+		// Current range is not empty. Consume 1 rune from start.
+		if toStart != toEnd {
+			toStart += toRangeStep
+			return
+		}
+
+		// No more rune. Repeat the last rune.
+		if to == "" {
+			toEnd = utf8.RuneError
+			return
+		}
+
+		// Both start and end are used. Read two more runes from the to pattern.
+		to, toStart, toEnd, toRangeStep = nextRuneRange(to, utf8.RuneError)
+	}
+
+	if deletion {
+		toStart = utf8.RuneError
+		toEnd = utf8.RuneError
+	} else {
+		// If from pattern is reverted, only the last rune in the to pattern will be used.
+		if reverted {
+			var size int
+
+			for len(to) > 0 {
+				toStart, size = utf8.DecodeRuneInString(to)
+				to = to[size:]
+			}
+
+			toEnd = utf8.RuneError
+		} else {
+			to, toStart, toEnd, toRangeStep = nextRuneRange(to, utf8.RuneError)
+		}
+	}
+
+	fromEnd = utf8.RuneError
+
+	for len(from) > 0 {
+		from, fromStart, fromEnd, fromRangeStep = nextRuneRange(from, fromEnd)
+
+		// fromStart is a single character. Just map it with a rune in the to pattern.
+		if fromRangeStep == 0 {
+			singleRunes = tr.addRune(fromStart, toStart, singleRunes)
+			updateRange()
+			continue
+		}
+
+		for toEnd != utf8.RuneError && fromStart != fromEnd {
+			// If mapped rune is a single character instead of a range, simply shift first
+			// rune in the range.
+			if toRangeStep == 0 {
+				singleRunes = tr.addRune(fromStart, toStart, singleRunes)
+				updateRange()
+				fromStart += fromRangeStep
+				continue
+			}
+
+			fromRangeSize = (fromEnd - fromStart) * fromRangeStep
+			toRangeSize = (toEnd - toStart) * toRangeStep
+
+			// Not enough runes in the to pattern. Need to read more.
+			if fromRangeSize > toRangeSize {
+				fromStart, toStart = tr.addRuneRange(fromStart, fromStart+toRangeSize*fromRangeStep, toStart, toEnd, singleRunes)
+				fromStart += fromRangeStep
+				updateRange()
+
+				// Edge case: If fromRangeSize == toRangeSize + 1, the last fromStart value needs be considered
+				// as a single rune.
+				if fromStart == fromEnd {
+					singleRunes = tr.addRune(fromStart, toStart, singleRunes)
+					updateRange()
+				}
+
+				continue
+			}
+
+			fromStart, toStart = tr.addRuneRange(fromStart, fromEnd, toStart, toStart+fromRangeSize*toRangeStep, singleRunes)
+			updateRange()
+			break
+		}
+
+		if fromStart == fromEnd {
+			fromEnd = utf8.RuneError
+			continue
+		}
+
+		fromStart, toStart = tr.addRuneRange(fromStart, fromEnd, toStart, toStart, singleRunes)
+		fromEnd = utf8.RuneError
+	}
+
+	if fromEnd != utf8.RuneError {
+		singleRunes = tr.addRune(fromEnd, toStart, singleRunes)
+	}
+
+	tr.reverted = reverted
+	tr.mappedRune = -1
+	tr.hasPattern = true
+
+	// Translate RuneError only if in deletion or reverted mode.
+	if deletion || reverted {
+		tr.mappedRune = toStart
+	}
+
+	return tr
+}
+
+func (tr *Translator) addRune(from, to rune, singleRunes []rune) []rune {
+	if from <= unicode.MaxASCII {
+		if tr.quickDict == nil {
+			tr.quickDict = &runeDict{}
+		}
+
+		tr.quickDict.Dict[from] = to
+	} else {
+		if tr.runeMap == nil {
+			tr.runeMap = make(runeMap)
+		}
+
+		tr.runeMap[from] = to
+	}
+
+	singleRunes = append(singleRunes, from)
+	return singleRunes
+}
+
+func (tr *Translator) addRuneRange(fromLo, fromHi, toLo, toHi rune, singleRunes []rune) (rune, rune) {
+	var r rune
+	var rrm *runeRangeMap
+
+	if fromLo < fromHi {
+		rrm = &runeRangeMap{
+			FromLo: fromLo,
+			FromHi: fromHi,
+			ToLo:   toLo,
+			ToHi:   toHi,
+		}
+	} else {
+		rrm = &runeRangeMap{
+			FromLo: fromHi,
+			FromHi: fromLo,
+			ToLo:   toHi,
+			ToHi:   toLo,
+		}
+	}
+
+	// If there is any single rune conflicts with this rune range, clear single rune record.
+	for _, r = range singleRunes {
+		if rrm.FromLo <= r && r <= rrm.FromHi {
+			if r <= unicode.MaxASCII {
+				tr.quickDict.Dict[r] = 0
+			} else {
+				delete(tr.runeMap, r)
+			}
+		}
+	}
+
+	tr.ranges = append(tr.ranges, rrm)
+	return fromHi, toHi
+}
+
+func nextRuneRange(str string, last rune) (remaining string, start, end rune, rangeStep rune) {
+	var r rune
+	var size int
+
+	remaining = str
+	escaping := false
+	isRange := false
+
+	for len(remaining) > 0 {
+		r, size = utf8.DecodeRuneInString(remaining)
+		remaining = remaining[size:]
+
+		// Parse special characters.
+		if !escaping {
+			if r == '\\' {
+				escaping = true
+				continue
+			}
+
+			if r == '-' {
+				// Ignore slash at beginning of string.
+				if last == utf8.RuneError {
+					continue
+				}
+
+				start = last
+				isRange = true
+				continue
+			}
+		}
+
+		escaping = false
+
+		if last != utf8.RuneError {
+			// This is a range which start and end are the same.
+			// Considier it as a normal character.
+			if isRange && last == r {
+				isRange = false
+				continue
+			}
+
+			start = last
+			end = r
+
+			if isRange {
+				if start < end {
+					rangeStep = 1
+				} else {
+					rangeStep = -1
+				}
+			}
+
+			return
+		}
+
+		last = r
+	}
+
+	start = last
+	end = utf8.RuneError
+	return
+}
+
+// Translate str with a from/to pattern pair.
+//
+// See comment in Translate function for usage and samples.
+func (tr *Translator) Translate(str string) string {
+	if !tr.hasPattern || str == "" {
+		return str
+	}
+
+	var r rune
+	var size int
+	var needTr bool
+
+	orig := str
+
+	var output *bytes.Buffer
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+		r, needTr = tr.TranslateRune(r)
+
+		if needTr && output == nil {
+			output = allocBuffer(orig, str)
+		}
+
+		if r != utf8.RuneError && output != nil {
+			output.WriteRune(r)
+		}
+
+		str = str[size:]
+	}
+
+	// No character is translated.
+	if output == nil {
+		return orig
+	}
+
+	return output.String()
+}
+
+// TranslateRune return translated rune and true if r matches the from pattern.
+// If r doesn't match the pattern, original r is returned and translated is false.
+func (tr *Translator) TranslateRune(r rune) (result rune, translated bool) {
+	switch {
+	case tr.quickDict != nil:
+		if r <= unicode.MaxASCII {
+			result = tr.quickDict.Dict[r]
+
+			if result != 0 {
+				translated = true
+
+				if tr.mappedRune >= 0 {
+					result = tr.mappedRune
+				}
+
+				break
+			}
+		}
+
+		fallthrough
+
+	case tr.runeMap != nil:
+		var ok bool
+
+		if result, ok = tr.runeMap[r]; ok {
+			translated = true
+
+			if tr.mappedRune >= 0 {
+				result = tr.mappedRune
+			}
+
+			break
+		}
+
+		fallthrough
+
+	default:
+		var rrm *runeRangeMap
+		ranges := tr.ranges
+
+		for i := len(ranges) - 1; i >= 0; i-- {
+			rrm = ranges[i]
+
+			if rrm.FromLo <= r && r <= rrm.FromHi {
+				translated = true
+
+				if tr.mappedRune >= 0 {
+					result = tr.mappedRune
+					break
+				}
+
+				if rrm.ToLo < rrm.ToHi {
+					result = rrm.ToLo + r - rrm.FromLo
+				} else if rrm.ToLo > rrm.ToHi {
+					// ToHi can be smaller than ToLo if range is from higher to lower.
+					result = rrm.ToLo - r + rrm.FromLo
+				} else {
+					result = rrm.ToLo
+				}
+
+				break
+			}
+		}
+	}
+
+	if tr.reverted {
+		if !translated {
+			result = tr.mappedRune
+		}
+
+		translated = !translated
+	}
+
+	if !translated {
+		result = r
+	}
+
+	return
+}
+
+// HasPattern returns true if Translator has one pattern at least.
+func (tr *Translator) HasPattern() bool {
+	return tr.hasPattern
+}
+
+// Translate str with the characters defined in from replaced by characters defined in to.
+//
+// From and to are patterns representing a set of characters. Pattern is defined as following.
+//
+//     * Special characters
+//       * '-' means a range of runes, e.g.
+//         * "a-z" means all characters from 'a' to 'z' inclusive;
+//         * "z-a" means all characters from 'z' to 'a' inclusive.
+//       * '^' as first character means a set of all runes excepted listed, e.g.
+//         * "^a-z" means all characters except 'a' to 'z' inclusive.
+//       * '\' escapes special characters.
+//     * Normal character represents itself, e.g. "abc" is a set including 'a', 'b' and 'c'.
+//
+// Translate will try to find a 1:1 mapping from from to to.
+// If to is smaller than from, last rune in to will be used to map "out of range" characters in from.
+//
+// Note that '^' only works in the from pattern. It will be considered as a normal character in the to pattern.
+//
+// If the to pattern is an empty string, Translate works exactly the same as Delete.
+//
+// Samples:
+//     Translate("hello", "aeiou", "12345")    => "h2ll4"
+//     Translate("hello", "a-z", "A-Z")        => "HELLO"
+//     Translate("hello", "z-a", "a-z")        => "svool"
+//     Translate("hello", "aeiou", "*")        => "h*ll*"
+//     Translate("hello", "^l", "*")           => "**ll*"
+//     Translate("hello ^ world", `\^lo`, "*") => "he*** * w*r*d"
+func Translate(str, from, to string) string {
+	tr := NewTranslator(from, to)
+	return tr.Translate(str)
+}
+
+// Delete runes in str matching the pattern.
+// Pattern is defined in Translate function.
+//
+// Samples:
+//     Delete("hello", "aeiou") => "hll"
+//     Delete("hello", "a-k")   => "llo"
+//     Delete("hello", "^a-k")  => "he"
+func Delete(str, pattern string) string {
+	tr := NewTranslator(pattern, "")
+	return tr.Translate(str)
+}
+
+// Count how many runes in str match the pattern.
+// Pattern is defined in Translate function.
+//
+// Samples:
+//     Count("hello", "aeiou") => 3
+//     Count("hello", "a-k")   => 3
+//     Count("hello", "^a-k")  => 2
+func Count(str, pattern string) int {
+	if pattern == "" || str == "" {
+		return 0
+	}
+
+	var r rune
+	var size int
+	var matched bool
+
+	tr := NewTranslator(pattern, "")
+	cnt := 0
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+		str = str[size:]
+
+		if _, matched = tr.TranslateRune(r); matched {
+			cnt++
+		}
+	}
+
+	return cnt
+}
+
+// Squeeze deletes adjacent repeated runes in str.
+// If pattern is not empty, only runes matching the pattern will be squeezed.
+//
+// Samples:
+//     Squeeze("hello", "")             => "helo"
+//     Squeeze("hello", "m-z")          => "hello"
+//     Squeeze("hello   world", " ")    => "hello world"
+func Squeeze(str, pattern string) string {
+	var last, r rune
+	var size int
+	var skipSqueeze, matched bool
+	var tr *Translator
+	var output *bytes.Buffer
+
+	orig := str
+	last = -1
+
+	if len(pattern) > 0 {
+		tr = NewTranslator(pattern, "")
+	}
+
+	for len(str) > 0 {
+		r, size = utf8.DecodeRuneInString(str)
+
+		// Need to squeeze the str.
+		if last == r && !skipSqueeze {
+			if tr != nil {
+				if _, matched = tr.TranslateRune(r); !matched {
+					skipSqueeze = true
+				}
+			}
+
+			if output == nil {
+				output = allocBuffer(orig, str)
+			}
+
+			if skipSqueeze {
+				output.WriteRune(r)
+			}
+		} else {
+			if output != nil {
+				output.WriteRune(r)
+			}
+
+			last = r
+			skipSqueeze = false
+		}
+
+		str = str[size:]
+	}
+
+	if output == nil {
+		return orig
+	}
+
+	return output.String()
+}
diff --git a/backend/vendor/github.com/imdario/mergo/.gitignore b/backend/vendor/github.com/imdario/mergo/.gitignore
new file mode 100644
index 00000000..529c3412
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/.gitignore
@@ -0,0 +1,33 @@
+#### joe made this: http://goel.io/joe
+
+#### go ####
+# Binaries for programs and plugins
+*.exe
+*.dll
+*.so
+*.dylib
+
+# Test binary, build with `go test -c`
+*.test
+
+# Output of the go coverage tool, specifically when used with LiteIDE
+*.out
+
+# Project-local glide cache, RE: https://github.com/Masterminds/glide/issues/736
+.glide/
+
+#### vim ####
+# Swap
+[._]*.s[a-v][a-z]
+[._]*.sw[a-p]
+[._]s[a-v][a-z]
+[._]sw[a-p]
+
+# Session
+Session.vim
+
+# Temporary
+.netrwhist
+*~
+# Auto-generated tag files
+tags
diff --git a/backend/vendor/github.com/imdario/mergo/.travis.yml b/backend/vendor/github.com/imdario/mergo/.travis.yml
new file mode 100644
index 00000000..b13a50ed
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/.travis.yml
@@ -0,0 +1,7 @@
+language: go
+install:
+  - go get -t
+  - go get golang.org/x/tools/cmd/cover
+  - go get github.com/mattn/goveralls
+script:
+  - $HOME/gopath/bin/goveralls -service=travis-ci -repotoken $COVERALLS_TOKEN
diff --git a/backend/vendor/github.com/imdario/mergo/CODE_OF_CONDUCT.md b/backend/vendor/github.com/imdario/mergo/CODE_OF_CONDUCT.md
new file mode 100644
index 00000000..469b4490
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/CODE_OF_CONDUCT.md
@@ -0,0 +1,46 @@
+# Contributor Covenant Code of Conduct
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, nationality, personal appearance, race, religion, or sexual identity and orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be reported by contacting the project team at i@dario.im. The project team will review and investigate all complaints, and will respond in a way that it deems appropriate to the circumstances. The project team is obligated to maintain confidentiality with regard to the reporter of an incident. Further details of specific enforcement policies may be posted separately.
+
+Project maintainers who do not follow or enforce the Code of Conduct in good faith may face temporary or permanent repercussions as determined by other members of the project's leadership.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4, available at [http://contributor-covenant.org/version/1/4][version]
+
+[homepage]: http://contributor-covenant.org
+[version]: http://contributor-covenant.org/version/1/4/
diff --git a/backend/vendor/github.com/imdario/mergo/LICENSE b/backend/vendor/github.com/imdario/mergo/LICENSE
new file mode 100644
index 00000000..68668029
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/LICENSE
@@ -0,0 +1,28 @@
+Copyright (c) 2013 Dario Castañé. All rights reserved.
+Copyright (c) 2012 The Go Authors. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+   * Redistributions in binary form must reproduce the above
+copyright notice, this list of conditions and the following disclaimer
+in the documentation and/or other materials provided with the
+distribution.
+   * Neither the name of Google Inc. nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/backend/vendor/github.com/imdario/mergo/README.md b/backend/vendor/github.com/imdario/mergo/README.md
new file mode 100644
index 00000000..8b76f1fb
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/README.md
@@ -0,0 +1,222 @@
+# Mergo
+
+A helper to merge structs and maps in Golang. Useful for configuration default values, avoiding messy if-statements.
+
+Also a lovely [comune](http://en.wikipedia.org/wiki/Mergo) (municipality) in the Province of Ancona in the Italian region of Marche.
+
+## Status
+
+It is ready for production use. [It is used in several projects by Docker, Google, The Linux Foundation, VMWare, Shopify, etc](https://github.com/imdario/mergo#mergo-in-the-wild).
+
+[![GoDoc][3]][4]
+[![GoCard][5]][6]
+[![Build Status][1]][2]
+[![Coverage Status][7]][8]
+[![Sourcegraph][9]][10]
+
+[1]: https://travis-ci.org/imdario/mergo.png
+[2]: https://travis-ci.org/imdario/mergo
+[3]: https://godoc.org/github.com/imdario/mergo?status.svg
+[4]: https://godoc.org/github.com/imdario/mergo
+[5]: https://goreportcard.com/badge/imdario/mergo
+[6]: https://goreportcard.com/report/github.com/imdario/mergo
+[7]: https://coveralls.io/repos/github/imdario/mergo/badge.svg?branch=master
+[8]: https://coveralls.io/github/imdario/mergo?branch=master
+[9]: https://sourcegraph.com/github.com/imdario/mergo/-/badge.svg
+[10]: https://sourcegraph.com/github.com/imdario/mergo?badge
+
+### Latest release
+
+[Release v0.3.6](https://github.com/imdario/mergo/releases/tag/v0.3.6).
+
+### Important note
+
+Please keep in mind that in [0.3.2](//github.com/imdario/mergo/releases/tag/0.3.2) Mergo changed `Merge()`and `Map()` signatures to support [transformers](#transformers). An optional/variadic argument has been added, so it won't break existing code.
+
+If you were using Mergo **before** April 6th 2015, please check your project works as intended after updating your local copy with ```go get -u github.com/imdario/mergo```. I apologize for any issue caused by its previous behavior and any future bug that Mergo could cause (I hope it won't!) in existing projects after the change (release 0.2.0).
+
+### Donations
+
+If Mergo is useful to you, consider buying me a coffee, a beer or making a monthly donation so I can keep building great free software. :heart_eyes:
+
+<a href='https://ko-fi.com/B0B58839' target='_blank'><img height='36' style='border:0px;height:36px;' src='https://az743702.vo.msecnd.net/cdn/kofi1.png?v=0' border='0' alt='Buy Me a Coffee at ko-fi.com' /></a>
+[![Beerpay](https://beerpay.io/imdario/mergo/badge.svg)](https://beerpay.io/imdario/mergo)
+[![Beerpay](https://beerpay.io/imdario/mergo/make-wish.svg)](https://beerpay.io/imdario/mergo)
+<a href="https://liberapay.com/dario/donate"><img alt="Donate using Liberapay" src="https://liberapay.com/assets/widgets/donate.svg"></a>
+
+### Mergo in the wild
+
+- [moby/moby](https://github.com/moby/moby)
+- [kubernetes/kubernetes](https://github.com/kubernetes/kubernetes)
+- [vmware/dispatch](https://github.com/vmware/dispatch)
+- [Shopify/themekit](https://github.com/Shopify/themekit)
+- [imdario/zas](https://github.com/imdario/zas)
+- [matcornic/hermes](https://github.com/matcornic/hermes)
+- [OpenBazaar/openbazaar-go](https://github.com/OpenBazaar/openbazaar-go)
+- [kataras/iris](https://github.com/kataras/iris)
+- [michaelsauter/crane](https://github.com/michaelsauter/crane)
+- [go-task/task](https://github.com/go-task/task)
+- [sensu/uchiwa](https://github.com/sensu/uchiwa)
+- [ory/hydra](https://github.com/ory/hydra)
+- [sisatech/vcli](https://github.com/sisatech/vcli)
+- [dairycart/dairycart](https://github.com/dairycart/dairycart)
+- [projectcalico/felix](https://github.com/projectcalico/felix)
+- [resin-os/balena](https://github.com/resin-os/balena)
+- [go-kivik/kivik](https://github.com/go-kivik/kivik)
+- [Telefonica/govice](https://github.com/Telefonica/govice)
+- [supergiant/supergiant](supergiant/supergiant)
+- [SergeyTsalkov/brooce](https://github.com/SergeyTsalkov/brooce)
+- [soniah/dnsmadeeasy](https://github.com/soniah/dnsmadeeasy)
+- [ohsu-comp-bio/funnel](https://github.com/ohsu-comp-bio/funnel)
+- [EagerIO/Stout](https://github.com/EagerIO/Stout)
+- [lynndylanhurley/defsynth-api](https://github.com/lynndylanhurley/defsynth-api)
+- [russross/canvasassignments](https://github.com/russross/canvasassignments)
+- [rdegges/cryptly-api](https://github.com/rdegges/cryptly-api)
+- [casualjim/exeggutor](https://github.com/casualjim/exeggutor)
+- [divshot/gitling](https://github.com/divshot/gitling)
+- [RWJMurphy/gorl](https://github.com/RWJMurphy/gorl)
+- [andrerocker/deploy42](https://github.com/andrerocker/deploy42)
+- [elwinar/rambler](https://github.com/elwinar/rambler)
+- [tmaiaroto/gopartman](https://github.com/tmaiaroto/gopartman)
+- [jfbus/impressionist](https://github.com/jfbus/impressionist)
+- [Jmeyering/zealot](https://github.com/Jmeyering/zealot)
+- [godep-migrator/rigger-host](https://github.com/godep-migrator/rigger-host)
+- [Dronevery/MultiwaySwitch-Go](https://github.com/Dronevery/MultiwaySwitch-Go)
+- [thoas/picfit](https://github.com/thoas/picfit)
+- [mantasmatelis/whooplist-server](https://github.com/mantasmatelis/whooplist-server)
+- [jnuthong/item_search](https://github.com/jnuthong/item_search)
+- [bukalapak/snowboard](https://github.com/bukalapak/snowboard)
+
+## Installation
+
+    go get github.com/imdario/mergo
+
+    // use in your .go code
+    import (
+        "github.com/imdario/mergo"
+    )
+
+## Usage
+
+You can only merge same-type structs with exported fields initialized as zero value of their type and same-types maps. Mergo won't merge unexported (private) fields but will do recursively any exported one. It won't merge empty structs value as [they are not considered zero values](https://golang.org/ref/spec#The_zero_value) either. Also maps will be merged recursively except for structs inside maps (because they are not addressable using Go reflection).
+
+```go
+if err := mergo.Merge(&dst, src); err != nil {
+    // ...
+}
+```
+
+Also, you can merge overwriting values using the transformer `WithOverride`.
+
+```go
+if err := mergo.Merge(&dst, src, mergo.WithOverride); err != nil {
+    // ...
+}
+```
+
+Additionally, you can map a `map[string]interface{}` to a struct (and otherwise, from struct to map), following the same restrictions as in `Merge()`. Keys are capitalized to find each corresponding exported field.
+
+```go
+if err := mergo.Map(&dst, srcMap); err != nil {
+    // ...
+}
+```
+
+Warning: if you map a struct to map, it won't do it recursively. Don't expect Mergo to map struct members of your struct as `map[string]interface{}`. They will be just assigned as values.
+
+More information and examples in [godoc documentation](http://godoc.org/github.com/imdario/mergo).
+
+### Nice example
+
+```go
+package main
+
+import (
+	"fmt"
+	"github.com/imdario/mergo"
+)
+
+type Foo struct {
+	A string
+	B int64
+}
+
+func main() {
+	src := Foo{
+		A: "one",
+		B: 2,
+	}
+	dest := Foo{
+		A: "two",
+	}
+	mergo.Merge(&dest, src)
+	fmt.Println(dest)
+	// Will print
+	// {two 2}
+}
+```
+
+Note: if test are failing due missing package, please execute:
+
+    go get gopkg.in/yaml.v2
+
+### Transformers
+
+Transformers allow to merge specific types differently than in the default behavior. In other words, now you can customize how some types are merged. For example, `time.Time` is a struct; it doesn't have zero value but IsZero can return true because it has fields with zero value. How can we merge a non-zero `time.Time`?
+
+```go
+package main
+
+import (
+	"fmt"
+	"github.com/imdario/mergo"
+        "reflect"
+        "time"
+)
+
+type timeTransfomer struct {
+}
+
+func (t timeTransfomer) Transformer(typ reflect.Type) func(dst, src reflect.Value) error {
+	if typ == reflect.TypeOf(time.Time{}) {
+		return func(dst, src reflect.Value) error {
+			if dst.CanSet() {
+				isZero := dst.MethodByName("IsZero")
+				result := isZero.Call([]reflect.Value{})
+				if result[0].Bool() {
+					dst.Set(src)
+				}
+			}
+			return nil
+		}
+	}
+	return nil
+}
+
+type Snapshot struct {
+	Time time.Time
+	// ...
+}
+
+func main() {
+	src := Snapshot{time.Now()}
+	dest := Snapshot{}
+	mergo.Merge(&dest, src, mergo.WithTransformers(timeTransfomer{}))
+	fmt.Println(dest)
+	// Will print
+	// { 2018-01-12 01:15:00 +0000 UTC m=+0.000000001 }
+}
+```
+
+
+## Contact me
+
+If I can help you, you have an idea or you are using Mergo in your projects, don't hesitate to drop me a line (or a pull request): [@im_dario](https://twitter.com/im_dario)
+
+## About
+
+Written by [Dario Castañé](http://dario.im).
+
+## License
+
+[BSD 3-Clause](http://opensource.org/licenses/BSD-3-Clause) license, as [Go language](http://golang.org/LICENSE).
diff --git a/backend/vendor/github.com/imdario/mergo/doc.go b/backend/vendor/github.com/imdario/mergo/doc.go
new file mode 100644
index 00000000..6e9aa7ba
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/doc.go
@@ -0,0 +1,44 @@
+// Copyright 2013 Dario Castañé. All rights reserved.
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+/*
+Package mergo merges same-type structs and maps by setting default values in zero-value fields.
+
+Mergo won't merge unexported (private) fields but will do recursively any exported one. It also won't merge structs inside maps (because they are not addressable using Go reflection).
+
+Usage
+
+From my own work-in-progress project:
+
+	type networkConfig struct {
+		Protocol string
+		Address string
+		ServerType string `json: "server_type"`
+		Port uint16
+	}
+
+	type FssnConfig struct {
+		Network networkConfig
+	}
+
+	var fssnDefault = FssnConfig {
+		networkConfig {
+			"tcp",
+			"127.0.0.1",
+			"http",
+			31560,
+		},
+	}
+
+	// Inside a function [...]
+
+	if err := mergo.Merge(&config, fssnDefault); err != nil {
+		log.Fatal(err)
+	}
+
+	// More code [...]
+
+*/
+package mergo
diff --git a/backend/vendor/github.com/imdario/mergo/map.go b/backend/vendor/github.com/imdario/mergo/map.go
new file mode 100644
index 00000000..6ea38e63
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/map.go
@@ -0,0 +1,174 @@
+// Copyright 2014 Dario Castañé. All rights reserved.
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Based on src/pkg/reflect/deepequal.go from official
+// golang's stdlib.
+
+package mergo
+
+import (
+	"fmt"
+	"reflect"
+	"unicode"
+	"unicode/utf8"
+)
+
+func changeInitialCase(s string, mapper func(rune) rune) string {
+	if s == "" {
+		return s
+	}
+	r, n := utf8.DecodeRuneInString(s)
+	return string(mapper(r)) + s[n:]
+}
+
+func isExported(field reflect.StructField) bool {
+	r, _ := utf8.DecodeRuneInString(field.Name)
+	return r >= 'A' && r <= 'Z'
+}
+
+// Traverses recursively both values, assigning src's fields values to dst.
+// The map argument tracks comparisons that have already been seen, which allows
+// short circuiting on recursive types.
+func deepMap(dst, src reflect.Value, visited map[uintptr]*visit, depth int, config *Config) (err error) {
+	overwrite := config.Overwrite
+	if dst.CanAddr() {
+		addr := dst.UnsafeAddr()
+		h := 17 * addr
+		seen := visited[h]
+		typ := dst.Type()
+		for p := seen; p != nil; p = p.next {
+			if p.ptr == addr && p.typ == typ {
+				return nil
+			}
+		}
+		// Remember, remember...
+		visited[h] = &visit{addr, typ, seen}
+	}
+	zeroValue := reflect.Value{}
+	switch dst.Kind() {
+	case reflect.Map:
+		dstMap := dst.Interface().(map[string]interface{})
+		for i, n := 0, src.NumField(); i < n; i++ {
+			srcType := src.Type()
+			field := srcType.Field(i)
+			if !isExported(field) {
+				continue
+			}
+			fieldName := field.Name
+			fieldName = changeInitialCase(fieldName, unicode.ToLower)
+			if v, ok := dstMap[fieldName]; !ok || (isEmptyValue(reflect.ValueOf(v)) || overwrite) {
+				dstMap[fieldName] = src.Field(i).Interface()
+			}
+		}
+	case reflect.Ptr:
+		if dst.IsNil() {
+			v := reflect.New(dst.Type().Elem())
+			dst.Set(v)
+		}
+		dst = dst.Elem()
+		fallthrough
+	case reflect.Struct:
+		srcMap := src.Interface().(map[string]interface{})
+		for key := range srcMap {
+			srcValue := srcMap[key]
+			fieldName := changeInitialCase(key, unicode.ToUpper)
+			dstElement := dst.FieldByName(fieldName)
+			if dstElement == zeroValue {
+				// We discard it because the field doesn't exist.
+				continue
+			}
+			srcElement := reflect.ValueOf(srcValue)
+			dstKind := dstElement.Kind()
+			srcKind := srcElement.Kind()
+			if srcKind == reflect.Ptr && dstKind != reflect.Ptr {
+				srcElement = srcElement.Elem()
+				srcKind = reflect.TypeOf(srcElement.Interface()).Kind()
+			} else if dstKind == reflect.Ptr {
+				// Can this work? I guess it can't.
+				if srcKind != reflect.Ptr && srcElement.CanAddr() {
+					srcPtr := srcElement.Addr()
+					srcElement = reflect.ValueOf(srcPtr)
+					srcKind = reflect.Ptr
+				}
+			}
+
+			if !srcElement.IsValid() {
+				continue
+			}
+			if srcKind == dstKind {
+				if err = deepMerge(dstElement, srcElement, visited, depth+1, config); err != nil {
+					return
+				}
+			} else if dstKind == reflect.Interface && dstElement.Kind() == reflect.Interface {
+				if err = deepMerge(dstElement, srcElement, visited, depth+1, config); err != nil {
+					return
+				}
+			} else if srcKind == reflect.Map {
+				if err = deepMap(dstElement, srcElement, visited, depth+1, config); err != nil {
+					return
+				}
+			} else {
+				return fmt.Errorf("type mismatch on %s field: found %v, expected %v", fieldName, srcKind, dstKind)
+			}
+		}
+	}
+	return
+}
+
+// Map sets fields' values in dst from src.
+// src can be a map with string keys or a struct. dst must be the opposite:
+// if src is a map, dst must be a valid pointer to struct. If src is a struct,
+// dst must be map[string]interface{}.
+// It won't merge unexported (private) fields and will do recursively
+// any exported field.
+// If dst is a map, keys will be src fields' names in lower camel case.
+// Missing key in src that doesn't match a field in dst will be skipped. This
+// doesn't apply if dst is a map.
+// This is separated method from Merge because it is cleaner and it keeps sane
+// semantics: merging equal types, mapping different (restricted) types.
+func Map(dst, src interface{}, opts ...func(*Config)) error {
+	return _map(dst, src, opts...)
+}
+
+// MapWithOverwrite will do the same as Map except that non-empty dst attributes will be overridden by
+// non-empty src attribute values.
+// Deprecated: Use Map(…) with WithOverride
+func MapWithOverwrite(dst, src interface{}, opts ...func(*Config)) error {
+	return _map(dst, src, append(opts, WithOverride)...)
+}
+
+func _map(dst, src interface{}, opts ...func(*Config)) error {
+	var (
+		vDst, vSrc reflect.Value
+		err        error
+	)
+	config := &Config{}
+
+	for _, opt := range opts {
+		opt(config)
+	}
+
+	if vDst, vSrc, err = resolveValues(dst, src); err != nil {
+		return err
+	}
+	// To be friction-less, we redirect equal-type arguments
+	// to deepMerge. Only because arguments can be anything.
+	if vSrc.Kind() == vDst.Kind() {
+		return deepMerge(vDst, vSrc, make(map[uintptr]*visit), 0, config)
+	}
+	switch vSrc.Kind() {
+	case reflect.Struct:
+		if vDst.Kind() != reflect.Map {
+			return ErrExpectedMapAsDestination
+		}
+	case reflect.Map:
+		if vDst.Kind() != reflect.Struct {
+			return ErrExpectedStructAsDestination
+		}
+	default:
+		return ErrNotSupported
+	}
+	return deepMap(vDst, vSrc, make(map[uintptr]*visit), 0, config)
+}
diff --git a/backend/vendor/github.com/imdario/mergo/merge.go b/backend/vendor/github.com/imdario/mergo/merge.go
new file mode 100644
index 00000000..44f70a89
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/merge.go
@@ -0,0 +1,252 @@
+// Copyright 2013 Dario Castañé. All rights reserved.
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Based on src/pkg/reflect/deepequal.go from official
+// golang's stdlib.
+
+package mergo
+
+import (
+	"fmt"
+	"reflect"
+)
+
+func hasExportedField(dst reflect.Value) (exported bool) {
+	for i, n := 0, dst.NumField(); i < n; i++ {
+		field := dst.Type().Field(i)
+		if field.Anonymous && dst.Field(i).Kind() == reflect.Struct {
+			exported = exported || hasExportedField(dst.Field(i))
+		} else {
+			exported = exported || len(field.PkgPath) == 0
+		}
+	}
+	return
+}
+
+type Config struct {
+	Overwrite    bool
+	AppendSlice  bool
+	Transformers Transformers
+}
+
+type Transformers interface {
+	Transformer(reflect.Type) func(dst, src reflect.Value) error
+}
+
+// Traverses recursively both values, assigning src's fields values to dst.
+// The map argument tracks comparisons that have already been seen, which allows
+// short circuiting on recursive types.
+func deepMerge(dst, src reflect.Value, visited map[uintptr]*visit, depth int, config *Config) (err error) {
+	overwrite := config.Overwrite
+
+	if !src.IsValid() {
+		return
+	}
+	if dst.CanAddr() {
+		addr := dst.UnsafeAddr()
+		h := 17 * addr
+		seen := visited[h]
+		typ := dst.Type()
+		for p := seen; p != nil; p = p.next {
+			if p.ptr == addr && p.typ == typ {
+				return nil
+			}
+		}
+		// Remember, remember...
+		visited[h] = &visit{addr, typ, seen}
+	}
+
+	if config.Transformers != nil && !isEmptyValue(dst) {
+		if fn := config.Transformers.Transformer(dst.Type()); fn != nil {
+			err = fn(dst, src)
+			return
+		}
+	}
+
+	switch dst.Kind() {
+	case reflect.Struct:
+		if hasExportedField(dst) {
+			for i, n := 0, dst.NumField(); i < n; i++ {
+				if err = deepMerge(dst.Field(i), src.Field(i), visited, depth+1, config); err != nil {
+					return
+				}
+			}
+		} else {
+			if dst.CanSet() && !isEmptyValue(src) && (overwrite || isEmptyValue(dst)) {
+				dst.Set(src)
+			}
+		}
+	case reflect.Map:
+		if dst.IsNil() && !src.IsNil() {
+			dst.Set(reflect.MakeMap(dst.Type()))
+		}
+		for _, key := range src.MapKeys() {
+			srcElement := src.MapIndex(key)
+			if !srcElement.IsValid() {
+				continue
+			}
+			dstElement := dst.MapIndex(key)
+			switch srcElement.Kind() {
+			case reflect.Chan, reflect.Func, reflect.Map, reflect.Interface, reflect.Slice:
+				if srcElement.IsNil() {
+					continue
+				}
+				fallthrough
+			default:
+				if !srcElement.CanInterface() {
+					continue
+				}
+				switch reflect.TypeOf(srcElement.Interface()).Kind() {
+				case reflect.Struct:
+					fallthrough
+				case reflect.Ptr:
+					fallthrough
+				case reflect.Map:
+					srcMapElm := srcElement
+					dstMapElm := dstElement
+					if srcMapElm.CanInterface() {
+						srcMapElm = reflect.ValueOf(srcMapElm.Interface())
+						if dstMapElm.IsValid() {
+							dstMapElm = reflect.ValueOf(dstMapElm.Interface())
+						}
+					}
+					if err = deepMerge(dstMapElm, srcMapElm, visited, depth+1, config); err != nil {
+						return
+					}
+				case reflect.Slice:
+					srcSlice := reflect.ValueOf(srcElement.Interface())
+
+					var dstSlice reflect.Value
+					if !dstElement.IsValid() || dstElement.IsNil() {
+						dstSlice = reflect.MakeSlice(srcSlice.Type(), 0, srcSlice.Len())
+					} else {
+						dstSlice = reflect.ValueOf(dstElement.Interface())
+					}
+
+					if !isEmptyValue(src) && (overwrite || isEmptyValue(dst)) && !config.AppendSlice {
+						dstSlice = srcSlice
+					} else if config.AppendSlice {
+						if srcSlice.Type() != dstSlice.Type() {
+							return fmt.Errorf("cannot append two slice with different type (%s, %s)", srcSlice.Type(), dstSlice.Type())
+						}
+						dstSlice = reflect.AppendSlice(dstSlice, srcSlice)
+					}
+					dst.SetMapIndex(key, dstSlice)
+				}
+			}
+			if dstElement.IsValid() && reflect.TypeOf(srcElement.Interface()).Kind() == reflect.Map {
+				continue
+			}
+
+			if srcElement.IsValid() && (overwrite || (!dstElement.IsValid() || isEmptyValue(dstElement))) {
+				if dst.IsNil() {
+					dst.Set(reflect.MakeMap(dst.Type()))
+				}
+				dst.SetMapIndex(key, srcElement)
+			}
+		}
+	case reflect.Slice:
+		if !dst.CanSet() {
+			break
+		}
+		if !isEmptyValue(src) && (overwrite || isEmptyValue(dst)) && !config.AppendSlice {
+			dst.Set(src)
+		} else if config.AppendSlice {
+			if src.Type() != dst.Type() {
+				return fmt.Errorf("cannot append two slice with different type (%s, %s)", src.Type(), dst.Type())
+			}
+			dst.Set(reflect.AppendSlice(dst, src))
+		}
+	case reflect.Ptr:
+		fallthrough
+	case reflect.Interface:
+		if src.IsNil() {
+			break
+		}
+		if src.Kind() != reflect.Interface {
+			if dst.IsNil() || overwrite {
+				if dst.CanSet() && (overwrite || isEmptyValue(dst)) {
+					dst.Set(src)
+				}
+			} else if src.Kind() == reflect.Ptr {
+				if err = deepMerge(dst.Elem(), src.Elem(), visited, depth+1, config); err != nil {
+					return
+				}
+			} else if dst.Elem().Type() == src.Type() {
+				if err = deepMerge(dst.Elem(), src, visited, depth+1, config); err != nil {
+					return
+				}
+			} else {
+				return ErrDifferentArgumentsTypes
+			}
+			break
+		}
+		if dst.IsNil() || overwrite {
+			if dst.CanSet() && (overwrite || isEmptyValue(dst)) {
+				dst.Set(src)
+			}
+		} else if err = deepMerge(dst.Elem(), src.Elem(), visited, depth+1, config); err != nil {
+			return
+		}
+	default:
+		if dst.CanSet() && !isEmptyValue(src) && (overwrite || isEmptyValue(dst)) {
+			dst.Set(src)
+		}
+	}
+	return
+}
+
+// Merge will fill any empty for value type attributes on the dst struct using corresponding
+// src attributes if they themselves are not empty. dst and src must be valid same-type structs
+// and dst must be a pointer to struct.
+// It won't merge unexported (private) fields and will do recursively any exported field.
+func Merge(dst, src interface{}, opts ...func(*Config)) error {
+	return merge(dst, src, opts...)
+}
+
+// MergeWithOverwrite will do the same as Merge except that non-empty dst attributes will be overriden by
+// non-empty src attribute values.
+// Deprecated: use Merge(…) with WithOverride
+func MergeWithOverwrite(dst, src interface{}, opts ...func(*Config)) error {
+	return merge(dst, src, append(opts, WithOverride)...)
+}
+
+// WithTransformers adds transformers to merge, allowing to customize the merging of some types.
+func WithTransformers(transformers Transformers) func(*Config) {
+	return func(config *Config) {
+		config.Transformers = transformers
+	}
+}
+
+// WithOverride will make merge override non-empty dst attributes with non-empty src attributes values.
+func WithOverride(config *Config) {
+	config.Overwrite = true
+}
+
+// WithAppendSlice will make merge append slices instead of overwriting it
+func WithAppendSlice(config *Config) {
+	config.AppendSlice = true
+}
+
+func merge(dst, src interface{}, opts ...func(*Config)) error {
+	var (
+		vDst, vSrc reflect.Value
+		err        error
+	)
+
+	config := &Config{}
+
+	for _, opt := range opts {
+		opt(config)
+	}
+
+	if vDst, vSrc, err = resolveValues(dst, src); err != nil {
+		return err
+	}
+	if vDst.Type() != vSrc.Type() {
+		return ErrDifferentArgumentsTypes
+	}
+	return deepMerge(vDst, vSrc, make(map[uintptr]*visit), 0, config)
+}
diff --git a/backend/vendor/github.com/imdario/mergo/mergo.go b/backend/vendor/github.com/imdario/mergo/mergo.go
new file mode 100644
index 00000000..a82fea2f
--- /dev/null
+++ b/backend/vendor/github.com/imdario/mergo/mergo.go
@@ -0,0 +1,97 @@
+// Copyright 2013 Dario Castañé. All rights reserved.
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Based on src/pkg/reflect/deepequal.go from official
+// golang's stdlib.
+
+package mergo
+
+import (
+	"errors"
+	"reflect"
+)
+
+// Errors reported by Mergo when it finds invalid arguments.
+var (
+	ErrNilArguments                = errors.New("src and dst must not be nil")
+	ErrDifferentArgumentsTypes     = errors.New("src and dst must be of same type")
+	ErrNotSupported                = errors.New("only structs and maps are supported")
+	ErrExpectedMapAsDestination    = errors.New("dst was expected to be a map")
+	ErrExpectedStructAsDestination = errors.New("dst was expected to be a struct")
+)
+
+// During deepMerge, must keep track of checks that are
+// in progress.  The comparison algorithm assumes that all
+// checks in progress are true when it reencounters them.
+// Visited are stored in a map indexed by 17 * a1 + a2;
+type visit struct {
+	ptr  uintptr
+	typ  reflect.Type
+	next *visit
+}
+
+// From src/pkg/encoding/json/encode.go.
+func isEmptyValue(v reflect.Value) bool {
+	switch v.Kind() {
+	case reflect.Array, reflect.Map, reflect.Slice, reflect.String:
+		return v.Len() == 0
+	case reflect.Bool:
+		return !v.Bool()
+	case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+		return v.Int() == 0
+	case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr:
+		return v.Uint() == 0
+	case reflect.Float32, reflect.Float64:
+		return v.Float() == 0
+	case reflect.Interface, reflect.Ptr:
+		if v.IsNil() {
+			return true
+		}
+		return isEmptyValue(v.Elem())
+	case reflect.Func:
+		return v.IsNil()
+	case reflect.Invalid:
+		return true
+	}
+	return false
+}
+
+func resolveValues(dst, src interface{}) (vDst, vSrc reflect.Value, err error) {
+	if dst == nil || src == nil {
+		err = ErrNilArguments
+		return
+	}
+	vDst = reflect.ValueOf(dst).Elem()
+	if vDst.Kind() != reflect.Struct && vDst.Kind() != reflect.Map {
+		err = ErrNotSupported
+		return
+	}
+	vSrc = reflect.ValueOf(src)
+	// We check if vSrc is a pointer to dereference it.
+	if vSrc.Kind() == reflect.Ptr {
+		vSrc = vSrc.Elem()
+	}
+	return
+}
+
+// Traverses recursively both values, assigning src's fields values to dst.
+// The map argument tracks comparisons that have already been seen, which allows
+// short circuiting on recursive types.
+func deeper(dst, src reflect.Value, visited map[uintptr]*visit, depth int) (err error) {
+	if dst.CanAddr() {
+		addr := dst.UnsafeAddr()
+		h := 17 * addr
+		seen := visited[h]
+		typ := dst.Type()
+		for p := seen; p != nil; p = p.next {
+			if p.ptr == addr && p.typ == typ {
+				return nil
+			}
+		}
+		// Remember, remember...
+		visited[h] = &visit{addr, typ, seen}
+	}
+	return // TODO refactor
+}
diff --git a/backend/vendor/github.com/jaytaylor/html2text/.gitignore b/backend/vendor/github.com/jaytaylor/html2text/.gitignore
new file mode 100644
index 00000000..daf913b1
--- /dev/null
+++ b/backend/vendor/github.com/jaytaylor/html2text/.gitignore
@@ -0,0 +1,24 @@
+# Compiled Object files, Static and Dynamic libs (Shared Objects)
+*.o
+*.a
+*.so
+
+# Folders
+_obj
+_test
+
+# Architecture specific extensions/prefixes
+*.[568vq]
+[568vq].out
+
+*.cgo1.go
+*.cgo2.c
+_cgo_defun.c
+_cgo_gotypes.go
+_cgo_export.*
+
+_testmain.go
+
+*.exe
+*.test
+*.prof
diff --git a/backend/vendor/github.com/jaytaylor/html2text/.travis.yml b/backend/vendor/github.com/jaytaylor/html2text/.travis.yml
new file mode 100644
index 00000000..6c7f48ef
--- /dev/null
+++ b/backend/vendor/github.com/jaytaylor/html2text/.travis.yml
@@ -0,0 +1,14 @@
+language: go
+go:
+  - tip
+  - 1.8
+  - 1.7
+  - 1.6
+  - 1.5
+  - 1.4
+  - 1.3
+  - 1.2
+notifications:
+  email:
+    on_success: change
+    on_failure: always
diff --git a/backend/vendor/github.com/jaytaylor/html2text/LICENSE b/backend/vendor/github.com/jaytaylor/html2text/LICENSE
new file mode 100644
index 00000000..24dc4abe
--- /dev/null
+++ b/backend/vendor/github.com/jaytaylor/html2text/LICENSE
@@ -0,0 +1,22 @@
+The MIT License (MIT)
+
+Copyright (c) 2015 Jay Taylor
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
diff --git a/backend/vendor/github.com/jaytaylor/html2text/README.md b/backend/vendor/github.com/jaytaylor/html2text/README.md
new file mode 100644
index 00000000..6b2494ee
--- /dev/null
+++ b/backend/vendor/github.com/jaytaylor/html2text/README.md
@@ -0,0 +1,137 @@
+# html2text
+
+[![Documentation](https://godoc.org/github.com/jaytaylor/html2text?status.svg)](https://godoc.org/github.com/jaytaylor/html2text)
+[![Build Status](https://travis-ci.org/jaytaylor/html2text.svg?branch=master)](https://travis-ci.org/jaytaylor/html2text)
+[![Report Card](https://goreportcard.com/badge/github.com/jaytaylor/html2text)](https://goreportcard.com/report/github.com/jaytaylor/html2text)
+
+### Converts HTML into text of the markdown-flavored variety
+
+
+## Introduction
+
+Ensure your emails are readable by all!
+
+Turns HTML into raw text, useful for sending fancy HTML emails with an equivalently nicely formatted TXT document as a fallback (e.g. for people who don't allow HTML emails or have other display issues).
+
+html2text is a simple golang package for rendering HTML into plaintext.
+
+There are still lots of improvements to be had, but FWIW this has worked fine for my [basic] HTML-2-text needs.
+
+It requires go 1.x or newer ;)
+
+
+## Download the package
+
+```bash
+go get jaytaylor.com/html2text
+```
+
+## Example usage
+
+```go
+package main
+
+import (
+	"fmt"
+
+	"jaytaylor.com/html2text"
+)
+
+func main() {
+	inputHTML := `
+<html>
+  <head>
+    <title>My Mega Service</title>
+    <link rel=\"stylesheet\" href=\"main.css\">
+    <style type=\"text/css\">body { color: #fff; }</style>
+  </head>
+
+  <body>
+    <div class="logo">
+      <a href="http://jaytaylor.com/"><img src="/logo-image.jpg" alt="Mega Service"/></a>
+    </div>
+
+    <h1>Welcome to your new account on my service!</h1>
+
+    <p>
+      Here is some more information:
+
+      <ul>
+        <li>Link 1: <a href="https://example.com">Example.com</a></li>
+        <li>Link 2: <a href="https://example2.com">Example2.com</a></li>
+        <li>Something else</li>
+      </ul>
+    </p>
+
+    <table>
+      <thead>
+        <tr><th>Header 1</th><th>Header 2</th></tr>
+      </thead>
+      <tfoot>
+        <tr><td>Footer 1</td><td>Footer 2</td></tr>
+      </tfoot>
+      <tbody>
+        <tr><td>Row 1 Col 1</td><td>Row 1 Col 2</td></tr>
+        <tr><td>Row 2 Col 1</td><td>Row 2 Col 2</td></tr>
+      </tbody>
+    </table>
+  </body>
+</html>`
+
+	text, err := html2text.FromString(inputHTML, html2text.Options{PrettyTables: true})
+	if err != nil {
+		panic(err)
+	}
+	fmt.Println(text)
+}
+```
+
+Output:
+```
+Mega Service ( http://jaytaylor.com/ )
+
+******************************************
+Welcome to your new account on my service!
+******************************************
+
+Here is some more information:
+
+* Link 1: Example.com ( https://example.com )
+* Link 2: Example2.com ( https://example2.com )
+* Something else
+
++-------------+-------------+
+|  HEADER 1   |  HEADER 2   |
++-------------+-------------+
+| Row 1 Col 1 | Row 1 Col 2 |
+| Row 2 Col 1 | Row 2 Col 2 |
++-------------+-------------+
+|  FOOTER 1   |  FOOTER 2   |
++-------------+-------------+
+```
+
+
+## Unit-tests
+
+Running the unit-tests is straightforward and standard:
+
+```bash
+go test
+```
+
+
+# License
+
+Permissive MIT license.
+
+
+## Contact
+
+You are more than welcome to open issues and send pull requests if you find a bug or want a new feature.
+
+If you appreciate this library please feel free to drop me a line and tell me!  It's always nice to hear from people who have benefitted from my work.
+
+Email: jay at (my github username).com
+
+Twitter: [@jtaylor](https://twitter.com/jtaylor)
+
diff --git a/backend/vendor/github.com/jaytaylor/html2text/html2text.go b/backend/vendor/github.com/jaytaylor/html2text/html2text.go
new file mode 100644
index 00000000..fa369909
--- /dev/null
+++ b/backend/vendor/github.com/jaytaylor/html2text/html2text.go
@@ -0,0 +1,473 @@
+package html2text
+
+import (
+	"bytes"
+	"io"
+	"regexp"
+	"strings"
+	"unicode"
+
+	"github.com/olekukonko/tablewriter"
+	"github.com/ssor/bom"
+	"golang.org/x/net/html"
+	"golang.org/x/net/html/atom"
+)
+
+// Options provide toggles and overrides to control specific rendering behaviors.
+type Options struct {
+	PrettyTables bool // Turns on pretty ASCII rendering for table elements.
+	OmitLinks    bool // Turns on omitting links
+}
+
+// FromHTMLNode renders text output from a pre-parsed HTML document.
+func FromHTMLNode(doc *html.Node, o ...Options) (string, error) {
+	var options Options
+	if len(o) > 0 {
+		options = o[0]
+	}
+
+	ctx := textifyTraverseContext{
+		buf:     bytes.Buffer{},
+		options: options,
+	}
+	if err := ctx.traverse(doc); err != nil {
+		return "", err
+	}
+
+	text := strings.TrimSpace(newlineRe.ReplaceAllString(
+		strings.Replace(ctx.buf.String(), "\n ", "\n", -1), "\n\n"),
+	)
+	return text, nil
+}
+
+// FromReader renders text output after parsing HTML for the specified
+// io.Reader.
+func FromReader(reader io.Reader, options ...Options) (string, error) {
+	newReader, err := bom.NewReaderWithoutBom(reader)
+	if err != nil {
+		return "", err
+	}
+	doc, err := html.Parse(newReader)
+	if err != nil {
+		return "", err
+	}
+	return FromHTMLNode(doc, options...)
+}
+
+// FromString parses HTML from the input string, then renders the text form.
+func FromString(input string, options ...Options) (string, error) {
+	bs := bom.CleanBom([]byte(input))
+	text, err := FromReader(bytes.NewReader(bs), options...)
+	if err != nil {
+		return "", err
+	}
+	return text, nil
+}
+
+var (
+	spacingRe = regexp.MustCompile(`[ \r\n\t]+`)
+	newlineRe = regexp.MustCompile(`\n\n+`)
+)
+
+// traverseTableCtx holds text-related context.
+type textifyTraverseContext struct {
+	buf bytes.Buffer
+
+	prefix          string
+	tableCtx        tableTraverseContext
+	options         Options
+	endsWithSpace   bool
+	justClosedDiv   bool
+	blockquoteLevel int
+	lineLength      int
+	isPre           bool
+}
+
+// tableTraverseContext holds table ASCII-form related context.
+type tableTraverseContext struct {
+	header     []string
+	body       [][]string
+	footer     []string
+	tmpRow     int
+	isInFooter bool
+}
+
+func (tableCtx *tableTraverseContext) init() {
+	tableCtx.body = [][]string{}
+	tableCtx.header = []string{}
+	tableCtx.footer = []string{}
+	tableCtx.isInFooter = false
+	tableCtx.tmpRow = 0
+}
+
+func (ctx *textifyTraverseContext) handleElement(node *html.Node) error {
+	ctx.justClosedDiv = false
+
+	switch node.DataAtom {
+	case atom.Br:
+		return ctx.emit("\n")
+
+	case atom.H1, atom.H2, atom.H3:
+		subCtx := textifyTraverseContext{}
+		if err := subCtx.traverseChildren(node); err != nil {
+			return err
+		}
+
+		str := subCtx.buf.String()
+		dividerLen := 0
+		for _, line := range strings.Split(str, "\n") {
+			if lineLen := len([]rune(line)); lineLen-1 > dividerLen {
+				dividerLen = lineLen - 1
+			}
+		}
+		var divider string
+		if node.DataAtom == atom.H1 {
+			divider = strings.Repeat("*", dividerLen)
+		} else {
+			divider = strings.Repeat("-", dividerLen)
+		}
+
+		if node.DataAtom == atom.H3 {
+			return ctx.emit("\n\n" + str + "\n" + divider + "\n\n")
+		}
+		return ctx.emit("\n\n" + divider + "\n" + str + "\n" + divider + "\n\n")
+
+	case atom.Blockquote:
+		ctx.blockquoteLevel++
+		ctx.prefix = strings.Repeat(">", ctx.blockquoteLevel) + " "
+		if err := ctx.emit("\n"); err != nil {
+			return err
+		}
+		if ctx.blockquoteLevel == 1 {
+			if err := ctx.emit("\n"); err != nil {
+				return err
+			}
+		}
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+		ctx.blockquoteLevel--
+		ctx.prefix = strings.Repeat(">", ctx.blockquoteLevel)
+		if ctx.blockquoteLevel > 0 {
+			ctx.prefix += " "
+		}
+		return ctx.emit("\n\n")
+
+	case atom.Div:
+		if ctx.lineLength > 0 {
+			if err := ctx.emit("\n"); err != nil {
+				return err
+			}
+		}
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+		var err error
+		if !ctx.justClosedDiv {
+			err = ctx.emit("\n")
+		}
+		ctx.justClosedDiv = true
+		return err
+
+	case atom.Li:
+		if err := ctx.emit("* "); err != nil {
+			return err
+		}
+
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+
+		return ctx.emit("\n")
+
+	case atom.B, atom.Strong:
+		subCtx := textifyTraverseContext{}
+		subCtx.endsWithSpace = true
+		if err := subCtx.traverseChildren(node); err != nil {
+			return err
+		}
+		str := subCtx.buf.String()
+		return ctx.emit("*" + str + "*")
+
+	case atom.A:
+		linkText := ""
+		// For simple link element content with single text node only, peek at the link text.
+		if node.FirstChild != nil && node.FirstChild.NextSibling == nil && node.FirstChild.Type == html.TextNode {
+			linkText = node.FirstChild.Data
+		}
+
+		// If image is the only child, take its alt text as the link text.
+		if img := node.FirstChild; img != nil && node.LastChild == img && img.DataAtom == atom.Img {
+			if altText := getAttrVal(img, "alt"); altText != "" {
+				if err := ctx.emit(altText); err != nil {
+					return err
+				}
+			}
+		} else if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+
+		hrefLink := ""
+		if attrVal := getAttrVal(node, "href"); attrVal != "" {
+			attrVal = ctx.normalizeHrefLink(attrVal)
+			// Don't print link href if it matches link element content or if the link is empty.
+			if !ctx.options.OmitLinks && attrVal != "" && linkText != attrVal {
+				hrefLink = "( " + attrVal + " )"
+			}
+		}
+
+		return ctx.emit(hrefLink)
+
+	case atom.P, atom.Ul:
+		return ctx.paragraphHandler(node)
+
+	case atom.Table, atom.Tfoot, atom.Th, atom.Tr, atom.Td:
+		if ctx.options.PrettyTables {
+			return ctx.handleTableElement(node)
+		} else if node.DataAtom == atom.Table {
+			return ctx.paragraphHandler(node)
+		}
+		return ctx.traverseChildren(node)
+
+	case atom.Pre:
+		ctx.isPre = true
+		err := ctx.traverseChildren(node)
+		ctx.isPre = false
+		return err
+
+	case atom.Style, atom.Script, atom.Head:
+		// Ignore the subtree.
+		return nil
+
+	default:
+		return ctx.traverseChildren(node)
+	}
+}
+
+// paragraphHandler renders node children surrounded by double newlines.
+func (ctx *textifyTraverseContext) paragraphHandler(node *html.Node) error {
+	if err := ctx.emit("\n\n"); err != nil {
+		return err
+	}
+	if err := ctx.traverseChildren(node); err != nil {
+		return err
+	}
+	return ctx.emit("\n\n")
+}
+
+// handleTableElement is only to be invoked when options.PrettyTables is active.
+func (ctx *textifyTraverseContext) handleTableElement(node *html.Node) error {
+	if !ctx.options.PrettyTables {
+		panic("handleTableElement invoked when PrettyTables not active")
+	}
+
+	switch node.DataAtom {
+	case atom.Table:
+		if err := ctx.emit("\n\n"); err != nil {
+			return err
+		}
+
+		// Re-intialize all table context.
+		ctx.tableCtx.init()
+
+		// Browse children, enriching context with table data.
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+
+		buf := &bytes.Buffer{}
+		table := tablewriter.NewWriter(buf)
+		table.SetHeader(ctx.tableCtx.header)
+		table.SetFooter(ctx.tableCtx.footer)
+		table.AppendBulk(ctx.tableCtx.body)
+
+		// Render the table using ASCII.
+		table.Render()
+		if err := ctx.emit(buf.String()); err != nil {
+			return err
+		}
+
+		return ctx.emit("\n\n")
+
+	case atom.Tfoot:
+		ctx.tableCtx.isInFooter = true
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+		ctx.tableCtx.isInFooter = false
+
+	case atom.Tr:
+		ctx.tableCtx.body = append(ctx.tableCtx.body, []string{})
+		if err := ctx.traverseChildren(node); err != nil {
+			return err
+		}
+		ctx.tableCtx.tmpRow++
+
+	case atom.Th:
+		res, err := ctx.renderEachChild(node)
+		if err != nil {
+			return err
+		}
+
+		ctx.tableCtx.header = append(ctx.tableCtx.header, res)
+
+	case atom.Td:
+		res, err := ctx.renderEachChild(node)
+		if err != nil {
+			return err
+		}
+
+		if ctx.tableCtx.isInFooter {
+			ctx.tableCtx.footer = append(ctx.tableCtx.footer, res)
+		} else {
+			ctx.tableCtx.body[ctx.tableCtx.tmpRow] = append(ctx.tableCtx.body[ctx.tableCtx.tmpRow], res)
+		}
+
+	}
+	return nil
+}
+
+func (ctx *textifyTraverseContext) traverse(node *html.Node) error {
+	switch node.Type {
+	default:
+		return ctx.traverseChildren(node)
+
+	case html.TextNode:
+		var data string
+		if ctx.isPre {
+			data = node.Data
+		} else {
+			data = strings.Trim(spacingRe.ReplaceAllString(node.Data, " "), " ")
+		}
+		return ctx.emit(data)
+
+	case html.ElementNode:
+		return ctx.handleElement(node)
+	}
+}
+
+func (ctx *textifyTraverseContext) traverseChildren(node *html.Node) error {
+	for c := node.FirstChild; c != nil; c = c.NextSibling {
+		if err := ctx.traverse(c); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+
+func (ctx *textifyTraverseContext) emit(data string) error {
+	if data == "" {
+		return nil
+	}
+	var (
+		lines = ctx.breakLongLines(data)
+		err   error
+	)
+	for _, line := range lines {
+		runes := []rune(line)
+		startsWithSpace := unicode.IsSpace(runes[0])
+		if !startsWithSpace && !ctx.endsWithSpace && !strings.HasPrefix(data, ".") {
+			if err = ctx.buf.WriteByte(' '); err != nil {
+				return err
+			}
+			ctx.lineLength++
+		}
+		ctx.endsWithSpace = unicode.IsSpace(runes[len(runes)-1])
+		for _, c := range line {
+			if _, err = ctx.buf.WriteString(string(c)); err != nil {
+				return err
+			}
+			ctx.lineLength++
+			if c == '\n' {
+				ctx.lineLength = 0
+				if ctx.prefix != "" {
+					if _, err = ctx.buf.WriteString(ctx.prefix); err != nil {
+						return err
+					}
+				}
+			}
+		}
+	}
+	return nil
+}
+
+const maxLineLen = 74
+
+func (ctx *textifyTraverseContext) breakLongLines(data string) []string {
+	// Only break lines when in blockquotes.
+	if ctx.blockquoteLevel == 0 {
+		return []string{data}
+	}
+	var (
+		ret      = []string{}
+		runes    = []rune(data)
+		l        = len(runes)
+		existing = ctx.lineLength
+	)
+	if existing >= maxLineLen {
+		ret = append(ret, "\n")
+		existing = 0
+	}
+	for l+existing > maxLineLen {
+		i := maxLineLen - existing
+		for i >= 0 && !unicode.IsSpace(runes[i]) {
+			i--
+		}
+		if i == -1 {
+			// No spaces, so go the other way.
+			i = maxLineLen - existing
+			for i < l && !unicode.IsSpace(runes[i]) {
+				i++
+			}
+		}
+		ret = append(ret, string(runes[:i])+"\n")
+		for i < l && unicode.IsSpace(runes[i]) {
+			i++
+		}
+		runes = runes[i:]
+		l = len(runes)
+		existing = 0
+	}
+	if len(runes) > 0 {
+		ret = append(ret, string(runes))
+	}
+	return ret
+}
+
+func (ctx *textifyTraverseContext) normalizeHrefLink(link string) string {
+	link = strings.TrimSpace(link)
+	link = strings.TrimPrefix(link, "mailto:")
+	return link
+}
+
+// renderEachChild visits each direct child of a node and collects the sequence of
+// textuual representaitons separated by a single newline.
+func (ctx *textifyTraverseContext) renderEachChild(node *html.Node) (string, error) {
+	buf := &bytes.Buffer{}
+	for c := node.FirstChild; c != nil; c = c.NextSibling {
+		s, err := FromHTMLNode(c, ctx.options)
+		if err != nil {
+			return "", err
+		}
+		if _, err = buf.WriteString(s); err != nil {
+			return "", err
+		}
+		if c.NextSibling != nil {
+			if err = buf.WriteByte('\n'); err != nil {
+				return "", err
+			}
+		}
+	}
+	return buf.String(), nil
+}
+
+func getAttrVal(node *html.Node, attrName string) string {
+	for _, attr := range node.Attr {
+		if attr.Key == attrName {
+			return attr.Val
+		}
+	}
+
+	return ""
+}
diff --git a/backend/vendor/github.com/matcornic/hermes/.gitignore b/backend/vendor/github.com/matcornic/hermes/.gitignore
new file mode 100644
index 00000000..ea31aff7
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/.gitignore
@@ -0,0 +1,28 @@
+# Compiled Object files, Static and Dynamic libs (Shared Objects)
+*.o
+*.a
+*.so
+
+# Folders
+_obj
+_test
+
+# Architecture specific extensions/prefixes
+*.[568vq]
+[568vq].out
+
+*.cgo1.go
+*.cgo2.c
+_cgo_defun.c
+_cgo_gotypes.go
+_cgo_export.*
+
+_testmain.go
+
+*.exe
+*.test
+*.prof
+
+.DS_Store
+coverage.txt
+vendor/
diff --git a/backend/vendor/github.com/matcornic/hermes/.travis.yml b/backend/vendor/github.com/matcornic/hermes/.travis.yml
new file mode 100644
index 00000000..e0055bda
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/.travis.yml
@@ -0,0 +1,20 @@
+language: go
+
+go:
+  - 1.6
+  - 1.7
+  - 1.8
+  - 1.9
+
+install:
+- curl -L https://github.com/golang/dep/releases/download/v0.3.2/dep-linux-amd64 -o dep
+- chmod u+x dep && ./dep ensure
+- go get -u github.com/alecthomas/gometalinter
+- gometalinter -u -i -f
+
+script:
+- go test -race -coverprofile=coverage.txt -covermode=atomic
+- gometalinter --vendor --config=./gometalinter.json ./...
+
+after_success:
+- bash <(curl -s https://codecov.io/bash)
diff --git a/backend/vendor/github.com/matcornic/hermes/CONTRIBUTING.md b/backend/vendor/github.com/matcornic/hermes/CONTRIBUTING.md
new file mode 100644
index 00000000..82d7883e
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/CONTRIBUTING.md
@@ -0,0 +1,347 @@
+# Contributing
+
+Thanks so much for wanting to help! We really appreciate it.
+
+* Have an idea for a new feature?
+* Want to add a new built-in theme?
+
+Excellent! You've come to the right place.
+
+1. If you find a bug or wish to suggest a new feature, please create an issue first
+2. Make sure your code & comment conventions are in-line with the project's style (execute gometalinter as in [.travis.yml](.travis.yml) file)
+3. Make your commits and PRs as tiny as possible - one feature or bugfix at a time
+4. Write detailed commit messages, in-line with the project's commit naming conventions
+
+## Theming Instructions
+
+This file contains instructions on adding themes to Hermes:
+
+* [Using a Custom Theme](#using-a-custom-theme)
+* [Creating a Built-In Theme](#creating-a-built-in-theme)
+
+> We use Golang templates under the hood to inject the e-mail body into themes.
+> - [Official guide](https://golang.org/pkg/text/template/)
+> - [Tutorial](https://astaxie.gitbooks.io/build-web-application-with-golang/en/07.4.html)
+> - [Hugo guide](https://gohugo.io/templates/go-templates/)
+
+### Using a Custom Theme
+
+If you want to supply your own **custom theme** for Hermes to use (but don't want it included with Hermes):
+
+1. Create a new struct implementing `Theme` interface ([hermes.go](hermes.go)). A real-life example is in [default.go](default.go)
+2. Supply your new theme at hermes creation
+
+```go
+
+type MyCustomTheme struct{}
+
+func (dt *MyCustomTheme) Name() string {
+	return "mycustomthem"
+}
+
+func (dt *MyCustomTheme) HTMLTemplate() string {
+    // Get the template from a file (if you want to be able to change the template live without retstarting your application)
+    // Or write the template by returning pure string here (if you want embbeded template and do not bother with external dependencies)
+    return "<A go html template with wanted information>" 
+}
+
+func (dt *MyCustomTheme) PlainTextTemplate() string {
+    // Get the template from a file (if you want to be able to change the template live without retstarting your application)
+    // Or write the template by returning pure string here (if you want embbeded template and do not bother with external dependencies)
+    return "<A go plaintext template with wanter information>"
+}
+
+h := hermes.Hermes{
+    Theme: new(MyCustomTheme) // Set your fresh new theme here
+    Product: hermes.Product{
+        Name: "Hermes",
+        Link: "https://example-hermes.com/",
+    },
+}
+
+// ...
+// Continue with the rest as usual, create your email and generate the content.
+// ...
+```
+
+3. That's it.
+
+### Creating a Built-In Theme
+
+If you want to create a new **built-in** Hermes theme:
+
+1. Fork the repository to your GitHub account and clone it to your computer
+2. Create a new Go file named after your new theme
+3. Copy content of [default.go](default.go) file in new file and make any necessary changes
+4. Scroll down to the [injection snippets](#injection-snippets) and copy and paste each code snippet into the relevant area of your template markup
+5. Test the theme by adding the theme to slice of tested themes (see [hermes_test.go](hermes_test.go)). A set of tests will be run to check that your theme follows features of Hermes.
+6. Create examples in new folder for your theme in `examples` folder and run `go run *.go`. It will generate the different `html` and `plaintext` emails for your different examples. Follow the same examples as default theme (3 examples: Welcome, Reset and Receipt)
+7. Add the theme name, credit, and screenshots to the `README.md` file's [Supported Themes](README.md#supported-themes) section (copy one of the existing themes' markup and modify it accordingly)
+8. Submit a pull request with your changes and we'll let you know if anything's missing!
+
+Thanks again for your contribution!
+
+# Injection Snippets
+
+## Product Branding Injection
+
+The following will inject either the product logo or name into the template.
+
+```html
+<a href="{{.Hermes.Product.Link}}" target="_blank">
+        {{ if .Hermes.Product.Logo }}
+            <img src="{{.Hermes.Product.Logo}}" class="email-logo" />
+        {{ else }}
+            {{ .Hermes.Product.Name }}
+        {{ end }}
+</a>
+```
+
+It's a good idea to add the following CSS declaration to set `max-height: 50px` for the logo:
+
+```css
+.email-logo {
+    max-height: 50px;
+}
+```
+
+## Title Injection
+
+The following will inject the e-mail title (Hi John Appleseed,) or a custom title provided by the user:
+
+```html
+<h1>{{if .Email.Body.Title }}{{ .Email.Body.Title }}{{ else }}{{ .Email.Body.Greeting }} {{ .Email.Body.Name }},{{ end }}</h1>
+```
+
+## Intro Injection
+
+The following will inject the intro text (string or array) into the e-mail:
+
+```html
+{{ with .Email.Body.Intros }}
+    {{ if gt (len .) 0 }}
+        {{ range $line := . }}
+            <p>{{ $line }}</p>
+        {{ end }}
+    {{ end }}
+{{ end }}
+```
+
+## Dictionary Injection
+
+The following will inject a `<dl>` of key-value pairs into the e-mail:
+
+```html
+{{ with .Email.Body.Dictionary }} 
+    {{ if gt (len .) 0 }}
+        <dl class="body-dictionary">
+        {{ range $entry := . }}
+            <dt>{{ $entry.Key }}:</dt>
+            <dd>{{ $entry.Value }}</dd>
+        {{ end }}
+        </dl>
+    {{ end }}
+{{ end }}
+```
+
+It's a good idea to add this to the top of the template to improve the styling of the dictionary:
+
+```css
+/* Dictionary */
+.dictionary {
+    width: 100%;
+    overflow: hidden;
+    margin: 0 auto;
+    padding: 0;
+}
+.dictionary dt {
+    clear: both;
+    color: #000;
+    font-weight: bold;
+    margin-right: 4px;
+}
+.dictionary dd {
+    margin: 0 0 10px 0;
+}
+```
+
+## Table Injection
+
+The following will inject the table into the e-mail:
+
+```html
+<!-- Table -->
+{{ with .Email.Body.Table }}
+{{ $data := .Data }}
+{{ $columns := .Columns }}
+    {{ if gt (len $data) 0 }}
+    <table class="data-wrapper" width="100%" cellpadding="0" cellspacing="0">
+        <tr>
+        <td colspan="2">
+            <table class="data-table" width="100%" cellpadding="0" cellspacing="0">
+            <tr>
+                {{ $col := index $data 0 }}
+                {{ range $entry := $col }}
+                <th
+                    {{ with $columns }}
+                        {{ $width := index .CustomWidth $entry.Key }}
+                        {{ with $width }}
+                        width="{{ . }}"
+                        {{ end }}
+                        {{ $align := index .CustomAlignment $entry.Key }}
+                        {{ with $align }}
+                        style="text-align:{{ . }}"
+                        {{ end }}
+                    {{ end }}
+                >
+                    <p>{{ $entry.Key }}</p>
+                </th>
+                {{ end }}
+            </tr>
+            {{ range $row := $data }}
+            <tr>
+                {{ range $cell := $row }}
+                <td
+                    {{ with $columns }}
+                        {{ $align := index .CustomAlignment $cell.Key }}
+                        {{ with $align }}
+                        style="text-align:{{ . }}"
+                        {{ end }}
+                    {{ end }}
+                >
+                {{ $cell.Value }}
+                </td>
+                {{ end }}
+            </tr>
+            {{ end }}
+            </table>
+        </td>
+        </tr>
+    </table>
+    {{ end }}
+{{ end }}
+```
+
+It's a good idea to add this to the top of the template to improve the styling of the table:
+
+```css
+/* Table */
+.data-wrapper {
+    width: 100%;
+    margin: 0;
+    padding: 35px 0;
+}
+.data-table {
+    width: 100%;
+    margin: 0;
+}
+.data-table th {
+    text-align: left;
+    padding: 0px 5px;
+    padding-bottom: 8px;
+    border-bottom: 1px solid #DEDEDE;
+}
+.data-table th p {
+    margin: 0;
+    font-size: 12px;
+}
+.data-table td {
+    text-align: left;
+    padding: 10px 5px;
+    font-size: 15px;
+    line-height: 18px;
+}
+```
+
+## Action Injection
+
+The following will inject the action link (or button) into the e-mail:
+
+```html
+{{ with .Email.Body.Actions }}
+    {{ if gt (len .) 0 }}
+        {{ range $action := . }}
+        <p>{{ $action.Instructions }}</p>
+        <table class="body-action" align="center" width="100%" cellpadding="0" cellspacing="0">
+            <tr>
+                <td align="center">
+                <div>
+                    <a href="{{ $action.Button.Link }}" class="button" style="background-color: {{ $action.Button.Color }}" target="_blank">{{ $action.Button.Text }}</a>
+                </div>
+                </td>
+            </tr>
+        </table>
+        {{ end }}
+    {{ end }}
+{{ end }}
+```
+
+A good practice is to describe action in footer in case of problem when displaying button and CSS. The text for the description is provided through the `TroubleText` field of the `Product` struct. The text may contain a placeholder `{ACTION}` which is expected to be replaced with the text of the button. The default value of `TroubleText` is `If you’re having trouble with the button '{ACTION}', copy and paste the URL below into your web browser.`
+
+```html
+{{ with .Email.Body.Actions }}
+<table class="body-sub">
+    <tbody><tr>
+    {{ range $action := . }}
+    <td>
+        <p class="sub">{{$.Hermes.Product.TroubleText | replace "{ACTION}" $action.Button.Text}}</p>
+        <p class="sub"><a href="{{ $action.Button.Link }}">{{ $action.Button.Link }}</a></p>
+    </td>
+    {{ end }}
+    </tr>
+    </tbody>
+</table>
+{{ end }}
+```
+
+## Outro Injection
+
+The following will inject the outro text (string or array) into the e-mail:
+
+```html
+{{ with .Email.Body.Outros }} 
+    {{ if gt (len .) 0 }}
+        {{ range $line := . }}
+        <p>{{ $line }}</p>
+        {{ end }}
+    {{ end }}
+{{ end }}
+```
+
+## Signature Injection
+
+The following will inject the signature phrase (e.g. Yours truly) along with the product name into the e-mail:
+
+```html
+{{.Email.Body.Signature}},
+<br>
+{{.Hermes.Product.Name}}
+```
+
+## Copyright Injection
+
+The following will inject the copyright notice into the e-mail:
+
+```html
+{{.Hermes.Product.Copyright}}
+```
+
+## Text Direction Injection
+
+In order to support generating RTL e-mails, inject the `textDirection` variable into the `<body>` tag:
+
+```html
+<body dir="{{.Hermes.TextDirection}}">
+```
+
+## FreeMarkdown Injection
+
+In order to support Markdown free content, inject the following code:
+
+````html
+{{ if (ne .Email.Body.FreeMarkdown "") }}
+    {{ .Email.Body.FreeMarkdown.ToHTML }}
+{{ else }}
+    [... Here is the templating for dictionary, table and actions]
+{{ end }}
+```
diff --git a/backend/vendor/github.com/matcornic/hermes/Gopkg.toml b/backend/vendor/github.com/matcornic/hermes/Gopkg.toml
new file mode 100644
index 00000000..a7ccc7b4
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/Gopkg.toml
@@ -0,0 +1,42 @@
+
+# Gopkg.toml example
+#
+# Refer to https://github.com/golang/dep/blob/master/docs/Gopkg.toml.md
+# for detailed Gopkg.toml documentation.
+#
+# required = ["github.com/user/thing/cmd/thing"]
+# ignored = ["github.com/user/project/pkgX", "bitbucket.org/user/project/pkgA/pkgY"]
+#
+# [[constraint]]
+#   name = "github.com/user/project"
+#   version = "1.0.0"
+#
+# [[constraint]]
+#   name = "github.com/user/project2"
+#   branch = "dev"
+#   source = "github.com/myfork/project2"
+#
+# [[override]]
+#  name = "github.com/x/y"
+#  version = "2.4.0"
+
+
+[[constraint]]
+  name = "github.com/Masterminds/sprig"
+  version = "2.14.1"
+
+[[constraint]]
+  name = "github.com/imdario/mergo"
+  version = "0.2.4"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/jaytaylor/html2text"
+
+[[constraint]]
+  name = "github.com/stretchr/testify"
+  version = "1.1.4"
+
+[[constraint]]
+  name = "gopkg.in/russross/blackfriday.v2"
+  version = "2.0.0"
diff --git a/backend/vendor/github.com/matcornic/hermes/LICENSE b/backend/vendor/github.com/matcornic/hermes/LICENSE
new file mode 100644
index 00000000..12572726
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/LICENSE
@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "{}"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2017 Hermes - Mathieu Cornic
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
diff --git a/backend/vendor/github.com/matcornic/hermes/README.md b/backend/vendor/github.com/matcornic/hermes/README.md
new file mode 100644
index 00000000..4af40ea7
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/README.md
@@ -0,0 +1,373 @@
+# Hermes
+
+[![Build Status](https://travis-ci.org/matcornic/hermes.svg?branch=master)](https://travis-ci.org/matcornic/hermes)
+[![Go Report Card](https://goreportcard.com/badge/github.com/matcornic/hermes)](https://goreportcard.com/report/github.com/matcornic/hermes)
+[![Go Coverage](https://codecov.io/github/matcornic/hermes/coverage.svg)](https://codecov.io/github/matcornic/hermes/)
+[![Godoc](https://godoc.org/github.com/matcornic/hermes?status.svg)](https://godoc.org/github.com/matcornic/hermes)
+
+Hermes is the Go port of the great [mailgen](https://github.com/eladnava/mailgen) engine for Node.js. Check their work, it's awesome!
+It's a package that generates clean, responsive HTML e-mails for sending transactional e-mails (welcome e-mails, reset password e-mails, receipt e-mails and so on), and associated plain text fallback.
+
+# Demo
+
+<img src="screens/default/welcome.png" height="400" /> <img src="screens/default/reset.png" height="400" /> <img src="screens/default/receipt.png" height="400" />
+
+# Usage
+
+First install the package:
+
+```
+go get -u github.com/matcornic/hermes
+```
+
+Then, start using the package by importing and configuring it:
+
+```go
+// Configure hermes by setting a theme and your product info
+h := hermes.Hermes{
+    // Optional Theme
+    // Theme: new(Default) 
+    Product: hermes.Product{
+        // Appears in header & footer of e-mails
+        Name: "Hermes",
+        Link: "https://example-hermes.com/",
+        // Optional product logo
+        Logo: "http://www.duchess-france.org/wp-content/uploads/2016/01/gopher.png",
+    },
+}
+```
+
+Next, generate an e-mail using the following code:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        Name: "Jon Snow",
+        Intros: []string{
+            "Welcome to Hermes! We're very excited to have you on board.",
+        },
+        Actions: []hermes.Action{
+            {
+                Instructions: "To get started with Hermes, please click here:",
+                Button: hermes.Button{
+                    Color: "#22BC66", // Optional action button color
+                    Text:  "Confirm your account",
+                    Link:  "https://hermes-example.com/confirm?token=d9729feb74992cc3482b350163a1a010",
+                },
+            },
+        },
+        Outros: []string{
+            "Need help, or have questions? Just reply to this email, we'd love to help.",
+        },
+    },
+}
+
+// Generate an HTML email with the provided contents (for modern clients)
+emailBody, err := h.GenerateHTML(email)
+if err != nil {
+    panic(err) // Tip: Handle error with something else than a panic ;)
+}
+
+// Generate the plaintext version of the e-mail (for clients that do not support xHTML)
+emailText, err := h.GeneratePlainText(email)
+if err != nil {
+    panic(err) // Tip: Handle error with something else than a panic ;)
+}
+
+// Optionally, preview the generated HTML e-mail by writing it to a local file
+err = ioutil.WriteFile("preview.html", []byte(emailBody), 0644)
+if err != nil {
+    panic(err) // Tip: Handle error with something else than a panic ;)
+}
+```
+
+This code would output the following HTML template:
+
+<img src="screens/demo.png" height="400" />
+
+And the following plain text:
+
+```
+------------
+Hi Jon Snow,
+------------
+
+Welcome to Hermes! We're very excited to have you on board.
+
+To get started with Hermes, please click here: https://hermes-example.com/confirm?token=d9729feb74992cc3482b350163a1a010
+
+Need help, or have questions? Just reply to this email, we'd love to help.
+
+Yours truly,
+Hermes - https://example-hermes.com/
+
+Copyright © 2017 Hermes. All rights reserved.
+```
+
+> Theme templates will be embedded in your application binary. If you want to use external templates (for configuration), use your own theme by implementing `hermes.Theme` interface with code searching for your files.
+
+## More Examples
+
+* [Welcome](examples/welcome.go)
+* [Receipt](examples/receipt.go)
+* [Password Reset](examples/reset.go)
+* [Maintenance](examples/maintenance.go)
+
+To run the examples, go to `examples` folder, then run `go run *.go`. HTML and Plaintext example should be created in given theme folders.
+
+## Plaintext E-mails
+
+To generate a [plaintext version of the e-mail](https://litmus.com/blog/best-practices-for-plain-text-emails-a-look-at-why-theyre-important), simply call `GeneratePlainText` function:
+
+```go
+// Generate plaintext email using hermes
+emailText, err := h.GeneratePlainText(email)
+if err != nil {
+    panic(err) // Tip: Handle error with something else than a panic ;)
+}
+```
+
+## Supported Themes
+
+The following open-source themes are bundled with this package:
+
+* `default` by [Postmark Transactional Email Templates](https://github.com/wildbit/postmark-templates)
+
+<img src="screens/default/welcome.png" height="200" /> <img src="screens/default/reset.png" height="200" /> <img src="screens/default/receipt.png" height="200" />
+
+* `flat`, slightly modified from [Postmark Transactional Email Templates](https://github.com/wildbit/postmark-templates)
+
+<img src="screens/flat/welcome.png" height="200" /> <img src="screens/flat/reset.png" height="200" /> <img src="screens/flat/receipt.png" height="200" />
+
+## RTL Support
+
+To change the default text direction (left-to-right), simply override it as follows:
+
+```go
+// Configure hermes by setting a theme and your product info
+h := hermes.Hermes {
+    // Custom text direction
+    TextDirection: hermes.TDRightToLeft,
+}
+```
+
+## Language Customizations
+
+To customize the e-mail's greeting ("Hi") or signature ("Yours truly"), supply custom strings within the e-mail's `Body`:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        Greeting: "Dear",
+        Signature: "Sincerely",
+    },
+}
+```
+
+To use a custom title string rather than a greeting/name introduction, provide it instead of `Name`:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        // Title will override `Name`
+        Title: "Welcome to Hermes",
+    },
+}
+```
+
+To customize the `Copyright`, override it when initializing `Hermes` within your `Product` as follows:
+
+```go
+// Configure hermes by setting a theme and your product info
+h := hermes.Hermes{
+    // Optional Theme
+    // Theme: new(Default)
+    Product: hermes.Product{
+        // Appears in header & footer of e-mails
+        Name: "Hermes",
+        Link: "https://example-hermes.com/",
+        // Custom copyright notice
+        Copyright: "Copyright © 2017 Dharma Initiative. All rights reserved."
+    },
+}
+```
+
+To use a custom fallback text at the end of the email, change the `TroubleText` field of the `hermes.Product` struct. The default value is `If you’re having trouble with the button '{ACTION}', copy and paste the URL below into your web browser.`. The `{ACTION}` placeholder will be replaced with the corresponding text of the supplied action button:
+
+```go
+// Configure hermes by setting a theme and your product info
+h := hermes.Hermes{
+    // Optional Theme
+    // Theme: new(Default)
+    Product: hermes.Product{
+        // Custom trouble text
+        TroubleText: "If the {ACTION}-button is not working for you, just copy and paste the URL below into your web browser."
+    },
+}
+```
+
+## Elements
+
+Hermes supports injecting custom elements such as dictionaries, tables and action buttons into e-mails.
+
+### Action
+
+To inject an action button in to the e-mail, supply the `Actions` object as follows:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        Actions: []hermes.Action{
+            {
+                Instructions: "To get started with Hermes, please click here:",
+                Button: hermes.Button{
+                    Color: "#22BC66", // Optional action button color
+                    Text:  "Confirm your account",
+                    Link:  "https://hermes-example.com/confirm?token=d9729feb74992cc3482b350163a1a010",
+                },
+            },
+        },
+    },
+}
+```
+
+To inject multiple action buttons in to the e-mail, supply another struct in Actions slice `Action`.
+
+### Table
+
+To inject a table into the e-mail, supply the `Table` object as follows:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        Table: hermes.Table{
+            Data: [][]hermes.Entry{
+                // List of rows
+                {   
+                    // Key is the column name, Value is the cell value
+                    // First object defines what columns will be displayed
+                    {Key: "Item", Value: "Golang"},
+                    {Key: "Description", Value: "Open source programming language that makes it easy to build simple, reliable, and efficient software"},
+                    {Key: "Price", Value: "$10.99"},
+                },
+                {
+                    {Key: "Item", Value: "Hermes"},
+                    {Key: "Description", Value: "Programmatically create beautiful e-mails using Golang."},
+                    {Key: "Price", Value: "$1.99"},
+                },
+            },
+            Columns: hermes.Columns{
+                // Custom style for each rows
+                CustomWidth: map[string]string{
+                    "Item":  "20%",
+                    "Price": "15%",
+                },
+                CustomAlignment: map[string]string{
+                    "Price": "right",
+                },
+            },
+        },
+    },
+}
+```
+
+### Dictionary
+
+To inject key-value pairs of data into the e-mail, supply the `Dictionary` object as follows:
+
+```go
+email := hermes.Email{
+    Body: hermes.Body{
+        Dictionary: []hermes.Entry{
+            {Key: "Date", Value: "20 November 1887"},
+            {Key: "Address", Value: "221B Baker Street, London"},
+        },
+    },
+}
+```
+
+### Free Markdown
+
+If you need more flexibility in the content of your generated e-mail, while keeping the same format than any other e-mail, use Markdown content. Supply the `FreeMarkdown` object as follows:
+
+```go
+email := hermes.Email{
+		Body: hermes.Body{
+			FreeMarkdown: `
+> _Hermes_ service will shutdown the **1st August 2017** for maintenance operations. 
+
+Services will be unavailable based on the following schedule:
+
+| Services | Downtime |
+| :------:| :-----------: |
+| Service A | 2AM to 3AM |
+| Service B | 4AM to 5AM |
+| Service C | 5AM to 6AM |
+
+---
+
+Feel free to contact us for any question regarding this matter at [support@hermes-example.com](mailto:support@hermes-example.com) or in our [Gitter](https://gitter.im/)
+
+`,
+		},
+	}
+}
+```
+
+This code would output the following HTML template:
+
+<img src="screens/free-markdown.png" height="400" />
+
+And the following plaintext:
+
+```
+------------
+Hi Jon Snow,
+------------
+
+> 
+> 
+> 
+> Hermes service will shutdown the *1st August 2017* for maintenance
+> operations.
+> 
+> 
+
+Services will be unavailable based on the following schedule:
+
++-----------+------------+
+| SERVICES  |  DOWNTIME  |
++-----------+------------+
+| Service A | 2AM to 3AM |
+| Service B | 4AM to 5AM |
+| Service C | 5AM to 6AM |
++-----------+------------+
+
+Feel free to contact us for any question regarding this matter at support@hermes-example.com ( support@hermes-example.com ) or in our Gitter ( https://gitter.im/ )
+
+Yours truly,
+Hermes - https://example-hermes.com/
+
+Copyright © 2017 Hermes. All rights reserved.
+```
+
+Be aware that this content will replace existing tables, dictionary and actions. Only intros, outros, header and footer will be kept.
+
+This is helpful when your application needs sending e-mails, wrote on-the-fly by adminstrators.
+
+> Markdown is rendered with [Blackfriday](https://github.com/russross/blackfriday), so every thing Blackfriday can do, Hermes can do it as well.
+
+## Troubleshooting
+
+1. After sending multiple e-mails to the same Gmail / Inbox address, they become grouped and truncated since they contain similar text, breaking the responsive e-mail layout.
+
+> Simply sending the `X-Entity-Ref-ID` header with your e-mails will prevent grouping / truncation.
+
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md)
+
+## License
+
+Apache 2.0
+
diff --git a/backend/vendor/github.com/matcornic/hermes/default.go b/backend/vendor/github.com/matcornic/hermes/default.go
new file mode 100644
index 00000000..b8e2e941
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/default.go
@@ -0,0 +1,495 @@
+package hermes
+
+// Default is the theme by default
+type Default struct{}
+
+// Name returns the name of the default theme
+func (dt *Default) Name() string {
+	return "default"
+}
+
+// HTMLTemplate returns a Golang template that will generate an HTML email.
+func (dt *Default) HTMLTemplate() string {
+	return `
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
+  <style type="text/css" rel="stylesheet" media="all">
+    /* Base ------------------------------ */
+    *:not(br):not(tr):not(html) {
+      font-family: Arial, 'Helvetica Neue', Helvetica, sans-serif;
+      -webkit-box-sizing: border-box;
+      box-sizing: border-box;
+    }
+    body {
+      width: 100% !important;
+      height: 100%;
+      margin: 0;
+      line-height: 1.4;
+      background-color: #F2F4F6;
+      color: #74787E;
+      -webkit-text-size-adjust: none;
+    }
+    a {
+      color: #3869D4;
+    }
+    /* Layout ------------------------------ */
+    .email-wrapper {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+      background-color: #F2F4F6;
+    }
+    .email-content {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+    }
+    /* Masthead ----------------------- */
+    .email-masthead {
+      padding: 25px 0;
+      text-align: center;
+    }
+    .email-masthead_logo {
+      max-width: 400px;
+      border: 0;
+    }
+    .email-masthead_name {
+      font-size: 16px;
+      font-weight: bold;
+      color: #2F3133;
+      text-decoration: none;
+      text-shadow: 0 1px 0 white;
+    }
+    .email-logo {
+      max-height: 50px;
+    }
+    /* Body ------------------------------ */
+    .email-body {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+      border-top: 1px solid #EDEFF2;
+      border-bottom: 1px solid #EDEFF2;
+      background-color: #FFF;
+    }
+    .email-body_inner {
+      width: 570px;
+      margin: 0 auto;
+      padding: 0;
+    }
+    .email-footer {
+      width: 570px;
+      margin: 0 auto;
+      padding: 0;
+      text-align: center;
+    }
+    .email-footer p {
+      color: #AEAEAE;
+    }
+    .body-action {
+      width: 100%;
+      margin: 30px auto;
+      padding: 0;
+      text-align: center;
+    }
+    .body-dictionary {
+      width: 100%;
+      overflow: hidden;
+      margin: 20px auto 10px;
+      padding: 0;
+    }
+    .body-dictionary dd {
+      margin: 0 0 10px 0;
+    }
+    .body-dictionary dt {
+      clear: both;
+      color: #000;
+      font-weight: bold;
+    }
+    .body-dictionary dd {
+      margin-left: 0;
+      margin-bottom: 10px;
+    }
+    .body-sub {
+      margin-top: 25px;
+      padding-top: 25px;
+      border-top: 1px solid #EDEFF2;
+      table-layout: fixed;
+    }
+    .body-sub a {
+      word-break: break-all;
+    }
+    .content-cell {
+      padding: 35px;
+    }
+    .align-right {
+      text-align: right;
+    }
+    /* Type ------------------------------ */
+    h1 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 19px;
+      font-weight: bold;
+    }
+    h2 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 16px;
+      font-weight: bold;
+    }
+    h3 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 14px;
+      font-weight: bold;
+    }
+    blockquote {
+      margin: 1.7rem 0;
+      padding-left: 0.85rem;
+      border-left: 10px solid #F0F2F4;
+    }
+    blockquote p {
+        font-size: 1.1rem;
+        color: #999;
+    }
+    blockquote cite {
+        display: block;
+        text-align: right;
+        color: #666;
+        font-size: 1.2rem;
+    }
+    cite {
+      display: block;
+      font-size: 0.925rem; 
+    }
+    cite:before {
+      content: "\2014 \0020";
+    }
+    p {
+      margin-top: 0;
+      color: #74787E;
+      font-size: 16px;
+      line-height: 1.5em;
+    }
+    p.sub {
+      font-size: 12px;
+    }
+    p.center {
+      text-align: center;
+    }
+    table {
+      width: 100%;
+    }
+    th {
+      padding: 0px 5px;
+      padding-bottom: 8px;
+      border-bottom: 1px solid #EDEFF2;
+    }
+    th p {
+      margin: 0;
+      color: #9BA2AB;
+      font-size: 12px;
+    }
+    td {
+      padding: 10px 5px;
+      color: #74787E;
+      font-size: 15px;
+      line-height: 18px;
+    }
+    .content {
+      align: center;
+      padding: 0;
+    }
+    /* Data table ------------------------------ */
+    .data-wrapper {
+      width: 100%;
+      margin: 0;
+      padding: 35px 0;
+    }
+    .data-table {
+      width: 100%;
+      margin: 0;
+    }
+    .data-table th {
+      text-align: left;
+      padding: 0px 5px;
+      padding-bottom: 8px;
+      border-bottom: 1px solid #EDEFF2;
+    }
+    .data-table th p {
+      margin: 0;
+      color: #9BA2AB;
+      font-size: 12px;
+    }
+    .data-table td {
+      padding: 10px 5px;
+      color: #74787E;
+      font-size: 15px;
+      line-height: 18px;
+    }
+    /* Buttons ------------------------------ */
+    .button {
+      display: inline-block;
+      width: 200px;
+      background-color: #3869D4;
+      border-radius: 3px;
+      color: #ffffff;
+      font-size: 15px;
+      line-height: 45px;
+      text-align: center;
+      text-decoration: none;
+      -webkit-text-size-adjust: none;
+      mso-hide: all;
+    }
+    /*Media Queries ------------------------------ */
+    @media only screen and (max-width: 600px) {
+      .email-body_inner,
+      .email-footer {
+        width: 100% !important;
+      }
+    }
+    @media only screen and (max-width: 500px) {
+      .button {
+        width: 100% !important;
+      }
+    }
+  </style>
+</head>
+<body dir="{{.Hermes.TextDirection}}">
+  <table class="email-wrapper" width="100%" cellpadding="0" cellspacing="0">
+    <tr>
+      <td class="content">
+        <table class="email-content" width="100%" cellpadding="0" cellspacing="0">
+          <!-- Logo -->
+          <tr>
+            <td class="email-masthead">
+              <a class="email-masthead_name" href="{{.Hermes.Product.Link}}" target="_blank">
+                {{ if .Hermes.Product.Logo }}
+                  <img src="{{.Hermes.Product.Logo | url }}" class="email-logo" />
+                {{ else }}
+                  {{ .Hermes.Product.Name }}
+                {{ end }}
+                </a>
+            </td>
+          </tr>
+
+          <!-- Email Body -->
+          <tr>
+            <td class="email-body" width="100%">
+              <table class="email-body_inner" align="center" width="570" cellpadding="0" cellspacing="0">
+                <!-- Body content -->
+                <tr>
+                  <td class="content-cell">
+                    <h1>{{if .Email.Body.Title }}{{ .Email.Body.Title }}{{ else }}{{ .Email.Body.Greeting }} {{ .Email.Body.Name }},{{ end }}</h1>
+                    {{ with .Email.Body.Intros }}
+                        {{ if gt (len .) 0 }}
+                          {{ range $line := . }}
+                            <p>{{ $line }}</p>
+                          {{ end }}
+                        {{ end }}
+                    {{ end }}
+                    {{ if (ne .Email.Body.FreeMarkdown "") }}
+                      {{ .Email.Body.FreeMarkdown.ToHTML }}
+                    {{ else }}
+
+                      {{ with .Email.Body.Dictionary }} 
+                        {{ if gt (len .) 0 }}
+                          <dl class="body-dictionary">
+                            {{ range $entry := . }}
+                              <dt>{{ $entry.Key }}:</dt>
+                              <dd>{{ $entry.Value }}</dd>
+                            {{ end }}
+                          </dl>
+                        {{ end }}
+                      {{ end }}
+
+                      <!-- Table -->
+                      {{ with .Email.Body.Table }}
+                        {{ $data := .Data }}
+                        {{ $columns := .Columns }}
+                        {{ if gt (len $data) 0 }}
+                          <table class="data-wrapper" width="100%" cellpadding="0" cellspacing="0">
+                            <tr>
+                              <td colspan="2">
+                                <table class="data-table" width="100%" cellpadding="0" cellspacing="0">
+                                  <tr>
+                                    {{ $col := index $data 0 }}
+                                    {{ range $entry := $col }}
+                                      <th
+                                        {{ with $columns }}
+                                          {{ $width := index .CustomWidth $entry.Key }}
+                                          {{ with $width }}
+                                            width="{{ . }}"
+                                          {{ end }}
+                                          {{ $align := index .CustomAlignment $entry.Key }}
+                                          {{ with $align }}
+                                            style="text-align:{{ . }}"
+                                          {{ end }}
+                                        {{ end }}
+                                      >
+                                        <p>{{ $entry.Key }}</p>
+                                      </th>
+                                    {{ end }}
+                                  </tr>
+                                  {{ range $row := $data }}
+                                    <tr>
+                                      {{ range $cell := $row }}
+                                        <td
+                                          {{ with $columns }}
+                                            {{ $align := index .CustomAlignment $cell.Key }}
+                                            {{ with $align }}
+                                              style="text-align:{{ . }}"
+                                            {{ end }}
+                                          {{ end }}
+                                        >
+                                          {{ $cell.Value }}
+                                        </td>
+                                      {{ end }}
+                                    </tr>
+                                  {{ end }}
+                                </table>
+                              </td>
+                            </tr>
+                          </table>
+                        {{ end }}
+                      {{ end }}
+
+                      <!-- Action -->
+                      {{ with .Email.Body.Actions }}
+                        {{ if gt (len .) 0 }}
+                          {{ range $action := . }}
+                            <p>{{ $action.Instructions }}</p>
+                            <table class="body-action" align="center" width="100%" cellpadding="0" cellspacing="0">
+                              <tr>
+                                <td align="center">
+                                  <div>
+                                    <a href="{{ $action.Button.Link }}" class="button" style="background-color: {{ $action.Button.Color }}; color: {{ $action.Button.TextColor }};" target="_blank">
+                                      {{ $action.Button.Text }}
+                                    </a>
+                                  </div>
+                                </td>
+                              </tr>
+                            </table>
+                          {{ end }}
+                        {{ end }}
+                      {{ end }}
+
+                    {{ end }}
+                    {{ with .Email.Body.Outros }} 
+                        {{ if gt (len .) 0 }}
+                          {{ range $line := . }}
+                            <p>{{ $line }}</p>
+                          {{ end }}
+                        {{ end }}
+                      {{ end }}
+
+                    <p>
+                      {{.Email.Body.Signature}},
+                      <br />
+                      {{.Hermes.Product.Name}}
+                    </p>
+
+                    {{ if (eq .Email.Body.FreeMarkdown "") }}
+                      {{ with .Email.Body.Actions }} 
+                        <table class="body-sub">
+                          <tbody>
+                              {{ range $action := . }}
+                                <tr>
+                                  <td>
+                                    <p class="sub">{{$.Hermes.Product.TroubleText | replace "{ACTION}" $action.Button.Text}}</p>
+                                    <p class="sub"><a href="{{ $action.Button.Link }}">{{ $action.Button.Link }}</a></p>
+                                  </td>
+                                </tr>
+                              {{ end }}
+                          </tbody>
+                        </table>
+                      {{ end }}
+                    {{ end }}
+                  </td>
+                </tr>
+              </table>
+            </td>
+          </tr>
+          <tr>
+            <td>
+              <table class="email-footer" align="center" width="570" cellpadding="0" cellspacing="0">
+                <tr>
+                  <td class="content-cell">
+                    <p class="sub center">
+                      {{.Hermes.Product.Copyright}}
+                    </p>
+                  </td>
+                </tr>
+              </table>
+            </td>
+          </tr>
+        </table>
+      </td>
+    </tr>
+  </table>
+</body>
+</html>
+`
+}
+
+// PlainTextTemplate returns a Golang template that will generate an plain text email.
+func (dt *Default) PlainTextTemplate() string {
+	return `<h2>{{if .Email.Body.Title }}{{ .Email.Body.Title }}{{ else }}{{ .Email.Body.Greeting }} {{ .Email.Body.Name }},{{ end }}</h2>
+{{ with .Email.Body.Intros }}
+  {{ range $line := . }}
+    <p>{{ $line }}</p>
+  {{ end }}
+{{ end }}
+{{ if (ne .Email.Body.FreeMarkdown "") }}
+  {{ .Email.Body.FreeMarkdown.ToHTML }}
+{{ else }}
+  {{ with .Email.Body.Dictionary }}
+    <ul>
+    {{ range $entry := . }}
+      <li>{{ $entry.Key }}: {{ $entry.Value }}</li>
+    {{ end }}
+    </ul>
+  {{ end }}
+  {{ with .Email.Body.Table }}
+    {{ $data := .Data }}
+    {{ $columns := .Columns }}
+    {{ if gt (len $data) 0 }}
+      <table class="data-table" width="100%" cellpadding="0" cellspacing="0">
+        <tr>
+          {{ $col := index $data 0 }}
+          {{ range $entry := $col }}
+            <th>{{ $entry.Key }} </th>
+          {{ end }}
+        </tr>
+        {{ range $row := $data }}
+          <tr>
+            {{ range $cell := $row }}
+              <td>
+                {{ $cell.Value }}
+              </td>
+            {{ end }}
+          </tr>
+        {{ end }}
+      </table>
+    {{ end }}
+  {{ end }}
+  {{ with .Email.Body.Actions }} 
+    {{ range $action := . }}
+      <p>{{ $action.Instructions }} {{ $action.Button.Link }}</p> 
+    {{ end }}
+  {{ end }}
+{{ end }}
+{{ with .Email.Body.Outros }} 
+  {{ range $line := . }}
+    <p>{{ $line }}<p>
+  {{ end }}
+{{ end }}
+<p>{{.Email.Body.Signature}},<br>{{.Hermes.Product.Name}} - {{.Hermes.Product.Link}}</p>
+
+<p>{{.Hermes.Product.Copyright}}</p>
+`
+}
diff --git a/backend/vendor/github.com/matcornic/hermes/flat.go b/backend/vendor/github.com/matcornic/hermes/flat.go
new file mode 100644
index 00000000..955a5686
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/flat.go
@@ -0,0 +1,495 @@
+package hermes
+
+// Flat is a theme
+type Flat struct{}
+
+// Name returns the name of the flat theme
+func (dt *Flat) Name() string {
+	return "flat"
+}
+
+// HTMLTemplate returns a Golang template that will generate an HTML email.
+func (dt *Flat) HTMLTemplate() string {
+	return `
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
+  <style type="text/css" rel="stylesheet" media="all">
+    /* Base ------------------------------ */
+    *:not(br):not(tr):not(html) {
+      font-family: Arial, 'Helvetica Neue', Helvetica, sans-serif;
+      -webkit-box-sizing: border-box;
+      box-sizing: border-box;
+    }
+    body {
+      width: 100% !important;
+      height: 100%;
+      margin: 0;
+      line-height: 1.4;
+      background-color: #2c3e50;
+      color: #74787E;
+      -webkit-text-size-adjust: none;
+    }
+    a {
+      color: #3869D4;
+    }
+
+    /* Layout ------------------------------ */
+    .email-wrapper {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+      background-color: #2c3e50;
+    }
+    .email-content {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+    }
+    /* Masthead ----------------------- */
+    .email-masthead {
+      padding: 25px 0;
+      text-align: center;
+    }
+    .email-masthead_logo {
+      max-width: 400px;
+      border: 0;
+    }
+    .email-masthead_name {
+      font-size: 16px;
+      font-weight: bold;
+      color: #2F3133;
+      text-decoration: none;
+      text-shadow: 0 1px 0 white;
+    }
+    .email-logo {
+      max-height: 50px;
+    }
+    /* Body ------------------------------ */
+    .email-body {
+      width: 100%;
+      margin: 0;
+      padding: 0;
+      border-top: 1px solid #EDEFF2;
+      border-bottom: 1px solid #EDEFF2;
+      background-color: #FFF;
+    }
+    .email-body_inner {
+      width: 570px;
+      margin: 0 auto;
+      padding: 0;
+    }
+    .email-footer {
+      width: 570px;
+      margin: 0 auto;
+      padding: 0;
+      text-align: center;
+    }
+    .email-footer p {
+      color: #eaeaea;
+    }
+    .body-action {
+      width: 100%;
+      margin: 30px auto;
+      padding: 0;
+      text-align: center;
+    }
+
+    .body-dictionary {
+      width: 100%;
+      overflow: hidden;
+      margin: 20px auto 20px;
+      padding: 0;
+    }
+    .body-dictionary dt {
+      clear: both;
+      color: #000;
+      font-weight: bold;
+      float: left;
+      width: 50%;
+      padding: 0;
+      margin: 0;
+      margin-bottom: 0.3em;
+    }
+    .body-dictionary dd {
+      float: left;
+      width: 50%;
+      padding: 0;
+      margin: 0;
+    }
+    .body-sub {
+      margin-top: 25px;
+      padding-top: 25px;
+      border-top: 1px solid #EDEFF2;
+      table-layout: fixed;
+    }
+    .body-sub a {
+      word-break: break-all;
+    }
+    .content-cell {
+      padding: 35px;
+    }
+    .align-right {
+      text-align: right;
+    }
+    /* Type ------------------------------ */
+    h1 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 19px;
+      font-weight: bold;
+    }
+    h2 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 16px;
+      font-weight: bold;
+    }
+    h3 {
+      margin-top: 0;
+      color: #2F3133;
+      font-size: 14px;
+      font-weight: bold;
+    }
+    blockquote {
+      margin: 1.7rem 0;
+      padding-left: 0.85rem;
+      border-left: 10px solid #F0F2F4;
+    }
+    blockquote p {
+        font-size: 1.1rem;
+        color: #999;
+    }
+    blockquote cite {
+        display: block;
+        text-align: right;
+        color: #666;
+        font-size: 1.2rem;
+    }
+    cite {
+      display: block;
+      font-size: 0.925rem; 
+    }
+    cite:before {
+      content: "\2014 \0020";
+    }
+    p {
+      margin-top: 0;
+      color: #74787E;
+      font-size: 16px;
+      line-height: 1.5em;
+    }
+    p.sub {
+      font-size: 12px;
+    }
+    p.center {
+      text-align: center;
+    }
+    table {
+      width: 100%;
+    }
+    th {
+      padding: 0px 5px;
+      padding-bottom: 8px;
+      border-bottom: 1px solid #EDEFF2;
+    }
+    th p {
+      margin: 0;
+      color: #9BA2AB;
+      font-size: 12px;
+    }
+    td {
+      padding: 10px 5px;
+      color: #74787E;
+      font-size: 15px;
+      line-height: 18px;
+    }
+    .content {
+      align: center;
+      padding: 0;
+    }
+    /* Data table ------------------------------ */
+    .data-wrapper {
+      width: 100%;
+      margin: 0;
+      padding: 35px 0;
+    }
+    .data-table {
+      width: 100%;
+      margin: 0;
+    }
+    .data-table th {
+      text-align: left;
+      padding: 0px 5px;
+      padding-bottom: 8px;
+      border-bottom: 1px solid #EDEFF2;
+    }
+    .data-table th p {
+      margin: 0;
+      color: #9BA2AB;
+      font-size: 12px;
+    }
+    .data-table td {
+      padding: 10px 5px;
+      color: #74787E;
+      font-size: 15px;
+      line-height: 18px;
+    }
+    /* Buttons ------------------------------ */
+    .button {
+      display: inline-block;
+      width: 100%;
+      background-color: #00948d;
+      color: #ffffff;
+      font-size: 15px;
+      line-height: 45px;
+      text-align: center;
+      text-decoration: none;
+      -webkit-text-size-adjust: none;
+      mso-hide: all;
+    }
+    /*Media Queries ------------------------------ */
+    @media only screen and (max-width: 600px) {
+      .email-body_inner,
+      .email-footer {
+        width: 100% !important;
+      }
+    }
+  </style>
+</head>
+<body dir="{{.Hermes.TextDirection}}">
+  <table class="email-wrapper" width="100%" cellpadding="0" cellspacing="0">
+    <tr>
+      <td class="content">
+        <table class="email-content" width="100%" cellpadding="0" cellspacing="0">
+          <!-- Logo -->
+          <tr>
+            <td class="email-masthead">
+              <a class="email-masthead_name" href="{{.Hermes.Product.Link}}" target="_blank">
+                {{ if .Hermes.Product.Logo }}
+                  <img src="{{.Hermes.Product.Logo | url }}" class="email-logo" />
+                {{ else }}
+                  {{ .Hermes.Product.Name }}
+                {{ end }}
+                </a>
+            </td>
+          </tr>
+
+          <!-- Email Body -->
+          <tr>
+            <td class="email-body" width="100%">
+              <table class="email-body_inner" align="center" width="570" cellpadding="0" cellspacing="0">
+                <!-- Body content -->
+                <tr>
+                  <td class="content-cell">
+                    <h1>{{if .Email.Body.Title }}{{ .Email.Body.Title }}{{ else }}{{ .Email.Body.Greeting }} {{ .Email.Body.Name }},{{ end }}</h1>
+                    {{ with .Email.Body.Intros }}
+                        {{ if gt (len .) 0 }}
+                          {{ range $line := . }}
+                            <p>{{ $line }}</p>
+                          {{ end }}
+                        {{ end }}
+                    {{ end }}
+                    {{ if (ne .Email.Body.FreeMarkdown "") }}
+                      {{ .Email.Body.FreeMarkdown.ToHTML }}
+                    {{ else }}
+
+                      {{ with .Email.Body.Dictionary }} 
+                        {{ if gt (len .) 0 }}
+                          <dl class="body-dictionary">
+                            {{ range $entry := . }}
+                              <dt>{{ $entry.Key }}:</dt>
+                              <dd>{{ $entry.Value }}</dd>
+                            {{ end }}
+                          </dl>
+                        {{ end }}
+                      {{ end }}
+
+                      <!-- Table -->
+                      {{ with .Email.Body.Table }}
+                        {{ $data := .Data }}
+                        {{ $columns := .Columns }}
+                        {{ if gt (len $data) 0 }}
+                          <table class="data-wrapper" width="100%" cellpadding="0" cellspacing="0">
+                            <tr>
+                              <td colspan="2">
+                                <table class="data-table" width="100%" cellpadding="0" cellspacing="0">
+                                  <tr>
+                                    {{ $col := index $data 0 }}
+                                    {{ range $entry := $col }}
+                                      <th
+                                        {{ with $columns }}
+                                          {{ $width := index .CustomWidth $entry.Key }}
+                                          {{ with $width }}
+                                            width="{{ . }}"
+                                          {{ end }}
+                                          {{ $align := index .CustomAlignment $entry.Key }}
+                                          {{ with $align }}
+                                            style="text-align:{{ . }}"
+                                          {{ end }}
+                                        {{ end }}
+                                      >
+                                        <p>{{ $entry.Key }}</p>
+                                      </th>
+                                    {{ end }}
+                                  </tr>
+                                  {{ range $row := $data }}
+                                    <tr>
+                                      {{ range $cell := $row }}
+                                        <td
+                                          {{ with $columns }}
+                                            {{ $align := index .CustomAlignment $cell.Key }}
+                                            {{ with $align }}
+                                              style="text-align:{{ . }}"
+                                            {{ end }}
+                                          {{ end }}
+                                        >
+                                          {{ $cell.Value }}
+                                        </td>
+                                      {{ end }}
+                                    </tr>
+                                  {{ end }}
+                                </table>
+                              </td>
+                            </tr>
+                          </table>
+                        {{ end }}
+                      {{ end }}
+
+                      <!-- Action -->
+                      {{ with .Email.Body.Actions }}
+                        {{ if gt (len .) 0 }}
+                          {{ range $action := . }}
+                            <p>{{ $action.Instructions }}</p>
+                            <table class="body-action" align="center" width="100%" cellpadding="0" cellspacing="0">
+                              <tr>
+                                <td align="center">
+                                  <div>
+                                    <a href="{{ $action.Button.Link }}" class="button" style="background-color: {{ $action.Button.Color }}; color: {{ $action.Button.TextColor }}" target="_blank">
+                                      {{ $action.Button.Text }}
+                                    </a>
+                                  </div>
+                                </td>
+                              </tr>
+                            </table>
+                          {{ end }}
+                        {{ end }}
+                      {{ end }}
+
+                    {{ end }}
+                    {{ with .Email.Body.Outros }} 
+                        {{ if gt (len .) 0 }}
+                          {{ range $line := . }}
+                            <p>{{ $line }}</p>
+                          {{ end }}
+                        {{ end }}
+                      {{ end }}
+
+                    <p>
+                      {{.Email.Body.Signature}},
+                      <br />
+                      {{.Hermes.Product.Name}}
+                    </p>
+
+                    {{ if (eq .Email.Body.FreeMarkdown "") }}
+                      {{ with .Email.Body.Actions }} 
+                        <table class="body-sub">
+                          <tbody>
+                              {{ range $action := . }}
+                                <tr>
+                                  <td>
+                                    <p class="sub">{{$.Hermes.Product.TroubleText | replace "{ACTION}" $action.Button.Text}}</p>
+                                    <p class="sub"><a href="{{ $action.Button.Link }}">{{ $action.Button.Link }}</a></p>
+                                  </td>
+                                </tr>
+                              {{ end }}
+                          </tbody>
+                        </table>
+                      {{ end }}
+                    {{ end }}
+                  </td>
+                </tr>
+              </table>
+            </td>
+          </tr>
+          <tr>
+            <td>
+              <table class="email-footer" align="center" width="570" cellpadding="0" cellspacing="0">
+                <tr>
+                  <td class="content-cell">
+                    <p class="sub center">
+                      {{.Hermes.Product.Copyright}}
+                    </p>
+                  </td>
+                </tr>
+              </table>
+            </td>
+          </tr>
+        </table>
+      </td>
+    </tr>
+  </table>
+</body>
+</html>
+`
+}
+
+// PlainTextTemplate returns a Golang template that will generate an plain text email.
+func (dt *Flat) PlainTextTemplate() string {
+	return `<h2>{{if .Email.Body.Title }}{{ .Email.Body.Title }}{{ else }}{{ .Email.Body.Greeting }} {{ .Email.Body.Name }}{{ end }},</h2>
+{{ with .Email.Body.Intros }}
+  {{ range $line := . }}
+    <p>{{ $line }}</p>
+  {{ end }}
+{{ end }}
+{{ if (ne .Email.Body.FreeMarkdown "") }}
+  {{ .Email.Body.FreeMarkdown.ToHTML }}
+{{ else }}
+  {{ with .Email.Body.Dictionary }}
+    <ul>
+    {{ range $entry := . }}
+      <li>{{ $entry.Key }}: {{ $entry.Value }}</li>
+    {{ end }}
+    </ul>
+  {{ end }}
+  {{ with .Email.Body.Table }}
+    {{ $data := .Data }}
+    {{ $columns := .Columns }}
+    {{ if gt (len $data) 0 }}
+      <table class="data-table" width="100%" cellpadding="0" cellspacing="0">
+        <tr>
+          {{ $col := index $data 0 }}
+          {{ range $entry := $col }}
+            <th>{{ $entry.Key }} </th>
+          {{ end }}
+        </tr>
+        {{ range $row := $data }}
+          <tr>
+            {{ range $cell := $row }}
+              <td>
+                {{ $cell.Value }}
+              </td>
+            {{ end }}
+          </tr>
+        {{ end }}
+      </table>
+    {{ end }}
+  {{ end }}
+  {{ with .Email.Body.Actions }} 
+    {{ range $action := . }}
+      <p>{{ $action.Instructions }} {{ $action.Button.Link }}</p> 
+    {{ end }}
+  {{ end }}
+{{ end }}
+{{ with .Email.Body.Outros }} 
+  {{ range $line := . }}
+    <p>{{ $line }}<p>
+  {{ end }}
+{{ end }}
+<p>{{.Email.Body.Signature}},<br>{{.Hermes.Product.Name}} - {{.Hermes.Product.Link}}</p>
+
+<p>{{.Hermes.Product.Copyright}}</p>
+`
+}
diff --git a/backend/vendor/github.com/matcornic/hermes/gometalinter.json b/backend/vendor/github.com/matcornic/hermes/gometalinter.json
new file mode 100644
index 00000000..80d6aaf2
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/gometalinter.json
@@ -0,0 +1,17 @@
+{
+  "DisableAll": true,
+  "Enable": [
+    "unused",
+    "vet",
+    "vetshadow",
+    "deadcode",
+    "gofmt",
+    "golint",
+    "ineffassign",
+    "goconst",
+    "gosimple",
+    "staticcheck",
+    "misspell"
+  ],
+  "Test": true
+}
\ No newline at end of file
diff --git a/backend/vendor/github.com/matcornic/hermes/hermes.go b/backend/vendor/github.com/matcornic/hermes/hermes.go
new file mode 100644
index 00000000..6848ef49
--- /dev/null
+++ b/backend/vendor/github.com/matcornic/hermes/hermes.go
@@ -0,0 +1,200 @@
+package hermes
+
+import (
+	"bytes"
+	"github.com/Masterminds/sprig"
+	"github.com/imdario/mergo"
+	"github.com/jaytaylor/html2text"
+	"gopkg.in/russross/blackfriday.v2"
+	"html/template"
+)
+
+// Hermes is an instance of the hermes email generator
+type Hermes struct {
+	Theme         Theme
+	TextDirection TextDirection
+	Product       Product
+}
+
+// Theme is an interface to implement when creating a new theme
+type Theme interface {
+	Name() string              // The name of the theme
+	HTMLTemplate() string      // The golang template for HTML emails
+	PlainTextTemplate() string // The golang templte for plain text emails (can be basic HTML)
+}
+
+// TextDirection of the text in HTML email
+type TextDirection string
+
+var templateFuncs = template.FuncMap{
+	"url": func(s string) template.URL {
+		return template.URL(s)
+	},
+}
+
+// TDLeftToRight is the text direction from left to right (default)
+const TDLeftToRight TextDirection = "ltr"
+
+// TDRightToLeft is the text direction from right to left
+const TDRightToLeft TextDirection = "rtl"
+
+// Product represents your company product (brand)
+// Appears in header & footer of e-mails
+type Product struct {
+	Name        string
+	Link        string // e.g. https://matcornic.github.io
+	Logo        string // e.g. https://matcornic.github.io/img/logo.png
+	Copyright   string // Copyright © 2017 Hermes. All rights reserved.
+	TroubleText string // TroubleText is the sentence at the end of the email for users having trouble with the button (default to `If you’re having trouble with the button '{ACTION}', copy and paste the URL below into your web browser.`)
+}
+
+// Email is the email containing a body
+type Email struct {
+	Body Body
+}
+
+// Markdown is a HTML template (a string) representing Markdown content
+// https://en.wikipedia.org/wiki/Markdown
+type Markdown template.HTML
+
+// Body is the body of the email, containing all interesting data
+type Body struct {
+	Name         string   // The name of the contacted person
+	Intros       []string // Intro sentences, first displayed in the email
+	Dictionary   []Entry  // A list of key+value (useful for displaying parameters/settings/personal info)
+	Table        Table    // Table is an table where you can put data (pricing grid, a bill, and so on)
+	Actions      []Action // Actions are a list of actions that the user will be able to execute via a button click
+	Outros       []string // Outro sentences, last displayed in the email
+	Greeting     string   // Greeting for the contacted person (default to 'Hi')
+	Signature    string   // Signature for the contacted person (default to 'Yours truly')
+	Title        string   // Title replaces the greeting+name when set
+	FreeMarkdown Markdown // Free markdown content that replaces all content other than header and footer
+}
+
+// ToHTML converts Markdown to HTML
+func (c Markdown) ToHTML() template.HTML {
+	return template.HTML(blackfriday.Run([]byte(string(c))))
+}
+
+// Entry is a simple entry of a map
+// Allows using a slice of entries instead of a map
+// Because Golang maps are not ordered
+type Entry struct {
+	Key   string
+	Value string
+}
+
+// Table is an table where you can put data (pricing grid, a bill, and so on)
+type Table struct {
+	Data    [][]Entry // Contains data
+	Columns Columns   // Contains meta-data for display purpose (width, alignement)
+}
+
+// Columns contains meta-data for the different columns
+type Columns struct {
+	CustomWidth     map[string]string
+	CustomAlignment map[string]string
+}
+
+// Action is an action the user can do on the email (click on a button)
+type Action struct {
+	Instructions string
+	Button       Button
+}
+
+// Button defines an action to launch
+type Button struct {
+	Color     string
+	TextColor string
+	Text      string
+	Link      string
+}
+
+// Template is the struct given to Golang templating
+// Root object in a template is this struct
+type Template struct {
+	Hermes Hermes
+	Email  Email
+}
+
+func setDefaultEmailValues(e *Email) error {
+	// Default values of an email
+	defaultEmail := Email{
+		Body: Body{
+			Intros:     []string{},
+			Dictionary: []Entry{},
+			Outros:     []string{},
+			Signature:  "Yours truly",
+			Greeting:   "Hi",
+		},
+	}
+	// Merge the given email with default one
+	// Default one overrides all zero values
+	return mergo.Merge(e, defaultEmail)
+}
+
+// default values of the engine
+func setDefaultHermesValues(h *Hermes) error {
+	defaultTextDirection := TDLeftToRight
+	defaultHermes := Hermes{
+		Theme:         new(Default),
+		TextDirection: defaultTextDirection,
+		Product: Product{
+			Name:        "Hermes",
+			Copyright:   "Copyright © 2017 Hermes. All rights reserved.",
+			TroubleText: "If you’re having trouble with the button '{ACTION}', copy and paste the URL below into your web browser.",
+		},
+	}
+	// Merge the given hermes engine configuration with default one
+	// Default one overrides all zero values
+	err := mergo.Merge(h, defaultHermes)
+	if err != nil {
+		return err
+	}
+	if h.TextDirection != TDLeftToRight && h.TextDirection != TDRightToLeft {
+		h.TextDirection = defaultTextDirection
+	}
+	return nil
+}
+
+// GenerateHTML generates the email body from data to an HTML Reader
+// This is for modern email clients
+func (h *Hermes) GenerateHTML(email Email) (string, error) {
+	err := setDefaultHermesValues(h)
+	if err != nil {
+		return "", err
+	}
+	return h.generateTemplate(email, h.Theme.HTMLTemplate())
+}
+
+// GeneratePlainText generates the email body from data
+// This is for old email clients
+func (h *Hermes) GeneratePlainText(email Email) (string, error) {
+	err := setDefaultHermesValues(h)
+	if err != nil {
+		return "", err
+	}
+	template, err := h.generateTemplate(email, h.Theme.PlainTextTemplate())
+	if err != nil {
+		return "", err
+	}
+	return html2text.FromString(template, html2text.Options{PrettyTables: true})
+}
+
+func (h *Hermes) generateTemplate(email Email, tplt string) (string, error) {
+
+	err := setDefaultEmailValues(&email)
+	if err != nil {
+		return "", err
+	}
+
+	// Generate the email from Golang template
+	// Allow usage of simple function from sprig : https://github.com/Masterminds/sprig
+	t, err := template.New("hermes").Funcs(sprig.FuncMap()).Funcs(templateFuncs).Parse(tplt)
+	if err != nil {
+		return "", err
+	}
+	var b bytes.Buffer
+	t.Execute(&b, Template{*h, email})
+	return b.String(), nil
+}
diff --git a/backend/vendor/github.com/mattn/go-runewidth/.travis.yml b/backend/vendor/github.com/mattn/go-runewidth/.travis.yml
new file mode 100644
index 00000000..5c9c2a30
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/.travis.yml
@@ -0,0 +1,8 @@
+language: go
+go:
+  - tip
+before_install:
+  - go get github.com/mattn/goveralls
+  - go get golang.org/x/tools/cmd/cover
+script:
+    - $HOME/gopath/bin/goveralls -repotoken lAKAWPzcGsD3A8yBX3BGGtRUdJ6CaGERL
diff --git a/backend/vendor/github.com/mattn/go-runewidth/LICENSE b/backend/vendor/github.com/mattn/go-runewidth/LICENSE
new file mode 100644
index 00000000..91b5cef3
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/LICENSE
@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2016 Yasuhiro Matsumoto
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/backend/vendor/github.com/mattn/go-runewidth/README.mkd b/backend/vendor/github.com/mattn/go-runewidth/README.mkd
new file mode 100644
index 00000000..66663a94
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/README.mkd
@@ -0,0 +1,27 @@
+go-runewidth
+============
+
+[![Build Status](https://travis-ci.org/mattn/go-runewidth.png?branch=master)](https://travis-ci.org/mattn/go-runewidth)
+[![Coverage Status](https://coveralls.io/repos/mattn/go-runewidth/badge.png?branch=HEAD)](https://coveralls.io/r/mattn/go-runewidth?branch=HEAD)
+[![GoDoc](https://godoc.org/github.com/mattn/go-runewidth?status.svg)](http://godoc.org/github.com/mattn/go-runewidth)
+[![Go Report Card](https://goreportcard.com/badge/github.com/mattn/go-runewidth)](https://goreportcard.com/report/github.com/mattn/go-runewidth)
+
+Provides functions to get fixed width of the character or string.
+
+Usage
+-----
+
+```go
+runewidth.StringWidth("つのだ☆HIRO") == 12
+```
+
+
+Author
+------
+
+Yasuhiro Matsumoto
+
+License
+-------
+
+under the MIT License: http://mattn.mit-license.org/2013
diff --git a/backend/vendor/github.com/mattn/go-runewidth/runewidth.go b/backend/vendor/github.com/mattn/go-runewidth/runewidth.go
new file mode 100644
index 00000000..82568a1b
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/runewidth.go
@@ -0,0 +1,1235 @@
+package runewidth
+
+import "os"
+
+var (
+	// EastAsianWidth will be set true if the current locale is CJK
+	EastAsianWidth bool
+
+	// DefaultCondition is a condition in current locale
+	DefaultCondition = &Condition{EastAsianWidth}
+)
+
+func init() {
+	env := os.Getenv("RUNEWIDTH_EASTASIAN")
+	if env == "" {
+		EastAsianWidth = IsEastAsian()
+	} else {
+		EastAsianWidth = env == "1"
+	}
+}
+
+type interval struct {
+	first rune
+	last  rune
+}
+
+type table []interval
+
+func inTables(r rune, ts ...table) bool {
+	for _, t := range ts {
+		if inTable(r, t) {
+			return true
+		}
+	}
+	return false
+}
+
+func inTable(r rune, t table) bool {
+	// func (t table) IncludesRune(r rune) bool {
+	if r < t[0].first {
+		return false
+	}
+
+	bot := 0
+	top := len(t) - 1
+	for top >= bot {
+		mid := (bot + top) / 2
+
+		switch {
+		case t[mid].last < r:
+			bot = mid + 1
+		case t[mid].first > r:
+			top = mid - 1
+		default:
+			return true
+		}
+	}
+
+	return false
+}
+
+var private = table{
+	{0x00E000, 0x00F8FF}, {0x0F0000, 0x0FFFFD}, {0x100000, 0x10FFFD},
+}
+
+var nonprint = table{
+	{0x0000, 0x001F}, {0x007F, 0x009F}, {0x00AD, 0x00AD},
+	{0x070F, 0x070F}, {0x180B, 0x180E}, {0x200B, 0x200F},
+	{0x2028, 0x2029},
+	{0x202A, 0x202E}, {0x206A, 0x206F}, {0xD800, 0xDFFF},
+	{0xFEFF, 0xFEFF}, {0xFFF9, 0xFFFB}, {0xFFFE, 0xFFFF},
+}
+
+var combining = table{
+	{0x0300, 0x036F}, {0x0483, 0x0489}, {0x0591, 0x05BD},
+	{0x05BF, 0x05BF}, {0x05C1, 0x05C2}, {0x05C4, 0x05C5},
+	{0x05C7, 0x05C7}, {0x0610, 0x061A}, {0x064B, 0x065F},
+	{0x0670, 0x0670}, {0x06D6, 0x06DC}, {0x06DF, 0x06E4},
+	{0x06E7, 0x06E8}, {0x06EA, 0x06ED}, {0x0711, 0x0711},
+	{0x0730, 0x074A}, {0x07A6, 0x07B0}, {0x07EB, 0x07F3},
+	{0x0816, 0x0819}, {0x081B, 0x0823}, {0x0825, 0x0827},
+	{0x0829, 0x082D}, {0x0859, 0x085B}, {0x08D4, 0x08E1},
+	{0x08E3, 0x0903}, {0x093A, 0x093C}, {0x093E, 0x094F},
+	{0x0951, 0x0957}, {0x0962, 0x0963}, {0x0981, 0x0983},
+	{0x09BC, 0x09BC}, {0x09BE, 0x09C4}, {0x09C7, 0x09C8},
+	{0x09CB, 0x09CD}, {0x09D7, 0x09D7}, {0x09E2, 0x09E3},
+	{0x0A01, 0x0A03}, {0x0A3C, 0x0A3C}, {0x0A3E, 0x0A42},
+	{0x0A47, 0x0A48}, {0x0A4B, 0x0A4D}, {0x0A51, 0x0A51},
+	{0x0A70, 0x0A71}, {0x0A75, 0x0A75}, {0x0A81, 0x0A83},
+	{0x0ABC, 0x0ABC}, {0x0ABE, 0x0AC5}, {0x0AC7, 0x0AC9},
+	{0x0ACB, 0x0ACD}, {0x0AE2, 0x0AE3}, {0x0B01, 0x0B03},
+	{0x0B3C, 0x0B3C}, {0x0B3E, 0x0B44}, {0x0B47, 0x0B48},
+	{0x0B4B, 0x0B4D}, {0x0B56, 0x0B57}, {0x0B62, 0x0B63},
+	{0x0B82, 0x0B82}, {0x0BBE, 0x0BC2}, {0x0BC6, 0x0BC8},
+	{0x0BCA, 0x0BCD}, {0x0BD7, 0x0BD7}, {0x0C00, 0x0C03},
+	{0x0C3E, 0x0C44}, {0x0C46, 0x0C48}, {0x0C4A, 0x0C4D},
+	{0x0C55, 0x0C56}, {0x0C62, 0x0C63}, {0x0C81, 0x0C83},
+	{0x0CBC, 0x0CBC}, {0x0CBE, 0x0CC4}, {0x0CC6, 0x0CC8},
+	{0x0CCA, 0x0CCD}, {0x0CD5, 0x0CD6}, {0x0CE2, 0x0CE3},
+	{0x0D01, 0x0D03}, {0x0D3E, 0x0D44}, {0x0D46, 0x0D48},
+	{0x0D4A, 0x0D4D}, {0x0D57, 0x0D57}, {0x0D62, 0x0D63},
+	{0x0D82, 0x0D83}, {0x0DCA, 0x0DCA}, {0x0DCF, 0x0DD4},
+	{0x0DD6, 0x0DD6}, {0x0DD8, 0x0DDF}, {0x0DF2, 0x0DF3},
+	{0x0E31, 0x0E31}, {0x0E34, 0x0E3A}, {0x0E47, 0x0E4E},
+	{0x0EB1, 0x0EB1}, {0x0EB4, 0x0EB9}, {0x0EBB, 0x0EBC},
+	{0x0EC8, 0x0ECD}, {0x0F18, 0x0F19}, {0x0F35, 0x0F35},
+	{0x0F37, 0x0F37}, {0x0F39, 0x0F39}, {0x0F3E, 0x0F3F},
+	{0x0F71, 0x0F84}, {0x0F86, 0x0F87}, {0x0F8D, 0x0F97},
+	{0x0F99, 0x0FBC}, {0x0FC6, 0x0FC6}, {0x102B, 0x103E},
+	{0x1056, 0x1059}, {0x105E, 0x1060}, {0x1062, 0x1064},
+	{0x1067, 0x106D}, {0x1071, 0x1074}, {0x1082, 0x108D},
+	{0x108F, 0x108F}, {0x109A, 0x109D}, {0x135D, 0x135F},
+	{0x1712, 0x1714}, {0x1732, 0x1734}, {0x1752, 0x1753},
+	{0x1772, 0x1773}, {0x17B4, 0x17D3}, {0x17DD, 0x17DD},
+	{0x180B, 0x180D}, {0x1885, 0x1886}, {0x18A9, 0x18A9},
+	{0x1920, 0x192B}, {0x1930, 0x193B}, {0x1A17, 0x1A1B},
+	{0x1A55, 0x1A5E}, {0x1A60, 0x1A7C}, {0x1A7F, 0x1A7F},
+	{0x1AB0, 0x1ABE}, {0x1B00, 0x1B04}, {0x1B34, 0x1B44},
+	{0x1B6B, 0x1B73}, {0x1B80, 0x1B82}, {0x1BA1, 0x1BAD},
+	{0x1BE6, 0x1BF3}, {0x1C24, 0x1C37}, {0x1CD0, 0x1CD2},
+	{0x1CD4, 0x1CE8}, {0x1CED, 0x1CED}, {0x1CF2, 0x1CF4},
+	{0x1CF8, 0x1CF9}, {0x1DC0, 0x1DF5}, {0x1DFB, 0x1DFF},
+	{0x20D0, 0x20F0}, {0x2CEF, 0x2CF1}, {0x2D7F, 0x2D7F},
+	{0x2DE0, 0x2DFF}, {0x302A, 0x302F}, {0x3099, 0x309A},
+	{0xA66F, 0xA672}, {0xA674, 0xA67D}, {0xA69E, 0xA69F},
+	{0xA6F0, 0xA6F1}, {0xA802, 0xA802}, {0xA806, 0xA806},
+	{0xA80B, 0xA80B}, {0xA823, 0xA827}, {0xA880, 0xA881},
+	{0xA8B4, 0xA8C5}, {0xA8E0, 0xA8F1}, {0xA926, 0xA92D},
+	{0xA947, 0xA953}, {0xA980, 0xA983}, {0xA9B3, 0xA9C0},
+	{0xA9E5, 0xA9E5}, {0xAA29, 0xAA36}, {0xAA43, 0xAA43},
+	{0xAA4C, 0xAA4D}, {0xAA7B, 0xAA7D}, {0xAAB0, 0xAAB0},
+	{0xAAB2, 0xAAB4}, {0xAAB7, 0xAAB8}, {0xAABE, 0xAABF},
+	{0xAAC1, 0xAAC1}, {0xAAEB, 0xAAEF}, {0xAAF5, 0xAAF6},
+	{0xABE3, 0xABEA}, {0xABEC, 0xABED}, {0xFB1E, 0xFB1E},
+	{0xFE00, 0xFE0F}, {0xFE20, 0xFE2F}, {0x101FD, 0x101FD},
+	{0x102E0, 0x102E0}, {0x10376, 0x1037A}, {0x10A01, 0x10A03},
+	{0x10A05, 0x10A06}, {0x10A0C, 0x10A0F}, {0x10A38, 0x10A3A},
+	{0x10A3F, 0x10A3F}, {0x10AE5, 0x10AE6}, {0x11000, 0x11002},
+	{0x11038, 0x11046}, {0x1107F, 0x11082}, {0x110B0, 0x110BA},
+	{0x11100, 0x11102}, {0x11127, 0x11134}, {0x11173, 0x11173},
+	{0x11180, 0x11182}, {0x111B3, 0x111C0}, {0x111CA, 0x111CC},
+	{0x1122C, 0x11237}, {0x1123E, 0x1123E}, {0x112DF, 0x112EA},
+	{0x11300, 0x11303}, {0x1133C, 0x1133C}, {0x1133E, 0x11344},
+	{0x11347, 0x11348}, {0x1134B, 0x1134D}, {0x11357, 0x11357},
+	{0x11362, 0x11363}, {0x11366, 0x1136C}, {0x11370, 0x11374},
+	{0x11435, 0x11446}, {0x114B0, 0x114C3}, {0x115AF, 0x115B5},
+	{0x115B8, 0x115C0}, {0x115DC, 0x115DD}, {0x11630, 0x11640},
+	{0x116AB, 0x116B7}, {0x1171D, 0x1172B}, {0x11C2F, 0x11C36},
+	{0x11C38, 0x11C3F}, {0x11C92, 0x11CA7}, {0x11CA9, 0x11CB6},
+	{0x16AF0, 0x16AF4}, {0x16B30, 0x16B36}, {0x16F51, 0x16F7E},
+	{0x16F8F, 0x16F92}, {0x1BC9D, 0x1BC9E}, {0x1D165, 0x1D169},
+	{0x1D16D, 0x1D172}, {0x1D17B, 0x1D182}, {0x1D185, 0x1D18B},
+	{0x1D1AA, 0x1D1AD}, {0x1D242, 0x1D244}, {0x1DA00, 0x1DA36},
+	{0x1DA3B, 0x1DA6C}, {0x1DA75, 0x1DA75}, {0x1DA84, 0x1DA84},
+	{0x1DA9B, 0x1DA9F}, {0x1DAA1, 0x1DAAF}, {0x1E000, 0x1E006},
+	{0x1E008, 0x1E018}, {0x1E01B, 0x1E021}, {0x1E023, 0x1E024},
+	{0x1E026, 0x1E02A}, {0x1E8D0, 0x1E8D6}, {0x1E944, 0x1E94A},
+	{0xE0100, 0xE01EF},
+}
+
+var doublewidth = table{
+	{0x1100, 0x115F}, {0x231A, 0x231B}, {0x2329, 0x232A},
+	{0x23E9, 0x23EC}, {0x23F0, 0x23F0}, {0x23F3, 0x23F3},
+	{0x25FD, 0x25FE}, {0x2614, 0x2615}, {0x2648, 0x2653},
+	{0x267F, 0x267F}, {0x2693, 0x2693}, {0x26A1, 0x26A1},
+	{0x26AA, 0x26AB}, {0x26BD, 0x26BE}, {0x26C4, 0x26C5},
+	{0x26CE, 0x26CE}, {0x26D4, 0x26D4}, {0x26EA, 0x26EA},
+	{0x26F2, 0x26F3}, {0x26F5, 0x26F5}, {0x26FA, 0x26FA},
+	{0x26FD, 0x26FD}, {0x2705, 0x2705}, {0x270A, 0x270B},
+	{0x2728, 0x2728}, {0x274C, 0x274C}, {0x274E, 0x274E},
+	{0x2753, 0x2755}, {0x2757, 0x2757}, {0x2795, 0x2797},
+	{0x27B0, 0x27B0}, {0x27BF, 0x27BF}, {0x2B1B, 0x2B1C},
+	{0x2B50, 0x2B50}, {0x2B55, 0x2B55}, {0x2E80, 0x2E99},
+	{0x2E9B, 0x2EF3}, {0x2F00, 0x2FD5}, {0x2FF0, 0x2FFB},
+	{0x3000, 0x303E}, {0x3041, 0x3096}, {0x3099, 0x30FF},
+	{0x3105, 0x312D}, {0x3131, 0x318E}, {0x3190, 0x31BA},
+	{0x31C0, 0x31E3}, {0x31F0, 0x321E}, {0x3220, 0x3247},
+	{0x3250, 0x32FE}, {0x3300, 0x4DBF}, {0x4E00, 0xA48C},
+	{0xA490, 0xA4C6}, {0xA960, 0xA97C}, {0xAC00, 0xD7A3},
+	{0xF900, 0xFAFF}, {0xFE10, 0xFE19}, {0xFE30, 0xFE52},
+	{0xFE54, 0xFE66}, {0xFE68, 0xFE6B}, {0xFF01, 0xFF60},
+	{0xFFE0, 0xFFE6}, {0x16FE0, 0x16FE0}, {0x17000, 0x187EC},
+	{0x18800, 0x18AF2}, {0x1B000, 0x1B001}, {0x1F004, 0x1F004},
+	{0x1F0CF, 0x1F0CF}, {0x1F18E, 0x1F18E}, {0x1F191, 0x1F19A},
+	{0x1F200, 0x1F202}, {0x1F210, 0x1F23B}, {0x1F240, 0x1F248},
+	{0x1F250, 0x1F251}, {0x1F300, 0x1F320}, {0x1F32D, 0x1F335},
+	{0x1F337, 0x1F37C}, {0x1F37E, 0x1F393}, {0x1F3A0, 0x1F3CA},
+	{0x1F3CF, 0x1F3D3}, {0x1F3E0, 0x1F3F0}, {0x1F3F4, 0x1F3F4},
+	{0x1F3F8, 0x1F43E}, {0x1F440, 0x1F440}, {0x1F442, 0x1F4FC},
+	{0x1F4FF, 0x1F53D}, {0x1F54B, 0x1F54E}, {0x1F550, 0x1F567},
+	{0x1F57A, 0x1F57A}, {0x1F595, 0x1F596}, {0x1F5A4, 0x1F5A4},
+	{0x1F5FB, 0x1F64F}, {0x1F680, 0x1F6C5}, {0x1F6CC, 0x1F6CC},
+	{0x1F6D0, 0x1F6D2}, {0x1F6EB, 0x1F6EC}, {0x1F6F4, 0x1F6F6},
+	{0x1F910, 0x1F91E}, {0x1F920, 0x1F927}, {0x1F930, 0x1F930},
+	{0x1F933, 0x1F93E}, {0x1F940, 0x1F94B}, {0x1F950, 0x1F95E},
+	{0x1F980, 0x1F991}, {0x1F9C0, 0x1F9C0}, {0x20000, 0x2FFFD},
+	{0x30000, 0x3FFFD},
+}
+
+var ambiguous = table{
+	{0x00A1, 0x00A1}, {0x00A4, 0x00A4}, {0x00A7, 0x00A8},
+	{0x00AA, 0x00AA}, {0x00AD, 0x00AE}, {0x00B0, 0x00B4},
+	{0x00B6, 0x00BA}, {0x00BC, 0x00BF}, {0x00C6, 0x00C6},
+	{0x00D0, 0x00D0}, {0x00D7, 0x00D8}, {0x00DE, 0x00E1},
+	{0x00E6, 0x00E6}, {0x00E8, 0x00EA}, {0x00EC, 0x00ED},
+	{0x00F0, 0x00F0}, {0x00F2, 0x00F3}, {0x00F7, 0x00FA},
+	{0x00FC, 0x00FC}, {0x00FE, 0x00FE}, {0x0101, 0x0101},
+	{0x0111, 0x0111}, {0x0113, 0x0113}, {0x011B, 0x011B},
+	{0x0126, 0x0127}, {0x012B, 0x012B}, {0x0131, 0x0133},
+	{0x0138, 0x0138}, {0x013F, 0x0142}, {0x0144, 0x0144},
+	{0x0148, 0x014B}, {0x014D, 0x014D}, {0x0152, 0x0153},
+	{0x0166, 0x0167}, {0x016B, 0x016B}, {0x01CE, 0x01CE},
+	{0x01D0, 0x01D0}, {0x01D2, 0x01D2}, {0x01D4, 0x01D4},
+	{0x01D6, 0x01D6}, {0x01D8, 0x01D8}, {0x01DA, 0x01DA},
+	{0x01DC, 0x01DC}, {0x0251, 0x0251}, {0x0261, 0x0261},
+	{0x02C4, 0x02C4}, {0x02C7, 0x02C7}, {0x02C9, 0x02CB},
+	{0x02CD, 0x02CD}, {0x02D0, 0x02D0}, {0x02D8, 0x02DB},
+	{0x02DD, 0x02DD}, {0x02DF, 0x02DF}, {0x0300, 0x036F},
+	{0x0391, 0x03A1}, {0x03A3, 0x03A9}, {0x03B1, 0x03C1},
+	{0x03C3, 0x03C9}, {0x0401, 0x0401}, {0x0410, 0x044F},
+	{0x0451, 0x0451}, {0x2010, 0x2010}, {0x2013, 0x2016},
+	{0x2018, 0x2019}, {0x201C, 0x201D}, {0x2020, 0x2022},
+	{0x2024, 0x2027}, {0x2030, 0x2030}, {0x2032, 0x2033},
+	{0x2035, 0x2035}, {0x203B, 0x203B}, {0x203E, 0x203E},
+	{0x2074, 0x2074}, {0x207F, 0x207F}, {0x2081, 0x2084},
+	{0x20AC, 0x20AC}, {0x2103, 0x2103}, {0x2105, 0x2105},
+	{0x2109, 0x2109}, {0x2113, 0x2113}, {0x2116, 0x2116},
+	{0x2121, 0x2122}, {0x2126, 0x2126}, {0x212B, 0x212B},
+	{0x2153, 0x2154}, {0x215B, 0x215E}, {0x2160, 0x216B},
+	{0x2170, 0x2179}, {0x2189, 0x2189}, {0x2190, 0x2199},
+	{0x21B8, 0x21B9}, {0x21D2, 0x21D2}, {0x21D4, 0x21D4},
+	{0x21E7, 0x21E7}, {0x2200, 0x2200}, {0x2202, 0x2203},
+	{0x2207, 0x2208}, {0x220B, 0x220B}, {0x220F, 0x220F},
+	{0x2211, 0x2211}, {0x2215, 0x2215}, {0x221A, 0x221A},
+	{0x221D, 0x2220}, {0x2223, 0x2223}, {0x2225, 0x2225},
+	{0x2227, 0x222C}, {0x222E, 0x222E}, {0x2234, 0x2237},
+	{0x223C, 0x223D}, {0x2248, 0x2248}, {0x224C, 0x224C},
+	{0x2252, 0x2252}, {0x2260, 0x2261}, {0x2264, 0x2267},
+	{0x226A, 0x226B}, {0x226E, 0x226F}, {0x2282, 0x2283},
+	{0x2286, 0x2287}, {0x2295, 0x2295}, {0x2299, 0x2299},
+	{0x22A5, 0x22A5}, {0x22BF, 0x22BF}, {0x2312, 0x2312},
+	{0x2460, 0x24E9}, {0x24EB, 0x254B}, {0x2550, 0x2573},
+	{0x2580, 0x258F}, {0x2592, 0x2595}, {0x25A0, 0x25A1},
+	{0x25A3, 0x25A9}, {0x25B2, 0x25B3}, {0x25B6, 0x25B7},
+	{0x25BC, 0x25BD}, {0x25C0, 0x25C1}, {0x25C6, 0x25C8},
+	{0x25CB, 0x25CB}, {0x25CE, 0x25D1}, {0x25E2, 0x25E5},
+	{0x25EF, 0x25EF}, {0x2605, 0x2606}, {0x2609, 0x2609},
+	{0x260E, 0x260F}, {0x261C, 0x261C}, {0x261E, 0x261E},
+	{0x2640, 0x2640}, {0x2642, 0x2642}, {0x2660, 0x2661},
+	{0x2663, 0x2665}, {0x2667, 0x266A}, {0x266C, 0x266D},
+	{0x266F, 0x266F}, {0x269E, 0x269F}, {0x26BF, 0x26BF},
+	{0x26C6, 0x26CD}, {0x26CF, 0x26D3}, {0x26D5, 0x26E1},
+	{0x26E3, 0x26E3}, {0x26E8, 0x26E9}, {0x26EB, 0x26F1},
+	{0x26F4, 0x26F4}, {0x26F6, 0x26F9}, {0x26FB, 0x26FC},
+	{0x26FE, 0x26FF}, {0x273D, 0x273D}, {0x2776, 0x277F},
+	{0x2B56, 0x2B59}, {0x3248, 0x324F}, {0xE000, 0xF8FF},
+	{0xFE00, 0xFE0F}, {0xFFFD, 0xFFFD}, {0x1F100, 0x1F10A},
+	{0x1F110, 0x1F12D}, {0x1F130, 0x1F169}, {0x1F170, 0x1F18D},
+	{0x1F18F, 0x1F190}, {0x1F19B, 0x1F1AC}, {0xE0100, 0xE01EF},
+	{0xF0000, 0xFFFFD}, {0x100000, 0x10FFFD},
+}
+
+var emoji = table{
+	{0x1F1E6, 0x1F1FF}, {0x1F321, 0x1F321}, {0x1F324, 0x1F32C},
+	{0x1F336, 0x1F336}, {0x1F37D, 0x1F37D}, {0x1F396, 0x1F397},
+	{0x1F399, 0x1F39B}, {0x1F39E, 0x1F39F}, {0x1F3CB, 0x1F3CE},
+	{0x1F3D4, 0x1F3DF}, {0x1F3F3, 0x1F3F5}, {0x1F3F7, 0x1F3F7},
+	{0x1F43F, 0x1F43F}, {0x1F441, 0x1F441}, {0x1F4FD, 0x1F4FD},
+	{0x1F549, 0x1F54A}, {0x1F56F, 0x1F570}, {0x1F573, 0x1F579},
+	{0x1F587, 0x1F587}, {0x1F58A, 0x1F58D}, {0x1F590, 0x1F590},
+	{0x1F5A5, 0x1F5A5}, {0x1F5A8, 0x1F5A8}, {0x1F5B1, 0x1F5B2},
+	{0x1F5BC, 0x1F5BC}, {0x1F5C2, 0x1F5C4}, {0x1F5D1, 0x1F5D3},
+	{0x1F5DC, 0x1F5DE}, {0x1F5E1, 0x1F5E1}, {0x1F5E3, 0x1F5E3},
+	{0x1F5E8, 0x1F5E8}, {0x1F5EF, 0x1F5EF}, {0x1F5F3, 0x1F5F3},
+	{0x1F5FA, 0x1F5FA}, {0x1F6CB, 0x1F6CF}, {0x1F6E0, 0x1F6E5},
+	{0x1F6E9, 0x1F6E9}, {0x1F6F0, 0x1F6F0}, {0x1F6F3, 0x1F6F3},
+}
+
+var notassigned = table{
+	{0x0378, 0x0379}, {0x0380, 0x0383}, {0x038B, 0x038B},
+	{0x038D, 0x038D}, {0x03A2, 0x03A2}, {0x0530, 0x0530},
+	{0x0557, 0x0558}, {0x0560, 0x0560}, {0x0588, 0x0588},
+	{0x058B, 0x058C}, {0x0590, 0x0590}, {0x05C8, 0x05CF},
+	{0x05EB, 0x05EF}, {0x05F5, 0x05FF}, {0x061D, 0x061D},
+	{0x070E, 0x070E}, {0x074B, 0x074C}, {0x07B2, 0x07BF},
+	{0x07FB, 0x07FF}, {0x082E, 0x082F}, {0x083F, 0x083F},
+	{0x085C, 0x085D}, {0x085F, 0x089F}, {0x08B5, 0x08B5},
+	{0x08BE, 0x08D3}, {0x0984, 0x0984}, {0x098D, 0x098E},
+	{0x0991, 0x0992}, {0x09A9, 0x09A9}, {0x09B1, 0x09B1},
+	{0x09B3, 0x09B5}, {0x09BA, 0x09BB}, {0x09C5, 0x09C6},
+	{0x09C9, 0x09CA}, {0x09CF, 0x09D6}, {0x09D8, 0x09DB},
+	{0x09DE, 0x09DE}, {0x09E4, 0x09E5}, {0x09FC, 0x0A00},
+	{0x0A04, 0x0A04}, {0x0A0B, 0x0A0E}, {0x0A11, 0x0A12},
+	{0x0A29, 0x0A29}, {0x0A31, 0x0A31}, {0x0A34, 0x0A34},
+	{0x0A37, 0x0A37}, {0x0A3A, 0x0A3B}, {0x0A3D, 0x0A3D},
+	{0x0A43, 0x0A46}, {0x0A49, 0x0A4A}, {0x0A4E, 0x0A50},
+	{0x0A52, 0x0A58}, {0x0A5D, 0x0A5D}, {0x0A5F, 0x0A65},
+	{0x0A76, 0x0A80}, {0x0A84, 0x0A84}, {0x0A8E, 0x0A8E},
+	{0x0A92, 0x0A92}, {0x0AA9, 0x0AA9}, {0x0AB1, 0x0AB1},
+	{0x0AB4, 0x0AB4}, {0x0ABA, 0x0ABB}, {0x0AC6, 0x0AC6},
+	{0x0ACA, 0x0ACA}, {0x0ACE, 0x0ACF}, {0x0AD1, 0x0ADF},
+	{0x0AE4, 0x0AE5}, {0x0AF2, 0x0AF8}, {0x0AFA, 0x0B00},
+	{0x0B04, 0x0B04}, {0x0B0D, 0x0B0E}, {0x0B11, 0x0B12},
+	{0x0B29, 0x0B29}, {0x0B31, 0x0B31}, {0x0B34, 0x0B34},
+	{0x0B3A, 0x0B3B}, {0x0B45, 0x0B46}, {0x0B49, 0x0B4A},
+	{0x0B4E, 0x0B55}, {0x0B58, 0x0B5B}, {0x0B5E, 0x0B5E},
+	{0x0B64, 0x0B65}, {0x0B78, 0x0B81}, {0x0B84, 0x0B84},
+	{0x0B8B, 0x0B8D}, {0x0B91, 0x0B91}, {0x0B96, 0x0B98},
+	{0x0B9B, 0x0B9B}, {0x0B9D, 0x0B9D}, {0x0BA0, 0x0BA2},
+	{0x0BA5, 0x0BA7}, {0x0BAB, 0x0BAD}, {0x0BBA, 0x0BBD},
+	{0x0BC3, 0x0BC5}, {0x0BC9, 0x0BC9}, {0x0BCE, 0x0BCF},
+	{0x0BD1, 0x0BD6}, {0x0BD8, 0x0BE5}, {0x0BFB, 0x0BFF},
+	{0x0C04, 0x0C04}, {0x0C0D, 0x0C0D}, {0x0C11, 0x0C11},
+	{0x0C29, 0x0C29}, {0x0C3A, 0x0C3C}, {0x0C45, 0x0C45},
+	{0x0C49, 0x0C49}, {0x0C4E, 0x0C54}, {0x0C57, 0x0C57},
+	{0x0C5B, 0x0C5F}, {0x0C64, 0x0C65}, {0x0C70, 0x0C77},
+	{0x0C84, 0x0C84}, {0x0C8D, 0x0C8D}, {0x0C91, 0x0C91},
+	{0x0CA9, 0x0CA9}, {0x0CB4, 0x0CB4}, {0x0CBA, 0x0CBB},
+	{0x0CC5, 0x0CC5}, {0x0CC9, 0x0CC9}, {0x0CCE, 0x0CD4},
+	{0x0CD7, 0x0CDD}, {0x0CDF, 0x0CDF}, {0x0CE4, 0x0CE5},
+	{0x0CF0, 0x0CF0}, {0x0CF3, 0x0D00}, {0x0D04, 0x0D04},
+	{0x0D0D, 0x0D0D}, {0x0D11, 0x0D11}, {0x0D3B, 0x0D3C},
+	{0x0D45, 0x0D45}, {0x0D49, 0x0D49}, {0x0D50, 0x0D53},
+	{0x0D64, 0x0D65}, {0x0D80, 0x0D81}, {0x0D84, 0x0D84},
+	{0x0D97, 0x0D99}, {0x0DB2, 0x0DB2}, {0x0DBC, 0x0DBC},
+	{0x0DBE, 0x0DBF}, {0x0DC7, 0x0DC9}, {0x0DCB, 0x0DCE},
+	{0x0DD5, 0x0DD5}, {0x0DD7, 0x0DD7}, {0x0DE0, 0x0DE5},
+	{0x0DF0, 0x0DF1}, {0x0DF5, 0x0E00}, {0x0E3B, 0x0E3E},
+	{0x0E5C, 0x0E80}, {0x0E83, 0x0E83}, {0x0E85, 0x0E86},
+	{0x0E89, 0x0E89}, {0x0E8B, 0x0E8C}, {0x0E8E, 0x0E93},
+	{0x0E98, 0x0E98}, {0x0EA0, 0x0EA0}, {0x0EA4, 0x0EA4},
+	{0x0EA6, 0x0EA6}, {0x0EA8, 0x0EA9}, {0x0EAC, 0x0EAC},
+	{0x0EBA, 0x0EBA}, {0x0EBE, 0x0EBF}, {0x0EC5, 0x0EC5},
+	{0x0EC7, 0x0EC7}, {0x0ECE, 0x0ECF}, {0x0EDA, 0x0EDB},
+	{0x0EE0, 0x0EFF}, {0x0F48, 0x0F48}, {0x0F6D, 0x0F70},
+	{0x0F98, 0x0F98}, {0x0FBD, 0x0FBD}, {0x0FCD, 0x0FCD},
+	{0x0FDB, 0x0FFF}, {0x10C6, 0x10C6}, {0x10C8, 0x10CC},
+	{0x10CE, 0x10CF}, {0x1249, 0x1249}, {0x124E, 0x124F},
+	{0x1257, 0x1257}, {0x1259, 0x1259}, {0x125E, 0x125F},
+	{0x1289, 0x1289}, {0x128E, 0x128F}, {0x12B1, 0x12B1},
+	{0x12B6, 0x12B7}, {0x12BF, 0x12BF}, {0x12C1, 0x12C1},
+	{0x12C6, 0x12C7}, {0x12D7, 0x12D7}, {0x1311, 0x1311},
+	{0x1316, 0x1317}, {0x135B, 0x135C}, {0x137D, 0x137F},
+	{0x139A, 0x139F}, {0x13F6, 0x13F7}, {0x13FE, 0x13FF},
+	{0x169D, 0x169F}, {0x16F9, 0x16FF}, {0x170D, 0x170D},
+	{0x1715, 0x171F}, {0x1737, 0x173F}, {0x1754, 0x175F},
+	{0x176D, 0x176D}, {0x1771, 0x1771}, {0x1774, 0x177F},
+	{0x17DE, 0x17DF}, {0x17EA, 0x17EF}, {0x17FA, 0x17FF},
+	{0x180F, 0x180F}, {0x181A, 0x181F}, {0x1878, 0x187F},
+	{0x18AB, 0x18AF}, {0x18F6, 0x18FF}, {0x191F, 0x191F},
+	{0x192C, 0x192F}, {0x193C, 0x193F}, {0x1941, 0x1943},
+	{0x196E, 0x196F}, {0x1975, 0x197F}, {0x19AC, 0x19AF},
+	{0x19CA, 0x19CF}, {0x19DB, 0x19DD}, {0x1A1C, 0x1A1D},
+	{0x1A5F, 0x1A5F}, {0x1A7D, 0x1A7E}, {0x1A8A, 0x1A8F},
+	{0x1A9A, 0x1A9F}, {0x1AAE, 0x1AAF}, {0x1ABF, 0x1AFF},
+	{0x1B4C, 0x1B4F}, {0x1B7D, 0x1B7F}, {0x1BF4, 0x1BFB},
+	{0x1C38, 0x1C3A}, {0x1C4A, 0x1C4C}, {0x1C89, 0x1CBF},
+	{0x1CC8, 0x1CCF}, {0x1CF7, 0x1CF7}, {0x1CFA, 0x1CFF},
+	{0x1DF6, 0x1DFA}, {0x1F16, 0x1F17}, {0x1F1E, 0x1F1F},
+	{0x1F46, 0x1F47}, {0x1F4E, 0x1F4F}, {0x1F58, 0x1F58},
+	{0x1F5A, 0x1F5A}, {0x1F5C, 0x1F5C}, {0x1F5E, 0x1F5E},
+	{0x1F7E, 0x1F7F}, {0x1FB5, 0x1FB5}, {0x1FC5, 0x1FC5},
+	{0x1FD4, 0x1FD5}, {0x1FDC, 0x1FDC}, {0x1FF0, 0x1FF1},
+	{0x1FF5, 0x1FF5}, {0x1FFF, 0x1FFF}, {0x2065, 0x2065},
+	{0x2072, 0x2073}, {0x208F, 0x208F}, {0x209D, 0x209F},
+	{0x20BF, 0x20CF}, {0x20F1, 0x20FF}, {0x218C, 0x218F},
+	{0x23FF, 0x23FF}, {0x2427, 0x243F}, {0x244B, 0x245F},
+	{0x2B74, 0x2B75}, {0x2B96, 0x2B97}, {0x2BBA, 0x2BBC},
+	{0x2BC9, 0x2BC9}, {0x2BD2, 0x2BEB}, {0x2BF0, 0x2BFF},
+	{0x2C2F, 0x2C2F}, {0x2C5F, 0x2C5F}, {0x2CF4, 0x2CF8},
+	{0x2D26, 0x2D26}, {0x2D28, 0x2D2C}, {0x2D2E, 0x2D2F},
+	{0x2D68, 0x2D6E}, {0x2D71, 0x2D7E}, {0x2D97, 0x2D9F},
+	{0x2DA7, 0x2DA7}, {0x2DAF, 0x2DAF}, {0x2DB7, 0x2DB7},
+	{0x2DBF, 0x2DBF}, {0x2DC7, 0x2DC7}, {0x2DCF, 0x2DCF},
+	{0x2DD7, 0x2DD7}, {0x2DDF, 0x2DDF}, {0x2E45, 0x2E7F},
+	{0x2E9A, 0x2E9A}, {0x2EF4, 0x2EFF}, {0x2FD6, 0x2FEF},
+	{0x2FFC, 0x2FFF}, {0x3040, 0x3040}, {0x3097, 0x3098},
+	{0x3100, 0x3104}, {0x312E, 0x3130}, {0x318F, 0x318F},
+	{0x31BB, 0x31BF}, {0x31E4, 0x31EF}, {0x321F, 0x321F},
+	{0x32FF, 0x32FF}, {0x4DB6, 0x4DBF}, {0x9FD6, 0x9FFF},
+	{0xA48D, 0xA48F}, {0xA4C7, 0xA4CF}, {0xA62C, 0xA63F},
+	{0xA6F8, 0xA6FF}, {0xA7AF, 0xA7AF}, {0xA7B8, 0xA7F6},
+	{0xA82C, 0xA82F}, {0xA83A, 0xA83F}, {0xA878, 0xA87F},
+	{0xA8C6, 0xA8CD}, {0xA8DA, 0xA8DF}, {0xA8FE, 0xA8FF},
+	{0xA954, 0xA95E}, {0xA97D, 0xA97F}, {0xA9CE, 0xA9CE},
+	{0xA9DA, 0xA9DD}, {0xA9FF, 0xA9FF}, {0xAA37, 0xAA3F},
+	{0xAA4E, 0xAA4F}, {0xAA5A, 0xAA5B}, {0xAAC3, 0xAADA},
+	{0xAAF7, 0xAB00}, {0xAB07, 0xAB08}, {0xAB0F, 0xAB10},
+	{0xAB17, 0xAB1F}, {0xAB27, 0xAB27}, {0xAB2F, 0xAB2F},
+	{0xAB66, 0xAB6F}, {0xABEE, 0xABEF}, {0xABFA, 0xABFF},
+	{0xD7A4, 0xD7AF}, {0xD7C7, 0xD7CA}, {0xD7FC, 0xD7FF},
+	{0xFA6E, 0xFA6F}, {0xFADA, 0xFAFF}, {0xFB07, 0xFB12},
+	{0xFB18, 0xFB1C}, {0xFB37, 0xFB37}, {0xFB3D, 0xFB3D},
+	{0xFB3F, 0xFB3F}, {0xFB42, 0xFB42}, {0xFB45, 0xFB45},
+	{0xFBC2, 0xFBD2}, {0xFD40, 0xFD4F}, {0xFD90, 0xFD91},
+	{0xFDC8, 0xFDEF}, {0xFDFE, 0xFDFF}, {0xFE1A, 0xFE1F},
+	{0xFE53, 0xFE53}, {0xFE67, 0xFE67}, {0xFE6C, 0xFE6F},
+	{0xFE75, 0xFE75}, {0xFEFD, 0xFEFE}, {0xFF00, 0xFF00},
+	{0xFFBF, 0xFFC1}, {0xFFC8, 0xFFC9}, {0xFFD0, 0xFFD1},
+	{0xFFD8, 0xFFD9}, {0xFFDD, 0xFFDF}, {0xFFE7, 0xFFE7},
+	{0xFFEF, 0xFFF8}, {0xFFFE, 0xFFFF}, {0x1000C, 0x1000C},
+	{0x10027, 0x10027}, {0x1003B, 0x1003B}, {0x1003E, 0x1003E},
+	{0x1004E, 0x1004F}, {0x1005E, 0x1007F}, {0x100FB, 0x100FF},
+	{0x10103, 0x10106}, {0x10134, 0x10136}, {0x1018F, 0x1018F},
+	{0x1019C, 0x1019F}, {0x101A1, 0x101CF}, {0x101FE, 0x1027F},
+	{0x1029D, 0x1029F}, {0x102D1, 0x102DF}, {0x102FC, 0x102FF},
+	{0x10324, 0x1032F}, {0x1034B, 0x1034F}, {0x1037B, 0x1037F},
+	{0x1039E, 0x1039E}, {0x103C4, 0x103C7}, {0x103D6, 0x103FF},
+	{0x1049E, 0x1049F}, {0x104AA, 0x104AF}, {0x104D4, 0x104D7},
+	{0x104FC, 0x104FF}, {0x10528, 0x1052F}, {0x10564, 0x1056E},
+	{0x10570, 0x105FF}, {0x10737, 0x1073F}, {0x10756, 0x1075F},
+	{0x10768, 0x107FF}, {0x10806, 0x10807}, {0x10809, 0x10809},
+	{0x10836, 0x10836}, {0x10839, 0x1083B}, {0x1083D, 0x1083E},
+	{0x10856, 0x10856}, {0x1089F, 0x108A6}, {0x108B0, 0x108DF},
+	{0x108F3, 0x108F3}, {0x108F6, 0x108FA}, {0x1091C, 0x1091E},
+	{0x1093A, 0x1093E}, {0x10940, 0x1097F}, {0x109B8, 0x109BB},
+	{0x109D0, 0x109D1}, {0x10A04, 0x10A04}, {0x10A07, 0x10A0B},
+	{0x10A14, 0x10A14}, {0x10A18, 0x10A18}, {0x10A34, 0x10A37},
+	{0x10A3B, 0x10A3E}, {0x10A48, 0x10A4F}, {0x10A59, 0x10A5F},
+	{0x10AA0, 0x10ABF}, {0x10AE7, 0x10AEA}, {0x10AF7, 0x10AFF},
+	{0x10B36, 0x10B38}, {0x10B56, 0x10B57}, {0x10B73, 0x10B77},
+	{0x10B92, 0x10B98}, {0x10B9D, 0x10BA8}, {0x10BB0, 0x10BFF},
+	{0x10C49, 0x10C7F}, {0x10CB3, 0x10CBF}, {0x10CF3, 0x10CF9},
+	{0x10D00, 0x10E5F}, {0x10E7F, 0x10FFF}, {0x1104E, 0x11051},
+	{0x11070, 0x1107E}, {0x110C2, 0x110CF}, {0x110E9, 0x110EF},
+	{0x110FA, 0x110FF}, {0x11135, 0x11135}, {0x11144, 0x1114F},
+	{0x11177, 0x1117F}, {0x111CE, 0x111CF}, {0x111E0, 0x111E0},
+	{0x111F5, 0x111FF}, {0x11212, 0x11212}, {0x1123F, 0x1127F},
+	{0x11287, 0x11287}, {0x11289, 0x11289}, {0x1128E, 0x1128E},
+	{0x1129E, 0x1129E}, {0x112AA, 0x112AF}, {0x112EB, 0x112EF},
+	{0x112FA, 0x112FF}, {0x11304, 0x11304}, {0x1130D, 0x1130E},
+	{0x11311, 0x11312}, {0x11329, 0x11329}, {0x11331, 0x11331},
+	{0x11334, 0x11334}, {0x1133A, 0x1133B}, {0x11345, 0x11346},
+	{0x11349, 0x1134A}, {0x1134E, 0x1134F}, {0x11351, 0x11356},
+	{0x11358, 0x1135C}, {0x11364, 0x11365}, {0x1136D, 0x1136F},
+	{0x11375, 0x113FF}, {0x1145A, 0x1145A}, {0x1145C, 0x1145C},
+	{0x1145E, 0x1147F}, {0x114C8, 0x114CF}, {0x114DA, 0x1157F},
+	{0x115B6, 0x115B7}, {0x115DE, 0x115FF}, {0x11645, 0x1164F},
+	{0x1165A, 0x1165F}, {0x1166D, 0x1167F}, {0x116B8, 0x116BF},
+	{0x116CA, 0x116FF}, {0x1171A, 0x1171C}, {0x1172C, 0x1172F},
+	{0x11740, 0x1189F}, {0x118F3, 0x118FE}, {0x11900, 0x11ABF},
+	{0x11AF9, 0x11BFF}, {0x11C09, 0x11C09}, {0x11C37, 0x11C37},
+	{0x11C46, 0x11C4F}, {0x11C6D, 0x11C6F}, {0x11C90, 0x11C91},
+	{0x11CA8, 0x11CA8}, {0x11CB7, 0x11FFF}, {0x1239A, 0x123FF},
+	{0x1246F, 0x1246F}, {0x12475, 0x1247F}, {0x12544, 0x12FFF},
+	{0x1342F, 0x143FF}, {0x14647, 0x167FF}, {0x16A39, 0x16A3F},
+	{0x16A5F, 0x16A5F}, {0x16A6A, 0x16A6D}, {0x16A70, 0x16ACF},
+	{0x16AEE, 0x16AEF}, {0x16AF6, 0x16AFF}, {0x16B46, 0x16B4F},
+	{0x16B5A, 0x16B5A}, {0x16B62, 0x16B62}, {0x16B78, 0x16B7C},
+	{0x16B90, 0x16EFF}, {0x16F45, 0x16F4F}, {0x16F7F, 0x16F8E},
+	{0x16FA0, 0x16FDF}, {0x16FE1, 0x16FFF}, {0x187ED, 0x187FF},
+	{0x18AF3, 0x1AFFF}, {0x1B002, 0x1BBFF}, {0x1BC6B, 0x1BC6F},
+	{0x1BC7D, 0x1BC7F}, {0x1BC89, 0x1BC8F}, {0x1BC9A, 0x1BC9B},
+	{0x1BCA4, 0x1CFFF}, {0x1D0F6, 0x1D0FF}, {0x1D127, 0x1D128},
+	{0x1D1E9, 0x1D1FF}, {0x1D246, 0x1D2FF}, {0x1D357, 0x1D35F},
+	{0x1D372, 0x1D3FF}, {0x1D455, 0x1D455}, {0x1D49D, 0x1D49D},
+	{0x1D4A0, 0x1D4A1}, {0x1D4A3, 0x1D4A4}, {0x1D4A7, 0x1D4A8},
+	{0x1D4AD, 0x1D4AD}, {0x1D4BA, 0x1D4BA}, {0x1D4BC, 0x1D4BC},
+	{0x1D4C4, 0x1D4C4}, {0x1D506, 0x1D506}, {0x1D50B, 0x1D50C},
+	{0x1D515, 0x1D515}, {0x1D51D, 0x1D51D}, {0x1D53A, 0x1D53A},
+	{0x1D53F, 0x1D53F}, {0x1D545, 0x1D545}, {0x1D547, 0x1D549},
+	{0x1D551, 0x1D551}, {0x1D6A6, 0x1D6A7}, {0x1D7CC, 0x1D7CD},
+	{0x1DA8C, 0x1DA9A}, {0x1DAA0, 0x1DAA0}, {0x1DAB0, 0x1DFFF},
+	{0x1E007, 0x1E007}, {0x1E019, 0x1E01A}, {0x1E022, 0x1E022},
+	{0x1E025, 0x1E025}, {0x1E02B, 0x1E7FF}, {0x1E8C5, 0x1E8C6},
+	{0x1E8D7, 0x1E8FF}, {0x1E94B, 0x1E94F}, {0x1E95A, 0x1E95D},
+	{0x1E960, 0x1EDFF}, {0x1EE04, 0x1EE04}, {0x1EE20, 0x1EE20},
+	{0x1EE23, 0x1EE23}, {0x1EE25, 0x1EE26}, {0x1EE28, 0x1EE28},
+	{0x1EE33, 0x1EE33}, {0x1EE38, 0x1EE38}, {0x1EE3A, 0x1EE3A},
+	{0x1EE3C, 0x1EE41}, {0x1EE43, 0x1EE46}, {0x1EE48, 0x1EE48},
+	{0x1EE4A, 0x1EE4A}, {0x1EE4C, 0x1EE4C}, {0x1EE50, 0x1EE50},
+	{0x1EE53, 0x1EE53}, {0x1EE55, 0x1EE56}, {0x1EE58, 0x1EE58},
+	{0x1EE5A, 0x1EE5A}, {0x1EE5C, 0x1EE5C}, {0x1EE5E, 0x1EE5E},
+	{0x1EE60, 0x1EE60}, {0x1EE63, 0x1EE63}, {0x1EE65, 0x1EE66},
+	{0x1EE6B, 0x1EE6B}, {0x1EE73, 0x1EE73}, {0x1EE78, 0x1EE78},
+	{0x1EE7D, 0x1EE7D}, {0x1EE7F, 0x1EE7F}, {0x1EE8A, 0x1EE8A},
+	{0x1EE9C, 0x1EEA0}, {0x1EEA4, 0x1EEA4}, {0x1EEAA, 0x1EEAA},
+	{0x1EEBC, 0x1EEEF}, {0x1EEF2, 0x1EFFF}, {0x1F02C, 0x1F02F},
+	{0x1F094, 0x1F09F}, {0x1F0AF, 0x1F0B0}, {0x1F0C0, 0x1F0C0},
+	{0x1F0D0, 0x1F0D0}, {0x1F0F6, 0x1F0FF}, {0x1F10D, 0x1F10F},
+	{0x1F12F, 0x1F12F}, {0x1F16C, 0x1F16F}, {0x1F1AD, 0x1F1E5},
+	{0x1F203, 0x1F20F}, {0x1F23C, 0x1F23F}, {0x1F249, 0x1F24F},
+	{0x1F252, 0x1F2FF}, {0x1F6D3, 0x1F6DF}, {0x1F6ED, 0x1F6EF},
+	{0x1F6F7, 0x1F6FF}, {0x1F774, 0x1F77F}, {0x1F7D5, 0x1F7FF},
+	{0x1F80C, 0x1F80F}, {0x1F848, 0x1F84F}, {0x1F85A, 0x1F85F},
+	{0x1F888, 0x1F88F}, {0x1F8AE, 0x1F90F}, {0x1F91F, 0x1F91F},
+	{0x1F928, 0x1F92F}, {0x1F931, 0x1F932}, {0x1F93F, 0x1F93F},
+	{0x1F94C, 0x1F94F}, {0x1F95F, 0x1F97F}, {0x1F992, 0x1F9BF},
+	{0x1F9C1, 0x1FFFF}, {0x2A6D7, 0x2A6FF}, {0x2B735, 0x2B73F},
+	{0x2B81E, 0x2B81F}, {0x2CEA2, 0x2F7FF}, {0x2FA1E, 0xE0000},
+	{0xE0002, 0xE001F}, {0xE0080, 0xE00FF}, {0xE01F0, 0xEFFFF},
+	{0xFFFFE, 0xFFFFF},
+}
+
+var neutral = table{
+	{0x0000, 0x001F}, {0x007F, 0x007F}, {0x0080, 0x009F},
+	{0x00A0, 0x00A0}, {0x00A9, 0x00A9}, {0x00AB, 0x00AB},
+	{0x00B5, 0x00B5}, {0x00BB, 0x00BB}, {0x00C0, 0x00C5},
+	{0x00C7, 0x00CF}, {0x00D1, 0x00D6}, {0x00D9, 0x00DD},
+	{0x00E2, 0x00E5}, {0x00E7, 0x00E7}, {0x00EB, 0x00EB},
+	{0x00EE, 0x00EF}, {0x00F1, 0x00F1}, {0x00F4, 0x00F6},
+	{0x00FB, 0x00FB}, {0x00FD, 0x00FD}, {0x00FF, 0x00FF},
+	{0x0100, 0x0100}, {0x0102, 0x0110}, {0x0112, 0x0112},
+	{0x0114, 0x011A}, {0x011C, 0x0125}, {0x0128, 0x012A},
+	{0x012C, 0x0130}, {0x0134, 0x0137}, {0x0139, 0x013E},
+	{0x0143, 0x0143}, {0x0145, 0x0147}, {0x014C, 0x014C},
+	{0x014E, 0x0151}, {0x0154, 0x0165}, {0x0168, 0x016A},
+	{0x016C, 0x017F}, {0x0180, 0x01BA}, {0x01BB, 0x01BB},
+	{0x01BC, 0x01BF}, {0x01C0, 0x01C3}, {0x01C4, 0x01CD},
+	{0x01CF, 0x01CF}, {0x01D1, 0x01D1}, {0x01D3, 0x01D3},
+	{0x01D5, 0x01D5}, {0x01D7, 0x01D7}, {0x01D9, 0x01D9},
+	{0x01DB, 0x01DB}, {0x01DD, 0x024F}, {0x0250, 0x0250},
+	{0x0252, 0x0260}, {0x0262, 0x0293}, {0x0294, 0x0294},
+	{0x0295, 0x02AF}, {0x02B0, 0x02C1}, {0x02C2, 0x02C3},
+	{0x02C5, 0x02C5}, {0x02C6, 0x02C6}, {0x02C8, 0x02C8},
+	{0x02CC, 0x02CC}, {0x02CE, 0x02CF}, {0x02D1, 0x02D1},
+	{0x02D2, 0x02D7}, {0x02DC, 0x02DC}, {0x02DE, 0x02DE},
+	{0x02E0, 0x02E4}, {0x02E5, 0x02EB}, {0x02EC, 0x02EC},
+	{0x02ED, 0x02ED}, {0x02EE, 0x02EE}, {0x02EF, 0x02FF},
+	{0x0370, 0x0373}, {0x0374, 0x0374}, {0x0375, 0x0375},
+	{0x0376, 0x0377}, {0x037A, 0x037A}, {0x037B, 0x037D},
+	{0x037E, 0x037E}, {0x037F, 0x037F}, {0x0384, 0x0385},
+	{0x0386, 0x0386}, {0x0387, 0x0387}, {0x0388, 0x038A},
+	{0x038C, 0x038C}, {0x038E, 0x0390}, {0x03AA, 0x03B0},
+	{0x03C2, 0x03C2}, {0x03CA, 0x03F5}, {0x03F6, 0x03F6},
+	{0x03F7, 0x03FF}, {0x0400, 0x0400}, {0x0402, 0x040F},
+	{0x0450, 0x0450}, {0x0452, 0x0481}, {0x0482, 0x0482},
+	{0x0483, 0x0487}, {0x0488, 0x0489}, {0x048A, 0x04FF},
+	{0x0500, 0x052F}, {0x0531, 0x0556}, {0x0559, 0x0559},
+	{0x055A, 0x055F}, {0x0561, 0x0587}, {0x0589, 0x0589},
+	{0x058A, 0x058A}, {0x058D, 0x058E}, {0x058F, 0x058F},
+	{0x0591, 0x05BD}, {0x05BE, 0x05BE}, {0x05BF, 0x05BF},
+	{0x05C0, 0x05C0}, {0x05C1, 0x05C2}, {0x05C3, 0x05C3},
+	{0x05C4, 0x05C5}, {0x05C6, 0x05C6}, {0x05C7, 0x05C7},
+	{0x05D0, 0x05EA}, {0x05F0, 0x05F2}, {0x05F3, 0x05F4},
+	{0x0600, 0x0605}, {0x0606, 0x0608}, {0x0609, 0x060A},
+	{0x060B, 0x060B}, {0x060C, 0x060D}, {0x060E, 0x060F},
+	{0x0610, 0x061A}, {0x061B, 0x061B}, {0x061C, 0x061C},
+	{0x061E, 0x061F}, {0x0620, 0x063F}, {0x0640, 0x0640},
+	{0x0641, 0x064A}, {0x064B, 0x065F}, {0x0660, 0x0669},
+	{0x066A, 0x066D}, {0x066E, 0x066F}, {0x0670, 0x0670},
+	{0x0671, 0x06D3}, {0x06D4, 0x06D4}, {0x06D5, 0x06D5},
+	{0x06D6, 0x06DC}, {0x06DD, 0x06DD}, {0x06DE, 0x06DE},
+	{0x06DF, 0x06E4}, {0x06E5, 0x06E6}, {0x06E7, 0x06E8},
+	{0x06E9, 0x06E9}, {0x06EA, 0x06ED}, {0x06EE, 0x06EF},
+	{0x06F0, 0x06F9}, {0x06FA, 0x06FC}, {0x06FD, 0x06FE},
+	{0x06FF, 0x06FF}, {0x0700, 0x070D}, {0x070F, 0x070F},
+	{0x0710, 0x0710}, {0x0711, 0x0711}, {0x0712, 0x072F},
+	{0x0730, 0x074A}, {0x074D, 0x074F}, {0x0750, 0x077F},
+	{0x0780, 0x07A5}, {0x07A6, 0x07B0}, {0x07B1, 0x07B1},
+	{0x07C0, 0x07C9}, {0x07CA, 0x07EA}, {0x07EB, 0x07F3},
+	{0x07F4, 0x07F5}, {0x07F6, 0x07F6}, {0x07F7, 0x07F9},
+	{0x07FA, 0x07FA}, {0x0800, 0x0815}, {0x0816, 0x0819},
+	{0x081A, 0x081A}, {0x081B, 0x0823}, {0x0824, 0x0824},
+	{0x0825, 0x0827}, {0x0828, 0x0828}, {0x0829, 0x082D},
+	{0x0830, 0x083E}, {0x0840, 0x0858}, {0x0859, 0x085B},
+	{0x085E, 0x085E}, {0x08A0, 0x08B4}, {0x08B6, 0x08BD},
+	{0x08D4, 0x08E1}, {0x08E2, 0x08E2}, {0x08E3, 0x08FF},
+	{0x0900, 0x0902}, {0x0903, 0x0903}, {0x0904, 0x0939},
+	{0x093A, 0x093A}, {0x093B, 0x093B}, {0x093C, 0x093C},
+	{0x093D, 0x093D}, {0x093E, 0x0940}, {0x0941, 0x0948},
+	{0x0949, 0x094C}, {0x094D, 0x094D}, {0x094E, 0x094F},
+	{0x0950, 0x0950}, {0x0951, 0x0957}, {0x0958, 0x0961},
+	{0x0962, 0x0963}, {0x0964, 0x0965}, {0x0966, 0x096F},
+	{0x0970, 0x0970}, {0x0971, 0x0971}, {0x0972, 0x097F},
+	{0x0980, 0x0980}, {0x0981, 0x0981}, {0x0982, 0x0983},
+	{0x0985, 0x098C}, {0x098F, 0x0990}, {0x0993, 0x09A8},
+	{0x09AA, 0x09B0}, {0x09B2, 0x09B2}, {0x09B6, 0x09B9},
+	{0x09BC, 0x09BC}, {0x09BD, 0x09BD}, {0x09BE, 0x09C0},
+	{0x09C1, 0x09C4}, {0x09C7, 0x09C8}, {0x09CB, 0x09CC},
+	{0x09CD, 0x09CD}, {0x09CE, 0x09CE}, {0x09D7, 0x09D7},
+	{0x09DC, 0x09DD}, {0x09DF, 0x09E1}, {0x09E2, 0x09E3},
+	{0x09E6, 0x09EF}, {0x09F0, 0x09F1}, {0x09F2, 0x09F3},
+	{0x09F4, 0x09F9}, {0x09FA, 0x09FA}, {0x09FB, 0x09FB},
+	{0x0A01, 0x0A02}, {0x0A03, 0x0A03}, {0x0A05, 0x0A0A},
+	{0x0A0F, 0x0A10}, {0x0A13, 0x0A28}, {0x0A2A, 0x0A30},
+	{0x0A32, 0x0A33}, {0x0A35, 0x0A36}, {0x0A38, 0x0A39},
+	{0x0A3C, 0x0A3C}, {0x0A3E, 0x0A40}, {0x0A41, 0x0A42},
+	{0x0A47, 0x0A48}, {0x0A4B, 0x0A4D}, {0x0A51, 0x0A51},
+	{0x0A59, 0x0A5C}, {0x0A5E, 0x0A5E}, {0x0A66, 0x0A6F},
+	{0x0A70, 0x0A71}, {0x0A72, 0x0A74}, {0x0A75, 0x0A75},
+	{0x0A81, 0x0A82}, {0x0A83, 0x0A83}, {0x0A85, 0x0A8D},
+	{0x0A8F, 0x0A91}, {0x0A93, 0x0AA8}, {0x0AAA, 0x0AB0},
+	{0x0AB2, 0x0AB3}, {0x0AB5, 0x0AB9}, {0x0ABC, 0x0ABC},
+	{0x0ABD, 0x0ABD}, {0x0ABE, 0x0AC0}, {0x0AC1, 0x0AC5},
+	{0x0AC7, 0x0AC8}, {0x0AC9, 0x0AC9}, {0x0ACB, 0x0ACC},
+	{0x0ACD, 0x0ACD}, {0x0AD0, 0x0AD0}, {0x0AE0, 0x0AE1},
+	{0x0AE2, 0x0AE3}, {0x0AE6, 0x0AEF}, {0x0AF0, 0x0AF0},
+	{0x0AF1, 0x0AF1}, {0x0AF9, 0x0AF9}, {0x0B01, 0x0B01},
+	{0x0B02, 0x0B03}, {0x0B05, 0x0B0C}, {0x0B0F, 0x0B10},
+	{0x0B13, 0x0B28}, {0x0B2A, 0x0B30}, {0x0B32, 0x0B33},
+	{0x0B35, 0x0B39}, {0x0B3C, 0x0B3C}, {0x0B3D, 0x0B3D},
+	{0x0B3E, 0x0B3E}, {0x0B3F, 0x0B3F}, {0x0B40, 0x0B40},
+	{0x0B41, 0x0B44}, {0x0B47, 0x0B48}, {0x0B4B, 0x0B4C},
+	{0x0B4D, 0x0B4D}, {0x0B56, 0x0B56}, {0x0B57, 0x0B57},
+	{0x0B5C, 0x0B5D}, {0x0B5F, 0x0B61}, {0x0B62, 0x0B63},
+	{0x0B66, 0x0B6F}, {0x0B70, 0x0B70}, {0x0B71, 0x0B71},
+	{0x0B72, 0x0B77}, {0x0B82, 0x0B82}, {0x0B83, 0x0B83},
+	{0x0B85, 0x0B8A}, {0x0B8E, 0x0B90}, {0x0B92, 0x0B95},
+	{0x0B99, 0x0B9A}, {0x0B9C, 0x0B9C}, {0x0B9E, 0x0B9F},
+	{0x0BA3, 0x0BA4}, {0x0BA8, 0x0BAA}, {0x0BAE, 0x0BB9},
+	{0x0BBE, 0x0BBF}, {0x0BC0, 0x0BC0}, {0x0BC1, 0x0BC2},
+	{0x0BC6, 0x0BC8}, {0x0BCA, 0x0BCC}, {0x0BCD, 0x0BCD},
+	{0x0BD0, 0x0BD0}, {0x0BD7, 0x0BD7}, {0x0BE6, 0x0BEF},
+	{0x0BF0, 0x0BF2}, {0x0BF3, 0x0BF8}, {0x0BF9, 0x0BF9},
+	{0x0BFA, 0x0BFA}, {0x0C00, 0x0C00}, {0x0C01, 0x0C03},
+	{0x0C05, 0x0C0C}, {0x0C0E, 0x0C10}, {0x0C12, 0x0C28},
+	{0x0C2A, 0x0C39}, {0x0C3D, 0x0C3D}, {0x0C3E, 0x0C40},
+	{0x0C41, 0x0C44}, {0x0C46, 0x0C48}, {0x0C4A, 0x0C4D},
+	{0x0C55, 0x0C56}, {0x0C58, 0x0C5A}, {0x0C60, 0x0C61},
+	{0x0C62, 0x0C63}, {0x0C66, 0x0C6F}, {0x0C78, 0x0C7E},
+	{0x0C7F, 0x0C7F}, {0x0C80, 0x0C80}, {0x0C81, 0x0C81},
+	{0x0C82, 0x0C83}, {0x0C85, 0x0C8C}, {0x0C8E, 0x0C90},
+	{0x0C92, 0x0CA8}, {0x0CAA, 0x0CB3}, {0x0CB5, 0x0CB9},
+	{0x0CBC, 0x0CBC}, {0x0CBD, 0x0CBD}, {0x0CBE, 0x0CBE},
+	{0x0CBF, 0x0CBF}, {0x0CC0, 0x0CC4}, {0x0CC6, 0x0CC6},
+	{0x0CC7, 0x0CC8}, {0x0CCA, 0x0CCB}, {0x0CCC, 0x0CCD},
+	{0x0CD5, 0x0CD6}, {0x0CDE, 0x0CDE}, {0x0CE0, 0x0CE1},
+	{0x0CE2, 0x0CE3}, {0x0CE6, 0x0CEF}, {0x0CF1, 0x0CF2},
+	{0x0D01, 0x0D01}, {0x0D02, 0x0D03}, {0x0D05, 0x0D0C},
+	{0x0D0E, 0x0D10}, {0x0D12, 0x0D3A}, {0x0D3D, 0x0D3D},
+	{0x0D3E, 0x0D40}, {0x0D41, 0x0D44}, {0x0D46, 0x0D48},
+	{0x0D4A, 0x0D4C}, {0x0D4D, 0x0D4D}, {0x0D4E, 0x0D4E},
+	{0x0D4F, 0x0D4F}, {0x0D54, 0x0D56}, {0x0D57, 0x0D57},
+	{0x0D58, 0x0D5E}, {0x0D5F, 0x0D61}, {0x0D62, 0x0D63},
+	{0x0D66, 0x0D6F}, {0x0D70, 0x0D78}, {0x0D79, 0x0D79},
+	{0x0D7A, 0x0D7F}, {0x0D82, 0x0D83}, {0x0D85, 0x0D96},
+	{0x0D9A, 0x0DB1}, {0x0DB3, 0x0DBB}, {0x0DBD, 0x0DBD},
+	{0x0DC0, 0x0DC6}, {0x0DCA, 0x0DCA}, {0x0DCF, 0x0DD1},
+	{0x0DD2, 0x0DD4}, {0x0DD6, 0x0DD6}, {0x0DD8, 0x0DDF},
+	{0x0DE6, 0x0DEF}, {0x0DF2, 0x0DF3}, {0x0DF4, 0x0DF4},
+	{0x0E01, 0x0E30}, {0x0E31, 0x0E31}, {0x0E32, 0x0E33},
+	{0x0E34, 0x0E3A}, {0x0E3F, 0x0E3F}, {0x0E40, 0x0E45},
+	{0x0E46, 0x0E46}, {0x0E47, 0x0E4E}, {0x0E4F, 0x0E4F},
+	{0x0E50, 0x0E59}, {0x0E5A, 0x0E5B}, {0x0E81, 0x0E82},
+	{0x0E84, 0x0E84}, {0x0E87, 0x0E88}, {0x0E8A, 0x0E8A},
+	{0x0E8D, 0x0E8D}, {0x0E94, 0x0E97}, {0x0E99, 0x0E9F},
+	{0x0EA1, 0x0EA3}, {0x0EA5, 0x0EA5}, {0x0EA7, 0x0EA7},
+	{0x0EAA, 0x0EAB}, {0x0EAD, 0x0EB0}, {0x0EB1, 0x0EB1},
+	{0x0EB2, 0x0EB3}, {0x0EB4, 0x0EB9}, {0x0EBB, 0x0EBC},
+	{0x0EBD, 0x0EBD}, {0x0EC0, 0x0EC4}, {0x0EC6, 0x0EC6},
+	{0x0EC8, 0x0ECD}, {0x0ED0, 0x0ED9}, {0x0EDC, 0x0EDF},
+	{0x0F00, 0x0F00}, {0x0F01, 0x0F03}, {0x0F04, 0x0F12},
+	{0x0F13, 0x0F13}, {0x0F14, 0x0F14}, {0x0F15, 0x0F17},
+	{0x0F18, 0x0F19}, {0x0F1A, 0x0F1F}, {0x0F20, 0x0F29},
+	{0x0F2A, 0x0F33}, {0x0F34, 0x0F34}, {0x0F35, 0x0F35},
+	{0x0F36, 0x0F36}, {0x0F37, 0x0F37}, {0x0F38, 0x0F38},
+	{0x0F39, 0x0F39}, {0x0F3A, 0x0F3A}, {0x0F3B, 0x0F3B},
+	{0x0F3C, 0x0F3C}, {0x0F3D, 0x0F3D}, {0x0F3E, 0x0F3F},
+	{0x0F40, 0x0F47}, {0x0F49, 0x0F6C}, {0x0F71, 0x0F7E},
+	{0x0F7F, 0x0F7F}, {0x0F80, 0x0F84}, {0x0F85, 0x0F85},
+	{0x0F86, 0x0F87}, {0x0F88, 0x0F8C}, {0x0F8D, 0x0F97},
+	{0x0F99, 0x0FBC}, {0x0FBE, 0x0FC5}, {0x0FC6, 0x0FC6},
+	{0x0FC7, 0x0FCC}, {0x0FCE, 0x0FCF}, {0x0FD0, 0x0FD4},
+	{0x0FD5, 0x0FD8}, {0x0FD9, 0x0FDA}, {0x1000, 0x102A},
+	{0x102B, 0x102C}, {0x102D, 0x1030}, {0x1031, 0x1031},
+	{0x1032, 0x1037}, {0x1038, 0x1038}, {0x1039, 0x103A},
+	{0x103B, 0x103C}, {0x103D, 0x103E}, {0x103F, 0x103F},
+	{0x1040, 0x1049}, {0x104A, 0x104F}, {0x1050, 0x1055},
+	{0x1056, 0x1057}, {0x1058, 0x1059}, {0x105A, 0x105D},
+	{0x105E, 0x1060}, {0x1061, 0x1061}, {0x1062, 0x1064},
+	{0x1065, 0x1066}, {0x1067, 0x106D}, {0x106E, 0x1070},
+	{0x1071, 0x1074}, {0x1075, 0x1081}, {0x1082, 0x1082},
+	{0x1083, 0x1084}, {0x1085, 0x1086}, {0x1087, 0x108C},
+	{0x108D, 0x108D}, {0x108E, 0x108E}, {0x108F, 0x108F},
+	{0x1090, 0x1099}, {0x109A, 0x109C}, {0x109D, 0x109D},
+	{0x109E, 0x109F}, {0x10A0, 0x10C5}, {0x10C7, 0x10C7},
+	{0x10CD, 0x10CD}, {0x10D0, 0x10FA}, {0x10FB, 0x10FB},
+	{0x10FC, 0x10FC}, {0x10FD, 0x10FF}, {0x1160, 0x11FF},
+	{0x1200, 0x1248}, {0x124A, 0x124D}, {0x1250, 0x1256},
+	{0x1258, 0x1258}, {0x125A, 0x125D}, {0x1260, 0x1288},
+	{0x128A, 0x128D}, {0x1290, 0x12B0}, {0x12B2, 0x12B5},
+	{0x12B8, 0x12BE}, {0x12C0, 0x12C0}, {0x12C2, 0x12C5},
+	{0x12C8, 0x12D6}, {0x12D8, 0x1310}, {0x1312, 0x1315},
+	{0x1318, 0x135A}, {0x135D, 0x135F}, {0x1360, 0x1368},
+	{0x1369, 0x137C}, {0x1380, 0x138F}, {0x1390, 0x1399},
+	{0x13A0, 0x13F5}, {0x13F8, 0x13FD}, {0x1400, 0x1400},
+	{0x1401, 0x166C}, {0x166D, 0x166E}, {0x166F, 0x167F},
+	{0x1680, 0x1680}, {0x1681, 0x169A}, {0x169B, 0x169B},
+	{0x169C, 0x169C}, {0x16A0, 0x16EA}, {0x16EB, 0x16ED},
+	{0x16EE, 0x16F0}, {0x16F1, 0x16F8}, {0x1700, 0x170C},
+	{0x170E, 0x1711}, {0x1712, 0x1714}, {0x1720, 0x1731},
+	{0x1732, 0x1734}, {0x1735, 0x1736}, {0x1740, 0x1751},
+	{0x1752, 0x1753}, {0x1760, 0x176C}, {0x176E, 0x1770},
+	{0x1772, 0x1773}, {0x1780, 0x17B3}, {0x17B4, 0x17B5},
+	{0x17B6, 0x17B6}, {0x17B7, 0x17BD}, {0x17BE, 0x17C5},
+	{0x17C6, 0x17C6}, {0x17C7, 0x17C8}, {0x17C9, 0x17D3},
+	{0x17D4, 0x17D6}, {0x17D7, 0x17D7}, {0x17D8, 0x17DA},
+	{0x17DB, 0x17DB}, {0x17DC, 0x17DC}, {0x17DD, 0x17DD},
+	{0x17E0, 0x17E9}, {0x17F0, 0x17F9}, {0x1800, 0x1805},
+	{0x1806, 0x1806}, {0x1807, 0x180A}, {0x180B, 0x180D},
+	{0x180E, 0x180E}, {0x1810, 0x1819}, {0x1820, 0x1842},
+	{0x1843, 0x1843}, {0x1844, 0x1877}, {0x1880, 0x1884},
+	{0x1885, 0x1886}, {0x1887, 0x18A8}, {0x18A9, 0x18A9},
+	{0x18AA, 0x18AA}, {0x18B0, 0x18F5}, {0x1900, 0x191E},
+	{0x1920, 0x1922}, {0x1923, 0x1926}, {0x1927, 0x1928},
+	{0x1929, 0x192B}, {0x1930, 0x1931}, {0x1932, 0x1932},
+	{0x1933, 0x1938}, {0x1939, 0x193B}, {0x1940, 0x1940},
+	{0x1944, 0x1945}, {0x1946, 0x194F}, {0x1950, 0x196D},
+	{0x1970, 0x1974}, {0x1980, 0x19AB}, {0x19B0, 0x19C9},
+	{0x19D0, 0x19D9}, {0x19DA, 0x19DA}, {0x19DE, 0x19DF},
+	{0x19E0, 0x19FF}, {0x1A00, 0x1A16}, {0x1A17, 0x1A18},
+	{0x1A19, 0x1A1A}, {0x1A1B, 0x1A1B}, {0x1A1E, 0x1A1F},
+	{0x1A20, 0x1A54}, {0x1A55, 0x1A55}, {0x1A56, 0x1A56},
+	{0x1A57, 0x1A57}, {0x1A58, 0x1A5E}, {0x1A60, 0x1A60},
+	{0x1A61, 0x1A61}, {0x1A62, 0x1A62}, {0x1A63, 0x1A64},
+	{0x1A65, 0x1A6C}, {0x1A6D, 0x1A72}, {0x1A73, 0x1A7C},
+	{0x1A7F, 0x1A7F}, {0x1A80, 0x1A89}, {0x1A90, 0x1A99},
+	{0x1AA0, 0x1AA6}, {0x1AA7, 0x1AA7}, {0x1AA8, 0x1AAD},
+	{0x1AB0, 0x1ABD}, {0x1ABE, 0x1ABE}, {0x1B00, 0x1B03},
+	{0x1B04, 0x1B04}, {0x1B05, 0x1B33}, {0x1B34, 0x1B34},
+	{0x1B35, 0x1B35}, {0x1B36, 0x1B3A}, {0x1B3B, 0x1B3B},
+	{0x1B3C, 0x1B3C}, {0x1B3D, 0x1B41}, {0x1B42, 0x1B42},
+	{0x1B43, 0x1B44}, {0x1B45, 0x1B4B}, {0x1B50, 0x1B59},
+	{0x1B5A, 0x1B60}, {0x1B61, 0x1B6A}, {0x1B6B, 0x1B73},
+	{0x1B74, 0x1B7C}, {0x1B80, 0x1B81}, {0x1B82, 0x1B82},
+	{0x1B83, 0x1BA0}, {0x1BA1, 0x1BA1}, {0x1BA2, 0x1BA5},
+	{0x1BA6, 0x1BA7}, {0x1BA8, 0x1BA9}, {0x1BAA, 0x1BAA},
+	{0x1BAB, 0x1BAD}, {0x1BAE, 0x1BAF}, {0x1BB0, 0x1BB9},
+	{0x1BBA, 0x1BBF}, {0x1BC0, 0x1BE5}, {0x1BE6, 0x1BE6},
+	{0x1BE7, 0x1BE7}, {0x1BE8, 0x1BE9}, {0x1BEA, 0x1BEC},
+	{0x1BED, 0x1BED}, {0x1BEE, 0x1BEE}, {0x1BEF, 0x1BF1},
+	{0x1BF2, 0x1BF3}, {0x1BFC, 0x1BFF}, {0x1C00, 0x1C23},
+	{0x1C24, 0x1C2B}, {0x1C2C, 0x1C33}, {0x1C34, 0x1C35},
+	{0x1C36, 0x1C37}, {0x1C3B, 0x1C3F}, {0x1C40, 0x1C49},
+	{0x1C4D, 0x1C4F}, {0x1C50, 0x1C59}, {0x1C5A, 0x1C77},
+	{0x1C78, 0x1C7D}, {0x1C7E, 0x1C7F}, {0x1C80, 0x1C88},
+	{0x1CC0, 0x1CC7}, {0x1CD0, 0x1CD2}, {0x1CD3, 0x1CD3},
+	{0x1CD4, 0x1CE0}, {0x1CE1, 0x1CE1}, {0x1CE2, 0x1CE8},
+	{0x1CE9, 0x1CEC}, {0x1CED, 0x1CED}, {0x1CEE, 0x1CF1},
+	{0x1CF2, 0x1CF3}, {0x1CF4, 0x1CF4}, {0x1CF5, 0x1CF6},
+	{0x1CF8, 0x1CF9}, {0x1D00, 0x1D2B}, {0x1D2C, 0x1D6A},
+	{0x1D6B, 0x1D77}, {0x1D78, 0x1D78}, {0x1D79, 0x1D7F},
+	{0x1D80, 0x1D9A}, {0x1D9B, 0x1DBF}, {0x1DC0, 0x1DF5},
+	{0x1DFB, 0x1DFF}, {0x1E00, 0x1EFF}, {0x1F00, 0x1F15},
+	{0x1F18, 0x1F1D}, {0x1F20, 0x1F45}, {0x1F48, 0x1F4D},
+	{0x1F50, 0x1F57}, {0x1F59, 0x1F59}, {0x1F5B, 0x1F5B},
+	{0x1F5D, 0x1F5D}, {0x1F5F, 0x1F7D}, {0x1F80, 0x1FB4},
+	{0x1FB6, 0x1FBC}, {0x1FBD, 0x1FBD}, {0x1FBE, 0x1FBE},
+	{0x1FBF, 0x1FC1}, {0x1FC2, 0x1FC4}, {0x1FC6, 0x1FCC},
+	{0x1FCD, 0x1FCF}, {0x1FD0, 0x1FD3}, {0x1FD6, 0x1FDB},
+	{0x1FDD, 0x1FDF}, {0x1FE0, 0x1FEC}, {0x1FED, 0x1FEF},
+	{0x1FF2, 0x1FF4}, {0x1FF6, 0x1FFC}, {0x1FFD, 0x1FFE},
+	{0x2000, 0x200A}, {0x200B, 0x200F}, {0x2011, 0x2012},
+	{0x2017, 0x2017}, {0x201A, 0x201A}, {0x201B, 0x201B},
+	{0x201E, 0x201E}, {0x201F, 0x201F}, {0x2023, 0x2023},
+	{0x2028, 0x2028}, {0x2029, 0x2029}, {0x202A, 0x202E},
+	{0x202F, 0x202F}, {0x2031, 0x2031}, {0x2034, 0x2034},
+	{0x2036, 0x2038}, {0x2039, 0x2039}, {0x203A, 0x203A},
+	{0x203C, 0x203D}, {0x203F, 0x2040}, {0x2041, 0x2043},
+	{0x2044, 0x2044}, {0x2045, 0x2045}, {0x2046, 0x2046},
+	{0x2047, 0x2051}, {0x2052, 0x2052}, {0x2053, 0x2053},
+	{0x2054, 0x2054}, {0x2055, 0x205E}, {0x205F, 0x205F},
+	{0x2060, 0x2064}, {0x2066, 0x206F}, {0x2070, 0x2070},
+	{0x2071, 0x2071}, {0x2075, 0x2079}, {0x207A, 0x207C},
+	{0x207D, 0x207D}, {0x207E, 0x207E}, {0x2080, 0x2080},
+	{0x2085, 0x2089}, {0x208A, 0x208C}, {0x208D, 0x208D},
+	{0x208E, 0x208E}, {0x2090, 0x209C}, {0x20A0, 0x20A8},
+	{0x20AA, 0x20AB}, {0x20AD, 0x20BE}, {0x20D0, 0x20DC},
+	{0x20DD, 0x20E0}, {0x20E1, 0x20E1}, {0x20E2, 0x20E4},
+	{0x20E5, 0x20F0}, {0x2100, 0x2101}, {0x2102, 0x2102},
+	{0x2104, 0x2104}, {0x2106, 0x2106}, {0x2107, 0x2107},
+	{0x2108, 0x2108}, {0x210A, 0x2112}, {0x2114, 0x2114},
+	{0x2115, 0x2115}, {0x2117, 0x2117}, {0x2118, 0x2118},
+	{0x2119, 0x211D}, {0x211E, 0x2120}, {0x2123, 0x2123},
+	{0x2124, 0x2124}, {0x2125, 0x2125}, {0x2127, 0x2127},
+	{0x2128, 0x2128}, {0x2129, 0x2129}, {0x212A, 0x212A},
+	{0x212C, 0x212D}, {0x212E, 0x212E}, {0x212F, 0x2134},
+	{0x2135, 0x2138}, {0x2139, 0x2139}, {0x213A, 0x213B},
+	{0x213C, 0x213F}, {0x2140, 0x2144}, {0x2145, 0x2149},
+	{0x214A, 0x214A}, {0x214B, 0x214B}, {0x214C, 0x214D},
+	{0x214E, 0x214E}, {0x214F, 0x214F}, {0x2150, 0x2152},
+	{0x2155, 0x215A}, {0x215F, 0x215F}, {0x216C, 0x216F},
+	{0x217A, 0x2182}, {0x2183, 0x2184}, {0x2185, 0x2188},
+	{0x218A, 0x218B}, {0x219A, 0x219B}, {0x219C, 0x219F},
+	{0x21A0, 0x21A0}, {0x21A1, 0x21A2}, {0x21A3, 0x21A3},
+	{0x21A4, 0x21A5}, {0x21A6, 0x21A6}, {0x21A7, 0x21AD},
+	{0x21AE, 0x21AE}, {0x21AF, 0x21B7}, {0x21BA, 0x21CD},
+	{0x21CE, 0x21CF}, {0x21D0, 0x21D1}, {0x21D3, 0x21D3},
+	{0x21D5, 0x21E6}, {0x21E8, 0x21F3}, {0x21F4, 0x21FF},
+	{0x2201, 0x2201}, {0x2204, 0x2206}, {0x2209, 0x220A},
+	{0x220C, 0x220E}, {0x2210, 0x2210}, {0x2212, 0x2214},
+	{0x2216, 0x2219}, {0x221B, 0x221C}, {0x2221, 0x2222},
+	{0x2224, 0x2224}, {0x2226, 0x2226}, {0x222D, 0x222D},
+	{0x222F, 0x2233}, {0x2238, 0x223B}, {0x223E, 0x2247},
+	{0x2249, 0x224B}, {0x224D, 0x2251}, {0x2253, 0x225F},
+	{0x2262, 0x2263}, {0x2268, 0x2269}, {0x226C, 0x226D},
+	{0x2270, 0x2281}, {0x2284, 0x2285}, {0x2288, 0x2294},
+	{0x2296, 0x2298}, {0x229A, 0x22A4}, {0x22A6, 0x22BE},
+	{0x22C0, 0x22FF}, {0x2300, 0x2307}, {0x2308, 0x2308},
+	{0x2309, 0x2309}, {0x230A, 0x230A}, {0x230B, 0x230B},
+	{0x230C, 0x2311}, {0x2313, 0x2319}, {0x231C, 0x231F},
+	{0x2320, 0x2321}, {0x2322, 0x2328}, {0x232B, 0x237B},
+	{0x237C, 0x237C}, {0x237D, 0x239A}, {0x239B, 0x23B3},
+	{0x23B4, 0x23DB}, {0x23DC, 0x23E1}, {0x23E2, 0x23E8},
+	{0x23ED, 0x23EF}, {0x23F1, 0x23F2}, {0x23F4, 0x23FE},
+	{0x2400, 0x2426}, {0x2440, 0x244A}, {0x24EA, 0x24EA},
+	{0x254C, 0x254F}, {0x2574, 0x257F}, {0x2590, 0x2591},
+	{0x2596, 0x259F}, {0x25A2, 0x25A2}, {0x25AA, 0x25B1},
+	{0x25B4, 0x25B5}, {0x25B8, 0x25BB}, {0x25BE, 0x25BF},
+	{0x25C2, 0x25C5}, {0x25C9, 0x25CA}, {0x25CC, 0x25CD},
+	{0x25D2, 0x25E1}, {0x25E6, 0x25EE}, {0x25F0, 0x25F7},
+	{0x25F8, 0x25FC}, {0x25FF, 0x25FF}, {0x2600, 0x2604},
+	{0x2607, 0x2608}, {0x260A, 0x260D}, {0x2610, 0x2613},
+	{0x2616, 0x261B}, {0x261D, 0x261D}, {0x261F, 0x263F},
+	{0x2641, 0x2641}, {0x2643, 0x2647}, {0x2654, 0x265F},
+	{0x2662, 0x2662}, {0x2666, 0x2666}, {0x266B, 0x266B},
+	{0x266E, 0x266E}, {0x2670, 0x267E}, {0x2680, 0x2692},
+	{0x2694, 0x269D}, {0x26A0, 0x26A0}, {0x26A2, 0x26A9},
+	{0x26AC, 0x26BC}, {0x26C0, 0x26C3}, {0x26E2, 0x26E2},
+	{0x26E4, 0x26E7}, {0x2700, 0x2704}, {0x2706, 0x2709},
+	{0x270C, 0x2727}, {0x2729, 0x273C}, {0x273E, 0x274B},
+	{0x274D, 0x274D}, {0x274F, 0x2752}, {0x2756, 0x2756},
+	{0x2758, 0x2767}, {0x2768, 0x2768}, {0x2769, 0x2769},
+	{0x276A, 0x276A}, {0x276B, 0x276B}, {0x276C, 0x276C},
+	{0x276D, 0x276D}, {0x276E, 0x276E}, {0x276F, 0x276F},
+	{0x2770, 0x2770}, {0x2771, 0x2771}, {0x2772, 0x2772},
+	{0x2773, 0x2773}, {0x2774, 0x2774}, {0x2775, 0x2775},
+	{0x2780, 0x2793}, {0x2794, 0x2794}, {0x2798, 0x27AF},
+	{0x27B1, 0x27BE}, {0x27C0, 0x27C4}, {0x27C5, 0x27C5},
+	{0x27C6, 0x27C6}, {0x27C7, 0x27E5}, {0x27EE, 0x27EE},
+	{0x27EF, 0x27EF}, {0x27F0, 0x27FF}, {0x2800, 0x28FF},
+	{0x2900, 0x297F}, {0x2980, 0x2982}, {0x2983, 0x2983},
+	{0x2984, 0x2984}, {0x2987, 0x2987}, {0x2988, 0x2988},
+	{0x2989, 0x2989}, {0x298A, 0x298A}, {0x298B, 0x298B},
+	{0x298C, 0x298C}, {0x298D, 0x298D}, {0x298E, 0x298E},
+	{0x298F, 0x298F}, {0x2990, 0x2990}, {0x2991, 0x2991},
+	{0x2992, 0x2992}, {0x2993, 0x2993}, {0x2994, 0x2994},
+	{0x2995, 0x2995}, {0x2996, 0x2996}, {0x2997, 0x2997},
+	{0x2998, 0x2998}, {0x2999, 0x29D7}, {0x29D8, 0x29D8},
+	{0x29D9, 0x29D9}, {0x29DA, 0x29DA}, {0x29DB, 0x29DB},
+	{0x29DC, 0x29FB}, {0x29FC, 0x29FC}, {0x29FD, 0x29FD},
+	{0x29FE, 0x29FF}, {0x2A00, 0x2AFF}, {0x2B00, 0x2B1A},
+	{0x2B1D, 0x2B2F}, {0x2B30, 0x2B44}, {0x2B45, 0x2B46},
+	{0x2B47, 0x2B4C}, {0x2B4D, 0x2B4F}, {0x2B51, 0x2B54},
+	{0x2B5A, 0x2B73}, {0x2B76, 0x2B95}, {0x2B98, 0x2BB9},
+	{0x2BBD, 0x2BC8}, {0x2BCA, 0x2BD1}, {0x2BEC, 0x2BEF},
+	{0x2C00, 0x2C2E}, {0x2C30, 0x2C5E}, {0x2C60, 0x2C7B},
+	{0x2C7C, 0x2C7D}, {0x2C7E, 0x2C7F}, {0x2C80, 0x2CE4},
+	{0x2CE5, 0x2CEA}, {0x2CEB, 0x2CEE}, {0x2CEF, 0x2CF1},
+	{0x2CF2, 0x2CF3}, {0x2CF9, 0x2CFC}, {0x2CFD, 0x2CFD},
+	{0x2CFE, 0x2CFF}, {0x2D00, 0x2D25}, {0x2D27, 0x2D27},
+	{0x2D2D, 0x2D2D}, {0x2D30, 0x2D67}, {0x2D6F, 0x2D6F},
+	{0x2D70, 0x2D70}, {0x2D7F, 0x2D7F}, {0x2D80, 0x2D96},
+	{0x2DA0, 0x2DA6}, {0x2DA8, 0x2DAE}, {0x2DB0, 0x2DB6},
+	{0x2DB8, 0x2DBE}, {0x2DC0, 0x2DC6}, {0x2DC8, 0x2DCE},
+	{0x2DD0, 0x2DD6}, {0x2DD8, 0x2DDE}, {0x2DE0, 0x2DFF},
+	{0x2E00, 0x2E01}, {0x2E02, 0x2E02}, {0x2E03, 0x2E03},
+	{0x2E04, 0x2E04}, {0x2E05, 0x2E05}, {0x2E06, 0x2E08},
+	{0x2E09, 0x2E09}, {0x2E0A, 0x2E0A}, {0x2E0B, 0x2E0B},
+	{0x2E0C, 0x2E0C}, {0x2E0D, 0x2E0D}, {0x2E0E, 0x2E16},
+	{0x2E17, 0x2E17}, {0x2E18, 0x2E19}, {0x2E1A, 0x2E1A},
+	{0x2E1B, 0x2E1B}, {0x2E1C, 0x2E1C}, {0x2E1D, 0x2E1D},
+	{0x2E1E, 0x2E1F}, {0x2E20, 0x2E20}, {0x2E21, 0x2E21},
+	{0x2E22, 0x2E22}, {0x2E23, 0x2E23}, {0x2E24, 0x2E24},
+	{0x2E25, 0x2E25}, {0x2E26, 0x2E26}, {0x2E27, 0x2E27},
+	{0x2E28, 0x2E28}, {0x2E29, 0x2E29}, {0x2E2A, 0x2E2E},
+	{0x2E2F, 0x2E2F}, {0x2E30, 0x2E39}, {0x2E3A, 0x2E3B},
+	{0x2E3C, 0x2E3F}, {0x2E40, 0x2E40}, {0x2E41, 0x2E41},
+	{0x2E42, 0x2E42}, {0x2E43, 0x2E44}, {0x303F, 0x303F},
+	{0x4DC0, 0x4DFF}, {0xA4D0, 0xA4F7}, {0xA4F8, 0xA4FD},
+	{0xA4FE, 0xA4FF}, {0xA500, 0xA60B}, {0xA60C, 0xA60C},
+	{0xA60D, 0xA60F}, {0xA610, 0xA61F}, {0xA620, 0xA629},
+	{0xA62A, 0xA62B}, {0xA640, 0xA66D}, {0xA66E, 0xA66E},
+	{0xA66F, 0xA66F}, {0xA670, 0xA672}, {0xA673, 0xA673},
+	{0xA674, 0xA67D}, {0xA67E, 0xA67E}, {0xA67F, 0xA67F},
+	{0xA680, 0xA69B}, {0xA69C, 0xA69D}, {0xA69E, 0xA69F},
+	{0xA6A0, 0xA6E5}, {0xA6E6, 0xA6EF}, {0xA6F0, 0xA6F1},
+	{0xA6F2, 0xA6F7}, {0xA700, 0xA716}, {0xA717, 0xA71F},
+	{0xA720, 0xA721}, {0xA722, 0xA76F}, {0xA770, 0xA770},
+	{0xA771, 0xA787}, {0xA788, 0xA788}, {0xA789, 0xA78A},
+	{0xA78B, 0xA78E}, {0xA78F, 0xA78F}, {0xA790, 0xA7AE},
+	{0xA7B0, 0xA7B7}, {0xA7F7, 0xA7F7}, {0xA7F8, 0xA7F9},
+	{0xA7FA, 0xA7FA}, {0xA7FB, 0xA7FF}, {0xA800, 0xA801},
+	{0xA802, 0xA802}, {0xA803, 0xA805}, {0xA806, 0xA806},
+	{0xA807, 0xA80A}, {0xA80B, 0xA80B}, {0xA80C, 0xA822},
+	{0xA823, 0xA824}, {0xA825, 0xA826}, {0xA827, 0xA827},
+	{0xA828, 0xA82B}, {0xA830, 0xA835}, {0xA836, 0xA837},
+	{0xA838, 0xA838}, {0xA839, 0xA839}, {0xA840, 0xA873},
+	{0xA874, 0xA877}, {0xA880, 0xA881}, {0xA882, 0xA8B3},
+	{0xA8B4, 0xA8C3}, {0xA8C4, 0xA8C5}, {0xA8CE, 0xA8CF},
+	{0xA8D0, 0xA8D9}, {0xA8E0, 0xA8F1}, {0xA8F2, 0xA8F7},
+	{0xA8F8, 0xA8FA}, {0xA8FB, 0xA8FB}, {0xA8FC, 0xA8FC},
+	{0xA8FD, 0xA8FD}, {0xA900, 0xA909}, {0xA90A, 0xA925},
+	{0xA926, 0xA92D}, {0xA92E, 0xA92F}, {0xA930, 0xA946},
+	{0xA947, 0xA951}, {0xA952, 0xA953}, {0xA95F, 0xA95F},
+	{0xA980, 0xA982}, {0xA983, 0xA983}, {0xA984, 0xA9B2},
+	{0xA9B3, 0xA9B3}, {0xA9B4, 0xA9B5}, {0xA9B6, 0xA9B9},
+	{0xA9BA, 0xA9BB}, {0xA9BC, 0xA9BC}, {0xA9BD, 0xA9C0},
+	{0xA9C1, 0xA9CD}, {0xA9CF, 0xA9CF}, {0xA9D0, 0xA9D9},
+	{0xA9DE, 0xA9DF}, {0xA9E0, 0xA9E4}, {0xA9E5, 0xA9E5},
+	{0xA9E6, 0xA9E6}, {0xA9E7, 0xA9EF}, {0xA9F0, 0xA9F9},
+	{0xA9FA, 0xA9FE}, {0xAA00, 0xAA28}, {0xAA29, 0xAA2E},
+	{0xAA2F, 0xAA30}, {0xAA31, 0xAA32}, {0xAA33, 0xAA34},
+	{0xAA35, 0xAA36}, {0xAA40, 0xAA42}, {0xAA43, 0xAA43},
+	{0xAA44, 0xAA4B}, {0xAA4C, 0xAA4C}, {0xAA4D, 0xAA4D},
+	{0xAA50, 0xAA59}, {0xAA5C, 0xAA5F}, {0xAA60, 0xAA6F},
+	{0xAA70, 0xAA70}, {0xAA71, 0xAA76}, {0xAA77, 0xAA79},
+	{0xAA7A, 0xAA7A}, {0xAA7B, 0xAA7B}, {0xAA7C, 0xAA7C},
+	{0xAA7D, 0xAA7D}, {0xAA7E, 0xAA7F}, {0xAA80, 0xAAAF},
+	{0xAAB0, 0xAAB0}, {0xAAB1, 0xAAB1}, {0xAAB2, 0xAAB4},
+	{0xAAB5, 0xAAB6}, {0xAAB7, 0xAAB8}, {0xAAB9, 0xAABD},
+	{0xAABE, 0xAABF}, {0xAAC0, 0xAAC0}, {0xAAC1, 0xAAC1},
+	{0xAAC2, 0xAAC2}, {0xAADB, 0xAADC}, {0xAADD, 0xAADD},
+	{0xAADE, 0xAADF}, {0xAAE0, 0xAAEA}, {0xAAEB, 0xAAEB},
+	{0xAAEC, 0xAAED}, {0xAAEE, 0xAAEF}, {0xAAF0, 0xAAF1},
+	{0xAAF2, 0xAAF2}, {0xAAF3, 0xAAF4}, {0xAAF5, 0xAAF5},
+	{0xAAF6, 0xAAF6}, {0xAB01, 0xAB06}, {0xAB09, 0xAB0E},
+	{0xAB11, 0xAB16}, {0xAB20, 0xAB26}, {0xAB28, 0xAB2E},
+	{0xAB30, 0xAB5A}, {0xAB5B, 0xAB5B}, {0xAB5C, 0xAB5F},
+	{0xAB60, 0xAB65}, {0xAB70, 0xABBF}, {0xABC0, 0xABE2},
+	{0xABE3, 0xABE4}, {0xABE5, 0xABE5}, {0xABE6, 0xABE7},
+	{0xABE8, 0xABE8}, {0xABE9, 0xABEA}, {0xABEB, 0xABEB},
+	{0xABEC, 0xABEC}, {0xABED, 0xABED}, {0xABF0, 0xABF9},
+	{0xD7B0, 0xD7C6}, {0xD7CB, 0xD7FB}, {0xD800, 0xDB7F},
+	{0xDB80, 0xDBFF}, {0xDC00, 0xDFFF}, {0xFB00, 0xFB06},
+	{0xFB13, 0xFB17}, {0xFB1D, 0xFB1D}, {0xFB1E, 0xFB1E},
+	{0xFB1F, 0xFB28}, {0xFB29, 0xFB29}, {0xFB2A, 0xFB36},
+	{0xFB38, 0xFB3C}, {0xFB3E, 0xFB3E}, {0xFB40, 0xFB41},
+	{0xFB43, 0xFB44}, {0xFB46, 0xFB4F}, {0xFB50, 0xFBB1},
+	{0xFBB2, 0xFBC1}, {0xFBD3, 0xFD3D}, {0xFD3E, 0xFD3E},
+	{0xFD3F, 0xFD3F}, {0xFD50, 0xFD8F}, {0xFD92, 0xFDC7},
+	{0xFDF0, 0xFDFB}, {0xFDFC, 0xFDFC}, {0xFDFD, 0xFDFD},
+	{0xFE20, 0xFE2F}, {0xFE70, 0xFE74}, {0xFE76, 0xFEFC},
+	{0xFEFF, 0xFEFF}, {0xFFF9, 0xFFFB}, {0xFFFC, 0xFFFC},
+	{0x10000, 0x1000B}, {0x1000D, 0x10026}, {0x10028, 0x1003A},
+	{0x1003C, 0x1003D}, {0x1003F, 0x1004D}, {0x10050, 0x1005D},
+	{0x10080, 0x100FA}, {0x10100, 0x10102}, {0x10107, 0x10133},
+	{0x10137, 0x1013F}, {0x10140, 0x10174}, {0x10175, 0x10178},
+	{0x10179, 0x10189}, {0x1018A, 0x1018B}, {0x1018C, 0x1018E},
+	{0x10190, 0x1019B}, {0x101A0, 0x101A0}, {0x101D0, 0x101FC},
+	{0x101FD, 0x101FD}, {0x10280, 0x1029C}, {0x102A0, 0x102D0},
+	{0x102E0, 0x102E0}, {0x102E1, 0x102FB}, {0x10300, 0x1031F},
+	{0x10320, 0x10323}, {0x10330, 0x10340}, {0x10341, 0x10341},
+	{0x10342, 0x10349}, {0x1034A, 0x1034A}, {0x10350, 0x10375},
+	{0x10376, 0x1037A}, {0x10380, 0x1039D}, {0x1039F, 0x1039F},
+	{0x103A0, 0x103C3}, {0x103C8, 0x103CF}, {0x103D0, 0x103D0},
+	{0x103D1, 0x103D5}, {0x10400, 0x1044F}, {0x10450, 0x1047F},
+	{0x10480, 0x1049D}, {0x104A0, 0x104A9}, {0x104B0, 0x104D3},
+	{0x104D8, 0x104FB}, {0x10500, 0x10527}, {0x10530, 0x10563},
+	{0x1056F, 0x1056F}, {0x10600, 0x10736}, {0x10740, 0x10755},
+	{0x10760, 0x10767}, {0x10800, 0x10805}, {0x10808, 0x10808},
+	{0x1080A, 0x10835}, {0x10837, 0x10838}, {0x1083C, 0x1083C},
+	{0x1083F, 0x1083F}, {0x10840, 0x10855}, {0x10857, 0x10857},
+	{0x10858, 0x1085F}, {0x10860, 0x10876}, {0x10877, 0x10878},
+	{0x10879, 0x1087F}, {0x10880, 0x1089E}, {0x108A7, 0x108AF},
+	{0x108E0, 0x108F2}, {0x108F4, 0x108F5}, {0x108FB, 0x108FF},
+	{0x10900, 0x10915}, {0x10916, 0x1091B}, {0x1091F, 0x1091F},
+	{0x10920, 0x10939}, {0x1093F, 0x1093F}, {0x10980, 0x1099F},
+	{0x109A0, 0x109B7}, {0x109BC, 0x109BD}, {0x109BE, 0x109BF},
+	{0x109C0, 0x109CF}, {0x109D2, 0x109FF}, {0x10A00, 0x10A00},
+	{0x10A01, 0x10A03}, {0x10A05, 0x10A06}, {0x10A0C, 0x10A0F},
+	{0x10A10, 0x10A13}, {0x10A15, 0x10A17}, {0x10A19, 0x10A33},
+	{0x10A38, 0x10A3A}, {0x10A3F, 0x10A3F}, {0x10A40, 0x10A47},
+	{0x10A50, 0x10A58}, {0x10A60, 0x10A7C}, {0x10A7D, 0x10A7E},
+	{0x10A7F, 0x10A7F}, {0x10A80, 0x10A9C}, {0x10A9D, 0x10A9F},
+	{0x10AC0, 0x10AC7}, {0x10AC8, 0x10AC8}, {0x10AC9, 0x10AE4},
+	{0x10AE5, 0x10AE6}, {0x10AEB, 0x10AEF}, {0x10AF0, 0x10AF6},
+	{0x10B00, 0x10B35}, {0x10B39, 0x10B3F}, {0x10B40, 0x10B55},
+	{0x10B58, 0x10B5F}, {0x10B60, 0x10B72}, {0x10B78, 0x10B7F},
+	{0x10B80, 0x10B91}, {0x10B99, 0x10B9C}, {0x10BA9, 0x10BAF},
+	{0x10C00, 0x10C48}, {0x10C80, 0x10CB2}, {0x10CC0, 0x10CF2},
+	{0x10CFA, 0x10CFF}, {0x10E60, 0x10E7E}, {0x11000, 0x11000},
+	{0x11001, 0x11001}, {0x11002, 0x11002}, {0x11003, 0x11037},
+	{0x11038, 0x11046}, {0x11047, 0x1104D}, {0x11052, 0x11065},
+	{0x11066, 0x1106F}, {0x1107F, 0x1107F}, {0x11080, 0x11081},
+	{0x11082, 0x11082}, {0x11083, 0x110AF}, {0x110B0, 0x110B2},
+	{0x110B3, 0x110B6}, {0x110B7, 0x110B8}, {0x110B9, 0x110BA},
+	{0x110BB, 0x110BC}, {0x110BD, 0x110BD}, {0x110BE, 0x110C1},
+	{0x110D0, 0x110E8}, {0x110F0, 0x110F9}, {0x11100, 0x11102},
+	{0x11103, 0x11126}, {0x11127, 0x1112B}, {0x1112C, 0x1112C},
+	{0x1112D, 0x11134}, {0x11136, 0x1113F}, {0x11140, 0x11143},
+	{0x11150, 0x11172}, {0x11173, 0x11173}, {0x11174, 0x11175},
+	{0x11176, 0x11176}, {0x11180, 0x11181}, {0x11182, 0x11182},
+	{0x11183, 0x111B2}, {0x111B3, 0x111B5}, {0x111B6, 0x111BE},
+	{0x111BF, 0x111C0}, {0x111C1, 0x111C4}, {0x111C5, 0x111C9},
+	{0x111CA, 0x111CC}, {0x111CD, 0x111CD}, {0x111D0, 0x111D9},
+	{0x111DA, 0x111DA}, {0x111DB, 0x111DB}, {0x111DC, 0x111DC},
+	{0x111DD, 0x111DF}, {0x111E1, 0x111F4}, {0x11200, 0x11211},
+	{0x11213, 0x1122B}, {0x1122C, 0x1122E}, {0x1122F, 0x11231},
+	{0x11232, 0x11233}, {0x11234, 0x11234}, {0x11235, 0x11235},
+	{0x11236, 0x11237}, {0x11238, 0x1123D}, {0x1123E, 0x1123E},
+	{0x11280, 0x11286}, {0x11288, 0x11288}, {0x1128A, 0x1128D},
+	{0x1128F, 0x1129D}, {0x1129F, 0x112A8}, {0x112A9, 0x112A9},
+	{0x112B0, 0x112DE}, {0x112DF, 0x112DF}, {0x112E0, 0x112E2},
+	{0x112E3, 0x112EA}, {0x112F0, 0x112F9}, {0x11300, 0x11301},
+	{0x11302, 0x11303}, {0x11305, 0x1130C}, {0x1130F, 0x11310},
+	{0x11313, 0x11328}, {0x1132A, 0x11330}, {0x11332, 0x11333},
+	{0x11335, 0x11339}, {0x1133C, 0x1133C}, {0x1133D, 0x1133D},
+	{0x1133E, 0x1133F}, {0x11340, 0x11340}, {0x11341, 0x11344},
+	{0x11347, 0x11348}, {0x1134B, 0x1134D}, {0x11350, 0x11350},
+	{0x11357, 0x11357}, {0x1135D, 0x11361}, {0x11362, 0x11363},
+	{0x11366, 0x1136C}, {0x11370, 0x11374}, {0x11400, 0x11434},
+	{0x11435, 0x11437}, {0x11438, 0x1143F}, {0x11440, 0x11441},
+	{0x11442, 0x11444}, {0x11445, 0x11445}, {0x11446, 0x11446},
+	{0x11447, 0x1144A}, {0x1144B, 0x1144F}, {0x11450, 0x11459},
+	{0x1145B, 0x1145B}, {0x1145D, 0x1145D}, {0x11480, 0x114AF},
+	{0x114B0, 0x114B2}, {0x114B3, 0x114B8}, {0x114B9, 0x114B9},
+	{0x114BA, 0x114BA}, {0x114BB, 0x114BE}, {0x114BF, 0x114C0},
+	{0x114C1, 0x114C1}, {0x114C2, 0x114C3}, {0x114C4, 0x114C5},
+	{0x114C6, 0x114C6}, {0x114C7, 0x114C7}, {0x114D0, 0x114D9},
+	{0x11580, 0x115AE}, {0x115AF, 0x115B1}, {0x115B2, 0x115B5},
+	{0x115B8, 0x115BB}, {0x115BC, 0x115BD}, {0x115BE, 0x115BE},
+	{0x115BF, 0x115C0}, {0x115C1, 0x115D7}, {0x115D8, 0x115DB},
+	{0x115DC, 0x115DD}, {0x11600, 0x1162F}, {0x11630, 0x11632},
+	{0x11633, 0x1163A}, {0x1163B, 0x1163C}, {0x1163D, 0x1163D},
+	{0x1163E, 0x1163E}, {0x1163F, 0x11640}, {0x11641, 0x11643},
+	{0x11644, 0x11644}, {0x11650, 0x11659}, {0x11660, 0x1166C},
+	{0x11680, 0x116AA}, {0x116AB, 0x116AB}, {0x116AC, 0x116AC},
+	{0x116AD, 0x116AD}, {0x116AE, 0x116AF}, {0x116B0, 0x116B5},
+	{0x116B6, 0x116B6}, {0x116B7, 0x116B7}, {0x116C0, 0x116C9},
+	{0x11700, 0x11719}, {0x1171D, 0x1171F}, {0x11720, 0x11721},
+	{0x11722, 0x11725}, {0x11726, 0x11726}, {0x11727, 0x1172B},
+	{0x11730, 0x11739}, {0x1173A, 0x1173B}, {0x1173C, 0x1173E},
+	{0x1173F, 0x1173F}, {0x118A0, 0x118DF}, {0x118E0, 0x118E9},
+	{0x118EA, 0x118F2}, {0x118FF, 0x118FF}, {0x11AC0, 0x11AF8},
+	{0x11C00, 0x11C08}, {0x11C0A, 0x11C2E}, {0x11C2F, 0x11C2F},
+	{0x11C30, 0x11C36}, {0x11C38, 0x11C3D}, {0x11C3E, 0x11C3E},
+	{0x11C3F, 0x11C3F}, {0x11C40, 0x11C40}, {0x11C41, 0x11C45},
+	{0x11C50, 0x11C59}, {0x11C5A, 0x11C6C}, {0x11C70, 0x11C71},
+	{0x11C72, 0x11C8F}, {0x11C92, 0x11CA7}, {0x11CA9, 0x11CA9},
+	{0x11CAA, 0x11CB0}, {0x11CB1, 0x11CB1}, {0x11CB2, 0x11CB3},
+	{0x11CB4, 0x11CB4}, {0x11CB5, 0x11CB6}, {0x12000, 0x12399},
+	{0x12400, 0x1246E}, {0x12470, 0x12474}, {0x12480, 0x12543},
+	{0x13000, 0x1342E}, {0x14400, 0x14646}, {0x16800, 0x16A38},
+	{0x16A40, 0x16A5E}, {0x16A60, 0x16A69}, {0x16A6E, 0x16A6F},
+	{0x16AD0, 0x16AED}, {0x16AF0, 0x16AF4}, {0x16AF5, 0x16AF5},
+	{0x16B00, 0x16B2F}, {0x16B30, 0x16B36}, {0x16B37, 0x16B3B},
+	{0x16B3C, 0x16B3F}, {0x16B40, 0x16B43}, {0x16B44, 0x16B44},
+	{0x16B45, 0x16B45}, {0x16B50, 0x16B59}, {0x16B5B, 0x16B61},
+	{0x16B63, 0x16B77}, {0x16B7D, 0x16B8F}, {0x16F00, 0x16F44},
+	{0x16F50, 0x16F50}, {0x16F51, 0x16F7E}, {0x16F8F, 0x16F92},
+	{0x16F93, 0x16F9F}, {0x1BC00, 0x1BC6A}, {0x1BC70, 0x1BC7C},
+	{0x1BC80, 0x1BC88}, {0x1BC90, 0x1BC99}, {0x1BC9C, 0x1BC9C},
+	{0x1BC9D, 0x1BC9E}, {0x1BC9F, 0x1BC9F}, {0x1BCA0, 0x1BCA3},
+	{0x1D000, 0x1D0F5}, {0x1D100, 0x1D126}, {0x1D129, 0x1D164},
+	{0x1D165, 0x1D166}, {0x1D167, 0x1D169}, {0x1D16A, 0x1D16C},
+	{0x1D16D, 0x1D172}, {0x1D173, 0x1D17A}, {0x1D17B, 0x1D182},
+	{0x1D183, 0x1D184}, {0x1D185, 0x1D18B}, {0x1D18C, 0x1D1A9},
+	{0x1D1AA, 0x1D1AD}, {0x1D1AE, 0x1D1E8}, {0x1D200, 0x1D241},
+	{0x1D242, 0x1D244}, {0x1D245, 0x1D245}, {0x1D300, 0x1D356},
+	{0x1D360, 0x1D371}, {0x1D400, 0x1D454}, {0x1D456, 0x1D49C},
+	{0x1D49E, 0x1D49F}, {0x1D4A2, 0x1D4A2}, {0x1D4A5, 0x1D4A6},
+	{0x1D4A9, 0x1D4AC}, {0x1D4AE, 0x1D4B9}, {0x1D4BB, 0x1D4BB},
+	{0x1D4BD, 0x1D4C3}, {0x1D4C5, 0x1D505}, {0x1D507, 0x1D50A},
+	{0x1D50D, 0x1D514}, {0x1D516, 0x1D51C}, {0x1D51E, 0x1D539},
+	{0x1D53B, 0x1D53E}, {0x1D540, 0x1D544}, {0x1D546, 0x1D546},
+	{0x1D54A, 0x1D550}, {0x1D552, 0x1D6A5}, {0x1D6A8, 0x1D6C0},
+	{0x1D6C1, 0x1D6C1}, {0x1D6C2, 0x1D6DA}, {0x1D6DB, 0x1D6DB},
+	{0x1D6DC, 0x1D6FA}, {0x1D6FB, 0x1D6FB}, {0x1D6FC, 0x1D714},
+	{0x1D715, 0x1D715}, {0x1D716, 0x1D734}, {0x1D735, 0x1D735},
+	{0x1D736, 0x1D74E}, {0x1D74F, 0x1D74F}, {0x1D750, 0x1D76E},
+	{0x1D76F, 0x1D76F}, {0x1D770, 0x1D788}, {0x1D789, 0x1D789},
+	{0x1D78A, 0x1D7A8}, {0x1D7A9, 0x1D7A9}, {0x1D7AA, 0x1D7C2},
+	{0x1D7C3, 0x1D7C3}, {0x1D7C4, 0x1D7CB}, {0x1D7CE, 0x1D7FF},
+	{0x1D800, 0x1D9FF}, {0x1DA00, 0x1DA36}, {0x1DA37, 0x1DA3A},
+	{0x1DA3B, 0x1DA6C}, {0x1DA6D, 0x1DA74}, {0x1DA75, 0x1DA75},
+	{0x1DA76, 0x1DA83}, {0x1DA84, 0x1DA84}, {0x1DA85, 0x1DA86},
+	{0x1DA87, 0x1DA8B}, {0x1DA9B, 0x1DA9F}, {0x1DAA1, 0x1DAAF},
+	{0x1E000, 0x1E006}, {0x1E008, 0x1E018}, {0x1E01B, 0x1E021},
+	{0x1E023, 0x1E024}, {0x1E026, 0x1E02A}, {0x1E800, 0x1E8C4},
+	{0x1E8C7, 0x1E8CF}, {0x1E8D0, 0x1E8D6}, {0x1E900, 0x1E943},
+	{0x1E944, 0x1E94A}, {0x1E950, 0x1E959}, {0x1E95E, 0x1E95F},
+	{0x1EE00, 0x1EE03}, {0x1EE05, 0x1EE1F}, {0x1EE21, 0x1EE22},
+	{0x1EE24, 0x1EE24}, {0x1EE27, 0x1EE27}, {0x1EE29, 0x1EE32},
+	{0x1EE34, 0x1EE37}, {0x1EE39, 0x1EE39}, {0x1EE3B, 0x1EE3B},
+	{0x1EE42, 0x1EE42}, {0x1EE47, 0x1EE47}, {0x1EE49, 0x1EE49},
+	{0x1EE4B, 0x1EE4B}, {0x1EE4D, 0x1EE4F}, {0x1EE51, 0x1EE52},
+	{0x1EE54, 0x1EE54}, {0x1EE57, 0x1EE57}, {0x1EE59, 0x1EE59},
+	{0x1EE5B, 0x1EE5B}, {0x1EE5D, 0x1EE5D}, {0x1EE5F, 0x1EE5F},
+	{0x1EE61, 0x1EE62}, {0x1EE64, 0x1EE64}, {0x1EE67, 0x1EE6A},
+	{0x1EE6C, 0x1EE72}, {0x1EE74, 0x1EE77}, {0x1EE79, 0x1EE7C},
+	{0x1EE7E, 0x1EE7E}, {0x1EE80, 0x1EE89}, {0x1EE8B, 0x1EE9B},
+	{0x1EEA1, 0x1EEA3}, {0x1EEA5, 0x1EEA9}, {0x1EEAB, 0x1EEBB},
+	{0x1EEF0, 0x1EEF1}, {0x1F000, 0x1F003}, {0x1F005, 0x1F02B},
+	{0x1F030, 0x1F093}, {0x1F0A0, 0x1F0AE}, {0x1F0B1, 0x1F0BF},
+	{0x1F0C1, 0x1F0CE}, {0x1F0D1, 0x1F0F5}, {0x1F10B, 0x1F10C},
+	{0x1F12E, 0x1F12E}, {0x1F16A, 0x1F16B}, {0x1F1E6, 0x1F1FF},
+	{0x1F321, 0x1F32C}, {0x1F336, 0x1F336}, {0x1F37D, 0x1F37D},
+	{0x1F394, 0x1F39F}, {0x1F3CB, 0x1F3CE}, {0x1F3D4, 0x1F3DF},
+	{0x1F3F1, 0x1F3F3}, {0x1F3F5, 0x1F3F7}, {0x1F43F, 0x1F43F},
+	{0x1F441, 0x1F441}, {0x1F4FD, 0x1F4FE}, {0x1F53E, 0x1F54A},
+	{0x1F54F, 0x1F54F}, {0x1F568, 0x1F579}, {0x1F57B, 0x1F594},
+	{0x1F597, 0x1F5A3}, {0x1F5A5, 0x1F5FA}, {0x1F650, 0x1F67F},
+	{0x1F6C6, 0x1F6CB}, {0x1F6CD, 0x1F6CF}, {0x1F6E0, 0x1F6EA},
+	{0x1F6F0, 0x1F6F3}, {0x1F700, 0x1F773}, {0x1F780, 0x1F7D4},
+	{0x1F800, 0x1F80B}, {0x1F810, 0x1F847}, {0x1F850, 0x1F859},
+	{0x1F860, 0x1F887}, {0x1F890, 0x1F8AD}, {0xE0001, 0xE0001},
+	{0xE0020, 0xE007F},
+}
+
+// Condition have flag EastAsianWidth whether the current locale is CJK or not.
+type Condition struct {
+	EastAsianWidth bool
+}
+
+// NewCondition return new instance of Condition which is current locale.
+func NewCondition() *Condition {
+	return &Condition{EastAsianWidth}
+}
+
+// RuneWidth returns the number of cells in r.
+// See http://www.unicode.org/reports/tr11/
+func (c *Condition) RuneWidth(r rune) int {
+	switch {
+	case r < 0 || r > 0x10FFFF ||
+		inTables(r, nonprint, combining, notassigned):
+		return 0
+	case (c.EastAsianWidth && IsAmbiguousWidth(r)) ||
+		inTables(r, doublewidth, emoji):
+		return 2
+	default:
+		return 1
+	}
+}
+
+// StringWidth return width as you can see
+func (c *Condition) StringWidth(s string) (width int) {
+	for _, r := range []rune(s) {
+		width += c.RuneWidth(r)
+	}
+	return width
+}
+
+// Truncate return string truncated with w cells
+func (c *Condition) Truncate(s string, w int, tail string) string {
+	if c.StringWidth(s) <= w {
+		return s
+	}
+	r := []rune(s)
+	tw := c.StringWidth(tail)
+	w -= tw
+	width := 0
+	i := 0
+	for ; i < len(r); i++ {
+		cw := c.RuneWidth(r[i])
+		if width+cw > w {
+			break
+		}
+		width += cw
+	}
+	return string(r[0:i]) + tail
+}
+
+// Wrap return string wrapped with w cells
+func (c *Condition) Wrap(s string, w int) string {
+	width := 0
+	out := ""
+	for _, r := range []rune(s) {
+		cw := RuneWidth(r)
+		if r == '\n' {
+			out += string(r)
+			width = 0
+			continue
+		} else if width+cw > w {
+			out += "\n"
+			width = 0
+			out += string(r)
+			width += cw
+			continue
+		}
+		out += string(r)
+		width += cw
+	}
+	return out
+}
+
+// FillLeft return string filled in left by spaces in w cells
+func (c *Condition) FillLeft(s string, w int) string {
+	width := c.StringWidth(s)
+	count := w - width
+	if count > 0 {
+		b := make([]byte, count)
+		for i := range b {
+			b[i] = ' '
+		}
+		return string(b) + s
+	}
+	return s
+}
+
+// FillRight return string filled in left by spaces in w cells
+func (c *Condition) FillRight(s string, w int) string {
+	width := c.StringWidth(s)
+	count := w - width
+	if count > 0 {
+		b := make([]byte, count)
+		for i := range b {
+			b[i] = ' '
+		}
+		return s + string(b)
+	}
+	return s
+}
+
+// RuneWidth returns the number of cells in r.
+// See http://www.unicode.org/reports/tr11/
+func RuneWidth(r rune) int {
+	return DefaultCondition.RuneWidth(r)
+}
+
+// IsAmbiguousWidth returns whether is ambiguous width or not.
+func IsAmbiguousWidth(r rune) bool {
+	return inTables(r, private, ambiguous)
+}
+
+// IsNeutralWidth returns whether is neutral width or not.
+func IsNeutralWidth(r rune) bool {
+	return inTable(r, neutral)
+}
+
+// StringWidth return width as you can see
+func StringWidth(s string) (width int) {
+	return DefaultCondition.StringWidth(s)
+}
+
+// Truncate return string truncated with w cells
+func Truncate(s string, w int, tail string) string {
+	return DefaultCondition.Truncate(s, w, tail)
+}
+
+// Wrap return string wrapped with w cells
+func Wrap(s string, w int) string {
+	return DefaultCondition.Wrap(s, w)
+}
+
+// FillLeft return string filled in left by spaces in w cells
+func FillLeft(s string, w int) string {
+	return DefaultCondition.FillLeft(s, w)
+}
+
+// FillRight return string filled in left by spaces in w cells
+func FillRight(s string, w int) string {
+	return DefaultCondition.FillRight(s, w)
+}
diff --git a/backend/vendor/github.com/mattn/go-runewidth/runewidth_js.go b/backend/vendor/github.com/mattn/go-runewidth/runewidth_js.go
new file mode 100644
index 00000000..0ce32c5e
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/runewidth_js.go
@@ -0,0 +1,8 @@
+// +build js
+
+package runewidth
+
+func IsEastAsian() bool {
+	// TODO: Implement this for the web. Detect east asian in a compatible way, and return true.
+	return false
+}
diff --git a/backend/vendor/github.com/mattn/go-runewidth/runewidth_posix.go b/backend/vendor/github.com/mattn/go-runewidth/runewidth_posix.go
new file mode 100644
index 00000000..c579e9a3
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/runewidth_posix.go
@@ -0,0 +1,77 @@
+// +build !windows,!js
+
+package runewidth
+
+import (
+	"os"
+	"regexp"
+	"strings"
+)
+
+var reLoc = regexp.MustCompile(`^[a-z][a-z][a-z]?(?:_[A-Z][A-Z])?\.(.+)`)
+
+var mblenTable = map[string]int{
+	"utf-8":   6,
+	"utf8":    6,
+	"jis":     8,
+	"eucjp":   3,
+	"euckr":   2,
+	"euccn":   2,
+	"sjis":    2,
+	"cp932":   2,
+	"cp51932": 2,
+	"cp936":   2,
+	"cp949":   2,
+	"cp950":   2,
+	"big5":    2,
+	"gbk":     2,
+	"gb2312":  2,
+}
+
+func isEastAsian(locale string) bool {
+	charset := strings.ToLower(locale)
+	r := reLoc.FindStringSubmatch(locale)
+	if len(r) == 2 {
+		charset = strings.ToLower(r[1])
+	}
+
+	if strings.HasSuffix(charset, "@cjk_narrow") {
+		return false
+	}
+
+	for pos, b := range []byte(charset) {
+		if b == '@' {
+			charset = charset[:pos]
+			break
+		}
+	}
+	max := 1
+	if m, ok := mblenTable[charset]; ok {
+		max = m
+	}
+	if max > 1 && (charset[0] != 'u' ||
+		strings.HasPrefix(locale, "ja") ||
+		strings.HasPrefix(locale, "ko") ||
+		strings.HasPrefix(locale, "zh")) {
+		return true
+	}
+	return false
+}
+
+// IsEastAsian return true if the current locale is CJK
+func IsEastAsian() bool {
+	locale := os.Getenv("LC_CTYPE")
+	if locale == "" {
+		locale = os.Getenv("LANG")
+	}
+
+	// ignore C locale
+	if locale == "POSIX" || locale == "C" {
+		return false
+	}
+	if len(locale) > 1 && locale[0] == 'C' && (locale[1] == '.' || locale[1] == '-') {
+		return false
+	}
+
+	return isEastAsian(locale)
+}
diff --git a/backend/vendor/github.com/mattn/go-runewidth/runewidth_windows.go b/backend/vendor/github.com/mattn/go-runewidth/runewidth_windows.go
new file mode 100644
index 00000000..0258876b
--- /dev/null
+++ b/backend/vendor/github.com/mattn/go-runewidth/runewidth_windows.go
@@ -0,0 +1,25 @@
+package runewidth
+
+import (
+	"syscall"
+)
+
+var (
+	kernel32               = syscall.NewLazyDLL("kernel32")
+	procGetConsoleOutputCP = kernel32.NewProc("GetConsoleOutputCP")
+)
+
+// IsEastAsian return true if the current locale is CJK
+func IsEastAsian() bool {
+	r1, _, _ := procGetConsoleOutputCP.Call()
+	if r1 == 0 {
+		return false
+	}
+
+	switch int(r1) {
+	case 932, 51932, 936, 949, 950:
+		return true
+	}
+
+	return false
+}
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/.gitignore b/backend/vendor/github.com/olekukonko/tablewriter/.gitignore
new file mode 100644
index 00000000..b66cec63
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/.gitignore
@@ -0,0 +1,15 @@
+# Created by .ignore support plugin (hsz.mobi)
+### Go template
+# Binaries for programs and plugins
+*.exe
+*.exe~
+*.dll
+*.so
+*.dylib
+
+# Test binary, build with `go test -c`
+*.test
+
+# Output of the go coverage tool, specifically when used with LiteIDE
+*.out
+
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/.travis.yml b/backend/vendor/github.com/olekukonko/tablewriter/.travis.yml
new file mode 100644
index 00000000..9c64270e
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/.travis.yml
@@ -0,0 +1,14 @@
+language: go
+
+go:
+  - 1.1
+  - 1.2
+  - 1.3
+  - 1.4
+  - 1.5
+  - 1.6
+  - 1.7
+  - 1.8
+  - 1.9
+  - "1.10"
+  - tip
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/LICENSE.md b/backend/vendor/github.com/olekukonko/tablewriter/LICENSE.md
new file mode 100644
index 00000000..a0769b5c
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/LICENSE.md
@@ -0,0 +1,19 @@
+Copyright (C) 2014 by Oleku Konko
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/README.md b/backend/vendor/github.com/olekukonko/tablewriter/README.md
new file mode 100644
index 00000000..9c2b139b
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/README.md
@@ -0,0 +1,277 @@
+ASCII Table Writer
+=========
+
+[![Build Status](https://travis-ci.org/olekukonko/tablewriter.png?branch=master)](https://travis-ci.org/olekukonko/tablewriter)
+[![Total views](https://img.shields.io/sourcegraph/rrc/github.com/olekukonko/tablewriter.svg)](https://sourcegraph.com/github.com/olekukonko/tablewriter)
+[![Godoc](https://godoc.org/github.com/olekukonko/tablewriter?status.svg)](https://godoc.org/github.com/olekukonko/tablewriter)
+
+Generate ASCII table on the fly ...  Installation is simple as
+
+    go get github.com/olekukonko/tablewriter
+
+
+#### Features
+- Automatic Padding
+- Support Multiple Lines
+- Supports Alignment
+- Support Custom Separators
+- Automatic Alignment of numbers & percentage
+- Write directly to http , file etc via `io.Writer`
+- Read directly from CSV file
+- Optional row line via `SetRowLine`
+- Normalise table header
+- Make CSV Headers optional
+- Enable or disable table border
+- Set custom footer support
+- Optional identical cells merging
+- Set custom caption
+- Optional reflowing of paragrpahs in multi-line cells.
+
+#### Example   1 - Basic
+```go
+data := [][]string{
+    []string{"A", "The Good", "500"},
+    []string{"B", "The Very very Bad Man", "288"},
+    []string{"C", "The Ugly", "120"},
+    []string{"D", "The Gopher", "800"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Name", "Sign", "Rating"})
+
+for _, v := range data {
+    table.Append(v)
+}
+table.Render() // Send output
+```
+
+##### Output  1
+```
++------+-----------------------+--------+
+| NAME |         SIGN          | RATING |
++------+-----------------------+--------+
+|  A   |       The Good        |    500 |
+|  B   | The Very very Bad Man |    288 |
+|  C   |       The Ugly        |    120 |
+|  D   |      The Gopher       |    800 |
++------+-----------------------+--------+
+```
+
+#### Example 2 - Without Border / Footer / Bulk Append
+```go
+data := [][]string{
+    []string{"1/1/2014", "Domain name", "2233", "$10.98"},
+    []string{"1/1/2014", "January Hosting", "2233", "$54.95"},
+    []string{"1/4/2014", "February Hosting", "2233", "$51.00"},
+    []string{"1/4/2014", "February Extra Bandwidth", "2233", "$30.00"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Date", "Description", "CV2", "Amount"})
+table.SetFooter([]string{"", "", "Total", "$146.93"}) // Add Footer
+table.SetBorder(false)                                // Set Border to false
+table.AppendBulk(data)                                // Add Bulk Data
+table.Render()
+```
+
+##### Output 2
+```
+
+    DATE   |       DESCRIPTION        |  CV2  | AMOUNT
++----------+--------------------------+-------+---------+
+  1/1/2014 | Domain name              |  2233 | $10.98
+  1/1/2014 | January Hosting          |  2233 | $54.95
+  1/4/2014 | February Hosting         |  2233 | $51.00
+  1/4/2014 | February Extra Bandwidth |  2233 | $30.00
++----------+--------------------------+-------+---------+
+                                        TOTAL | $146 93
+                                      +-------+---------+
+
+```
+
+
+#### Example 3 - CSV
+```go
+table, _ := tablewriter.NewCSV(os.Stdout, "testdata/test_info.csv", true)
+table.SetAlignment(tablewriter.ALIGN_LEFT)   // Set Alignment
+table.Render()
+```
+
+##### Output 3
+```
++----------+--------------+------+-----+---------+----------------+
+|  FIELD   |     TYPE     | NULL | KEY | DEFAULT |     EXTRA      |
++----------+--------------+------+-----+---------+----------------+
+| user_id  | smallint(5)  | NO   | PRI | NULL    | auto_increment |
+| username | varchar(10)  | NO   |     | NULL    |                |
+| password | varchar(100) | NO   |     | NULL    |                |
++----------+--------------+------+-----+---------+----------------+
+```
+
+#### Example 4  - Custom Separator
+```go
+table, _ := tablewriter.NewCSV(os.Stdout, "testdata/test.csv", true)
+table.SetRowLine(true)         // Enable row line
+
+// Change table lines
+table.SetCenterSeparator("*")
+table.SetColumnSeparator("╪")
+table.SetRowSeparator("-")
+
+table.SetAlignment(tablewriter.ALIGN_LEFT)
+table.Render()
+```
+
+##### Output 4
+```
+*------------*-----------*---------*
+╪ FIRST NAME ╪ LAST NAME ╪   SSN   ╪
+*------------*-----------*---------*
+╪ John       ╪ Barry     ╪ 123456  ╪
+*------------*-----------*---------*
+╪ Kathy      ╪ Smith     ╪ 687987  ╪
+*------------*-----------*---------*
+╪ Bob        ╪ McCornick ╪ 3979870 ╪
+*------------*-----------*---------*
+```
+
+#### Example 5 - Markdown Format
+```go
+data := [][]string{
+	[]string{"1/1/2014", "Domain name", "2233", "$10.98"},
+	[]string{"1/1/2014", "January Hosting", "2233", "$54.95"},
+	[]string{"1/4/2014", "February Hosting", "2233", "$51.00"},
+	[]string{"1/4/2014", "February Extra Bandwidth", "2233", "$30.00"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Date", "Description", "CV2", "Amount"})
+table.SetBorders(tablewriter.Border{Left: true, Top: false, Right: true, Bottom: false})
+table.SetCenterSeparator("|")
+table.AppendBulk(data) // Add Bulk Data
+table.Render()
+```
+
+##### Output 5
+```
+|   DATE   |       DESCRIPTION        | CV2  | AMOUNT |
+|----------|--------------------------|------|--------|
+| 1/1/2014 | Domain name              | 2233 | $10.98 |
+| 1/1/2014 | January Hosting          | 2233 | $54.95 |
+| 1/4/2014 | February Hosting         | 2233 | $51.00 |
+| 1/4/2014 | February Extra Bandwidth | 2233 | $30.00 |
+```
+
+#### Example 6  - Identical cells merging
+```go
+data := [][]string{
+  []string{"1/1/2014", "Domain name", "1234", "$10.98"},
+  []string{"1/1/2014", "January Hosting", "2345", "$54.95"},
+  []string{"1/4/2014", "February Hosting", "3456", "$51.00"},
+  []string{"1/4/2014", "February Extra Bandwidth", "4567", "$30.00"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Date", "Description", "CV2", "Amount"})
+table.SetFooter([]string{"", "", "Total", "$146.93"})
+table.SetAutoMergeCells(true)
+table.SetRowLine(true)
+table.AppendBulk(data)
+table.Render()
+```
+
+##### Output 6
+```
++----------+--------------------------+-------+---------+
+|   DATE   |       DESCRIPTION        |  CV2  | AMOUNT  |
++----------+--------------------------+-------+---------+
+| 1/1/2014 | Domain name              |  1234 | $10.98  |
++          +--------------------------+-------+---------+
+|          | January Hosting          |  2345 | $54.95  |
++----------+--------------------------+-------+---------+
+| 1/4/2014 | February Hosting         |  3456 | $51.00  |
++          +--------------------------+-------+---------+
+|          | February Extra Bandwidth |  4567 | $30.00  |
++----------+--------------------------+-------+---------+
+|                                       TOTAL | $146 93 |
++----------+--------------------------+-------+---------+
+```
+
+
+#### Table with color
+```go
+data := [][]string{
+	[]string{"1/1/2014", "Domain name", "2233", "$10.98"},
+	[]string{"1/1/2014", "January Hosting", "2233", "$54.95"},
+	[]string{"1/4/2014", "February Hosting", "2233", "$51.00"},
+	[]string{"1/4/2014", "February Extra Bandwidth", "2233", "$30.00"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Date", "Description", "CV2", "Amount"})
+table.SetFooter([]string{"", "", "Total", "$146.93"}) // Add Footer
+table.SetBorder(false)                                // Set Border to false
+
+table.SetHeaderColor(tablewriter.Colors{tablewriter.Bold, tablewriter.BgGreenColor},
+	tablewriter.Colors{tablewriter.FgHiRedColor, tablewriter.Bold, tablewriter.BgBlackColor},
+	tablewriter.Colors{tablewriter.BgRedColor, tablewriter.FgWhiteColor},
+	tablewriter.Colors{tablewriter.BgCyanColor, tablewriter.FgWhiteColor})
+
+table.SetColumnColor(tablewriter.Colors{tablewriter.Bold, tablewriter.FgHiBlackColor},
+	tablewriter.Colors{tablewriter.Bold, tablewriter.FgHiRedColor},
+	tablewriter.Colors{tablewriter.Bold, tablewriter.FgHiBlackColor},
+	tablewriter.Colors{tablewriter.Bold, tablewriter.FgBlackColor})
+
+table.SetFooterColor(tablewriter.Colors{}, tablewriter.Colors{},
+	tablewriter.Colors{tablewriter.Bold},
+	tablewriter.Colors{tablewriter.FgHiRedColor})
+
+table.AppendBulk(data)
+table.Render()
+```
+
+#### Table with color Output
+![Table with Color](https://cloud.githubusercontent.com/assets/6460392/21101956/bbc7b356-c0a1-11e6-9f36-dba694746efc.png)
+
+#### Example 6 - Set table caption
+```go
+data := [][]string{
+    []string{"A", "The Good", "500"},
+    []string{"B", "The Very very Bad Man", "288"},
+    []string{"C", "The Ugly", "120"},
+    []string{"D", "The Gopher", "800"},
+}
+
+table := tablewriter.NewWriter(os.Stdout)
+table.SetHeader([]string{"Name", "Sign", "Rating"})
+table.SetCaption(true, "Movie ratings.")
+
+for _, v := range data {
+    table.Append(v)
+}
+table.Render() // Send output
+```
+
+Note: Caption text will wrap with total width of rendered table.
+
+##### Output 6
+```
++------+-----------------------+--------+
+| NAME |         SIGN          | RATING |
++------+-----------------------+--------+
+|  A   |       The Good        |    500 |
+|  B   | The Very very Bad Man |    288 |
+|  C   |       The Ugly        |    120 |
+|  D   |      The Gopher       |    800 |
++------+-----------------------+--------+
+Movie ratings.
+```
+
+#### TODO
+- ~~Import Directly from CSV~~  - `done`
+- ~~Support for `SetFooter`~~  - `done`
+- ~~Support for `SetBorder`~~  - `done`
+- ~~Support table with uneven rows~~ - `done`
+- ~~Support custom alignment~~
+- General Improvement & Optimisation
+- `NewHTML` Parse table from HTML
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/csv.go b/backend/vendor/github.com/olekukonko/tablewriter/csv.go
new file mode 100644
index 00000000..98878303
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/csv.go
@@ -0,0 +1,52 @@
+// Copyright 2014 Oleku Konko All rights reserved.
+// Use of this source code is governed by a MIT
+// license that can be found in the LICENSE file.
+
+// This module is a Table Writer  API for the Go Programming Language.
+// The protocols were written in pure Go and works on windows and unix systems
+
+package tablewriter
+
+import (
+	"encoding/csv"
+	"io"
+	"os"
+)
+
+// Start A new table by importing from a CSV file
+// Takes io.Writer and csv File name
+func NewCSV(writer io.Writer, fileName string, hasHeader bool) (*Table, error) {
+	file, err := os.Open(fileName)
+	if err != nil {
+		return &Table{}, err
+	}
+	defer file.Close()
+	csvReader := csv.NewReader(file)
+	t, err := NewCSVReader(writer, csvReader, hasHeader)
+	return t, err
+}
+
+//  Start a New Table Writer with csv.Reader
+// This enables customisation such as reader.Comma = ';'
+// See http://golang.org/src/pkg/encoding/csv/reader.go?s=3213:3671#L94
+func NewCSVReader(writer io.Writer, csvReader *csv.Reader, hasHeader bool) (*Table, error) {
+	t := NewWriter(writer)
+	if hasHeader {
+		// Read the first row
+		headers, err := csvReader.Read()
+		if err != nil {
+			return &Table{}, err
+		}
+		t.SetHeader(headers)
+	}
+	for {
+		record, err := csvReader.Read()
+		if err == io.EOF {
+			break
+		} else if err != nil {
+			return &Table{}, err
+		}
+		t.Append(record)
+	}
+	return t, nil
+}
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/table.go b/backend/vendor/github.com/olekukonko/tablewriter/table.go
new file mode 100644
index 00000000..dec0385f
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/table.go
@@ -0,0 +1,839 @@
+// Copyright 2014 Oleku Konko All rights reserved.
+// Use of this source code is governed by a MIT
+// license that can be found in the LICENSE file.
+
+// This module is a Table Writer  API for the Go Programming Language.
+// The protocols were written in pure Go and works on windows and unix systems
+
+// Create & Generate text based table
+package tablewriter
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"regexp"
+	"strings"
+)
+
+const (
+	MAX_ROW_WIDTH = 30
+)
+
+const (
+	CENTER  = "+"
+	ROW     = "-"
+	COLUMN  = "|"
+	SPACE   = " "
+	NEWLINE = "\n"
+)
+
+const (
+	ALIGN_DEFAULT = iota
+	ALIGN_CENTER
+	ALIGN_RIGHT
+	ALIGN_LEFT
+)
+
+var (
+	decimal = regexp.MustCompile(`^-?(?:\d{1,3}(?:,\d{3})*|\d+)(?:\.\d+)?$`)
+	percent = regexp.MustCompile(`^-?\d+\.?\d*$%$`)
+)
+
+type Border struct {
+	Left   bool
+	Right  bool
+	Top    bool
+	Bottom bool
+}
+
+type Table struct {
+	out            io.Writer
+	rows           [][]string
+	lines          [][][]string
+	cs             map[int]int
+	rs             map[int]int
+	headers        [][]string
+	footers        [][]string
+	caption        bool
+	captionText    string
+	autoFmt        bool
+	autoWrap       bool
+	reflowText     bool
+	mW             int
+	pCenter        string
+	pRow           string
+	pColumn        string
+	tColumn        int
+	tRow           int
+	hAlign         int
+	fAlign         int
+	align          int
+	newLine        string
+	rowLine        bool
+	autoMergeCells bool
+	hdrLine        bool
+	borders        Border
+	colSize        int
+	headerParams   []string
+	columnsParams  []string
+	footerParams   []string
+	columnsAlign   []int
+}
+
+// Start New Table
+// Take io.Writer Directly
+func NewWriter(writer io.Writer) *Table {
+	t := &Table{
+		out:           writer,
+		rows:          [][]string{},
+		lines:         [][][]string{},
+		cs:            make(map[int]int),
+		rs:            make(map[int]int),
+		headers:       [][]string{},
+		footers:       [][]string{},
+		caption:       false,
+		captionText:   "Table caption.",
+		autoFmt:       true,
+		autoWrap:      true,
+		reflowText:    true,
+		mW:            MAX_ROW_WIDTH,
+		pCenter:       CENTER,
+		pRow:          ROW,
+		pColumn:       COLUMN,
+		tColumn:       -1,
+		tRow:          -1,
+		hAlign:        ALIGN_DEFAULT,
+		fAlign:        ALIGN_DEFAULT,
+		align:         ALIGN_DEFAULT,
+		newLine:       NEWLINE,
+		rowLine:       false,
+		hdrLine:       true,
+		borders:       Border{Left: true, Right: true, Bottom: true, Top: true},
+		colSize:       -1,
+		headerParams:  []string{},
+		columnsParams: []string{},
+		footerParams:  []string{},
+		columnsAlign:  []int{}}
+	return t
+}
+
+// Render table output
+func (t *Table) Render() {
+	if t.borders.Top {
+		t.printLine(true)
+	}
+	t.printHeading()
+	if t.autoMergeCells {
+		t.printRowsMergeCells()
+	} else {
+		t.printRows()
+	}
+	if !t.rowLine && t.borders.Bottom {
+		t.printLine(true)
+	}
+	t.printFooter()
+
+	if t.caption {
+		t.printCaption()
+	}
+}
+
+const (
+	headerRowIdx = -1
+	footerRowIdx = -2
+)
+
+// Set table header
+func (t *Table) SetHeader(keys []string) {
+	t.colSize = len(keys)
+	for i, v := range keys {
+		lines := t.parseDimension(v, i, headerRowIdx)
+		t.headers = append(t.headers, lines)
+	}
+}
+
+// Set table Footer
+func (t *Table) SetFooter(keys []string) {
+	//t.colSize = len(keys)
+	for i, v := range keys {
+		lines := t.parseDimension(v, i, footerRowIdx)
+		t.footers = append(t.footers, lines)
+	}
+}
+
+// Set table Caption
+func (t *Table) SetCaption(caption bool, captionText ...string) {
+	t.caption = caption
+	if len(captionText) == 1 {
+		t.captionText = captionText[0]
+	}
+}
+
+// Turn header autoformatting on/off. Default is on (true).
+func (t *Table) SetAutoFormatHeaders(auto bool) {
+	t.autoFmt = auto
+}
+
+// Turn automatic multiline text adjustment on/off. Default is on (true).
+func (t *Table) SetAutoWrapText(auto bool) {
+	t.autoWrap = auto
+}
+
+// Turn automatic reflowing of multiline text when rewrapping. Default is on (true).
+func (t *Table) SetReflowDuringAutoWrap(auto bool) {
+	t.reflowText = auto
+}
+
+// Set the Default column width
+func (t *Table) SetColWidth(width int) {
+	t.mW = width
+}
+
+// Set the minimal width for a column
+func (t *Table) SetColMinWidth(column int, width int) {
+	t.cs[column] = width
+}
+
+// Set the Column Separator
+func (t *Table) SetColumnSeparator(sep string) {
+	t.pColumn = sep
+}
+
+// Set the Row Separator
+func (t *Table) SetRowSeparator(sep string) {
+	t.pRow = sep
+}
+
+// Set the center Separator
+func (t *Table) SetCenterSeparator(sep string) {
+	t.pCenter = sep
+}
+
+// Set Header Alignment
+func (t *Table) SetHeaderAlignment(hAlign int) {
+	t.hAlign = hAlign
+}
+
+// Set Footer Alignment
+func (t *Table) SetFooterAlignment(fAlign int) {
+	t.fAlign = fAlign
+}
+
+// Set Table Alignment
+func (t *Table) SetAlignment(align int) {
+	t.align = align
+}
+
+func (t *Table) SetColumnAlignment(keys []int) {
+	for _, v := range keys {
+		switch v {
+		case ALIGN_CENTER:
+			break
+		case ALIGN_LEFT:
+			break
+		case ALIGN_RIGHT:
+			break
+		default:
+			v = ALIGN_DEFAULT
+		}
+		t.columnsAlign = append(t.columnsAlign, v)
+	}
+}
+
+// Set New Line
+func (t *Table) SetNewLine(nl string) {
+	t.newLine = nl
+}
+
+// Set Header Line
+// This would enable / disable a line after the header
+func (t *Table) SetHeaderLine(line bool) {
+	t.hdrLine = line
+}
+
+// Set Row Line
+// This would enable / disable a line on each row of the table
+func (t *Table) SetRowLine(line bool) {
+	t.rowLine = line
+}
+
+// Set Auto Merge Cells
+// This would enable / disable the merge of cells with identical values
+func (t *Table) SetAutoMergeCells(auto bool) {
+	t.autoMergeCells = auto
+}
+
+// Set Table Border
+// This would enable / disable line around the table
+func (t *Table) SetBorder(border bool) {
+	t.SetBorders(Border{border, border, border, border})
+}
+
+func (t *Table) SetBorders(border Border) {
+	t.borders = border
+}
+
+// Append row to table
+func (t *Table) Append(row []string) {
+	rowSize := len(t.headers)
+	if rowSize > t.colSize {
+		t.colSize = rowSize
+	}
+
+	n := len(t.lines)
+	line := [][]string{}
+	for i, v := range row {
+
+		// Detect string  width
+		// Detect String height
+		// Break strings into words
+		out := t.parseDimension(v, i, n)
+
+		// Append broken words
+		line = append(line, out)
+	}
+	t.lines = append(t.lines, line)
+}
+
+// Allow Support for Bulk Append
+// Eliminates repeated for loops
+func (t *Table) AppendBulk(rows [][]string) {
+	for _, row := range rows {
+		t.Append(row)
+	}
+}
+
+// NumLines to get the number of lines
+func (t *Table) NumLines() int {
+	return len(t.lines)
+}
+
+// Clear rows
+func (t *Table) ClearRows() {
+	t.lines = [][][]string{}
+}
+
+// Clear footer
+func (t *Table) ClearFooter() {
+	t.footers = [][]string{}
+}
+
+// Print line based on row width
+func (t *Table) printLine(nl bool) {
+	fmt.Fprint(t.out, t.pCenter)
+	for i := 0; i < len(t.cs); i++ {
+		v := t.cs[i]
+		fmt.Fprintf(t.out, "%s%s%s%s",
+			t.pRow,
+			strings.Repeat(string(t.pRow), v),
+			t.pRow,
+			t.pCenter)
+	}
+	if nl {
+		fmt.Fprint(t.out, t.newLine)
+	}
+}
+
+// Print line based on row width with our without cell separator
+func (t *Table) printLineOptionalCellSeparators(nl bool, displayCellSeparator []bool) {
+	fmt.Fprint(t.out, t.pCenter)
+	for i := 0; i < len(t.cs); i++ {
+		v := t.cs[i]
+		if i > len(displayCellSeparator) || displayCellSeparator[i] {
+			// Display the cell separator
+			fmt.Fprintf(t.out, "%s%s%s%s",
+				t.pRow,
+				strings.Repeat(string(t.pRow), v),
+				t.pRow,
+				t.pCenter)
+		} else {
+			// Don't display the cell separator for this cell
+			fmt.Fprintf(t.out, "%s%s",
+				strings.Repeat(" ", v+2),
+				t.pCenter)
+		}
+	}
+	if nl {
+		fmt.Fprint(t.out, t.newLine)
+	}
+}
+
+// Return the PadRight function if align is left, PadLeft if align is right,
+// and Pad by default
+func pad(align int) func(string, string, int) string {
+	padFunc := Pad
+	switch align {
+	case ALIGN_LEFT:
+		padFunc = PadRight
+	case ALIGN_RIGHT:
+		padFunc = PadLeft
+	}
+	return padFunc
+}
+
+// Print heading information
+func (t *Table) printHeading() {
+	// Check if headers is available
+	if len(t.headers) < 1 {
+		return
+	}
+
+	// Identify last column
+	end := len(t.cs) - 1
+
+	// Get pad function
+	padFunc := pad(t.hAlign)
+
+	// Checking for ANSI escape sequences for header
+	is_esc_seq := false
+	if len(t.headerParams) > 0 {
+		is_esc_seq = true
+	}
+
+	// Maximum height.
+	max := t.rs[headerRowIdx]
+
+	// Print Heading
+	for x := 0; x < max; x++ {
+		// Check if border is set
+		// Replace with space if not set
+		fmt.Fprint(t.out, ConditionString(t.borders.Left, t.pColumn, SPACE))
+
+		for y := 0; y <= end; y++ {
+			v := t.cs[y]
+			h := ""
+			if y < len(t.headers) && x < len(t.headers[y]) {
+				h = t.headers[y][x]
+			}
+			if t.autoFmt {
+				h = Title(h)
+			}
+			pad := ConditionString((y == end && !t.borders.Left), SPACE, t.pColumn)
+
+			if is_esc_seq {
+				fmt.Fprintf(t.out, " %s %s",
+					format(padFunc(h, SPACE, v),
+						t.headerParams[y]), pad)
+			} else {
+				fmt.Fprintf(t.out, " %s %s",
+					padFunc(h, SPACE, v),
+					pad)
+			}
+		}
+		// Next line
+		fmt.Fprint(t.out, t.newLine)
+	}
+	if t.hdrLine {
+		t.printLine(true)
+	}
+}
+
+// Print heading information
+func (t *Table) printFooter() {
+	// Check if headers is available
+	if len(t.footers) < 1 {
+		return
+	}
+
+	// Only print line if border is not set
+	if !t.borders.Bottom {
+		t.printLine(true)
+	}
+
+	// Identify last column
+	end := len(t.cs) - 1
+
+	// Get pad function
+	padFunc := pad(t.fAlign)
+
+	// Checking for ANSI escape sequences for header
+	is_esc_seq := false
+	if len(t.footerParams) > 0 {
+		is_esc_seq = true
+	}
+
+	// Maximum height.
+	max := t.rs[footerRowIdx]
+
+	// Print Footer
+	erasePad := make([]bool, len(t.footers))
+	for x := 0; x < max; x++ {
+		// Check if border is set
+		// Replace with space if not set
+		fmt.Fprint(t.out, ConditionString(t.borders.Bottom, t.pColumn, SPACE))
+
+		for y := 0; y <= end; y++ {
+			v := t.cs[y]
+			f := ""
+			if y < len(t.footers) && x < len(t.footers[y]) {
+				f = t.footers[y][x]
+			}
+			if t.autoFmt {
+				f = Title(f)
+			}
+			pad := ConditionString((y == end && !t.borders.Top), SPACE, t.pColumn)
+
+			if erasePad[y] || (x == 0 && len(f) == 0) {
+				pad = SPACE
+				erasePad[y] = true
+			}
+
+			if is_esc_seq {
+				fmt.Fprintf(t.out, " %s %s",
+					format(padFunc(f, SPACE, v),
+						t.footerParams[y]), pad)
+			} else {
+				fmt.Fprintf(t.out, " %s %s",
+					padFunc(f, SPACE, v),
+					pad)
+			}
+
+			//fmt.Fprintf(t.out, " %s %s",
+			//	padFunc(f, SPACE, v),
+			//	pad)
+		}
+		// Next line
+		fmt.Fprint(t.out, t.newLine)
+		//t.printLine(true)
+	}
+
+	hasPrinted := false
+
+	for i := 0; i <= end; i++ {
+		v := t.cs[i]
+		pad := t.pRow
+		center := t.pCenter
+		length := len(t.footers[i][0])
+
+		if length > 0 {
+			hasPrinted = true
+		}
+
+		// Set center to be space if length is 0
+		if length == 0 && !t.borders.Right {
+			center = SPACE
+		}
+
+		// Print first junction
+		if i == 0 {
+			fmt.Fprint(t.out, center)
+		}
+
+		// Pad With space of length is 0
+		if length == 0 {
+			pad = SPACE
+		}
+		// Ignore left space of it has printed before
+		if hasPrinted || t.borders.Left {
+			pad = t.pRow
+			center = t.pCenter
+		}
+
+		// Change Center start position
+		if center == SPACE {
+			if i < end && len(t.footers[i+1][0]) != 0 {
+				center = t.pCenter
+			}
+		}
+
+		// Print the footer
+		fmt.Fprintf(t.out, "%s%s%s%s",
+			pad,
+			strings.Repeat(string(pad), v),
+			pad,
+			center)
+
+	}
+
+	fmt.Fprint(t.out, t.newLine)
+}
+
+// Print caption text
+func (t Table) printCaption() {
+	width := t.getTableWidth()
+	paragraph, _ := WrapString(t.captionText, width)
+	for linecount := 0; linecount < len(paragraph); linecount++ {
+		fmt.Fprintln(t.out, paragraph[linecount])
+	}
+}
+
+// Calculate the total number of characters in a row
+func (t Table) getTableWidth() int {
+	var chars int
+	for _, v := range t.cs {
+		chars += v
+	}
+
+	// Add chars, spaces, seperators to calculate the total width of the table.
+	// ncols := t.colSize
+	// spaces := ncols * 2
+	// seps := ncols + 1
+
+	return (chars + (3 * t.colSize) + 2)
+}
+
+func (t Table) printRows() {
+	for i, lines := range t.lines {
+		t.printRow(lines, i)
+	}
+}
+
+func (t *Table) fillAlignment(num int) {
+	if len(t.columnsAlign) < num {
+		t.columnsAlign = make([]int, num)
+		for i := range t.columnsAlign {
+			t.columnsAlign[i] = t.align
+		}
+	}
+}
+
+// Print Row Information
+// Adjust column alignment based on type
+
+func (t *Table) printRow(columns [][]string, rowIdx int) {
+	// Get Maximum Height
+	max := t.rs[rowIdx]
+	total := len(columns)
+
+	// TODO Fix uneven col size
+	// if total < t.colSize {
+	//	for n := t.colSize - total; n < t.colSize ; n++ {
+	//		columns = append(columns, []string{SPACE})
+	//		t.cs[n] = t.mW
+	//	}
+	//}
+
+	// Pad Each Height
+	pads := []int{}
+
+	// Checking for ANSI escape sequences for columns
+	is_esc_seq := false
+	if len(t.columnsParams) > 0 {
+		is_esc_seq = true
+	}
+	t.fillAlignment(total)
+
+	for i, line := range columns {
+		length := len(line)
+		pad := max - length
+		pads = append(pads, pad)
+		for n := 0; n < pad; n++ {
+			columns[i] = append(columns[i], "  ")
+		}
+	}
+	//fmt.Println(max, "\n")
+	for x := 0; x < max; x++ {
+		for y := 0; y < total; y++ {
+
+			// Check if border is set
+			fmt.Fprint(t.out, ConditionString((!t.borders.Left && y == 0), SPACE, t.pColumn))
+
+			fmt.Fprintf(t.out, SPACE)
+			str := columns[y][x]
+
+			// Embedding escape sequence with column value
+			if is_esc_seq {
+				str = format(str, t.columnsParams[y])
+			}
+
+			// This would print alignment
+			// Default alignment  would use multiple configuration
+			switch t.columnsAlign[y] {
+			case ALIGN_CENTER: //
+				fmt.Fprintf(t.out, "%s", Pad(str, SPACE, t.cs[y]))
+			case ALIGN_RIGHT:
+				fmt.Fprintf(t.out, "%s", PadLeft(str, SPACE, t.cs[y]))
+			case ALIGN_LEFT:
+				fmt.Fprintf(t.out, "%s", PadRight(str, SPACE, t.cs[y]))
+			default:
+				if decimal.MatchString(strings.TrimSpace(str)) || percent.MatchString(strings.TrimSpace(str)) {
+					fmt.Fprintf(t.out, "%s", PadLeft(str, SPACE, t.cs[y]))
+				} else {
+					fmt.Fprintf(t.out, "%s", PadRight(str, SPACE, t.cs[y]))
+
+					// TODO Custom alignment per column
+					//if max == 1 || pads[y] > 0 {
+					//	fmt.Fprintf(t.out, "%s", Pad(str, SPACE, t.cs[y]))
+					//} else {
+					//	fmt.Fprintf(t.out, "%s", PadRight(str, SPACE, t.cs[y]))
+					//}
+
+				}
+			}
+			fmt.Fprintf(t.out, SPACE)
+		}
+		// Check if border is set
+		// Replace with space if not set
+		fmt.Fprint(t.out, ConditionString(t.borders.Left, t.pColumn, SPACE))
+		fmt.Fprint(t.out, t.newLine)
+	}
+
+	if t.rowLine {
+		t.printLine(true)
+	}
+}
+
+// Print the rows of the table and merge the cells that are identical
+func (t *Table) printRowsMergeCells() {
+	var previousLine []string
+	var displayCellBorder []bool
+	var tmpWriter bytes.Buffer
+	for i, lines := range t.lines {
+		// We store the display of the current line in a tmp writer, as we need to know which border needs to be print above
+		previousLine, displayCellBorder = t.printRowMergeCells(&tmpWriter, lines, i, previousLine)
+		if i > 0 { //We don't need to print borders above first line
+			if t.rowLine {
+				t.printLineOptionalCellSeparators(true, displayCellBorder)
+			}
+		}
+		tmpWriter.WriteTo(t.out)
+	}
+	//Print the end of the table
+	if t.rowLine {
+		t.printLine(true)
+	}
+}
+
+// Print Row Information to a writer and merge identical cells.
+// Adjust column alignment based on type
+
+func (t *Table) printRowMergeCells(writer io.Writer, columns [][]string, rowIdx int, previousLine []string) ([]string, []bool) {
+	// Get Maximum Height
+	max := t.rs[rowIdx]
+	total := len(columns)
+
+	// Pad Each Height
+	pads := []int{}
+
+	for i, line := range columns {
+		length := len(line)
+		pad := max - length
+		pads = append(pads, pad)
+		for n := 0; n < pad; n++ {
+			columns[i] = append(columns[i], "  ")
+		}
+	}
+
+	var displayCellBorder []bool
+	t.fillAlignment(total)
+	for x := 0; x < max; x++ {
+		for y := 0; y < total; y++ {
+
+			// Check if border is set
+			fmt.Fprint(writer, ConditionString((!t.borders.Left && y == 0), SPACE, t.pColumn))
+
+			fmt.Fprintf(writer, SPACE)
+
+			str := columns[y][x]
+
+			if t.autoMergeCells {
+				//Store the full line to merge mutli-lines cells
+				fullLine := strings.Join(columns[y], " ")
+				if len(previousLine) > y && fullLine == previousLine[y] && fullLine != "" {
+					// If this cell is identical to the one above but not empty, we don't display the border and keep the cell empty.
+					displayCellBorder = append(displayCellBorder, false)
+					str = ""
+				} else {
+					// First line or different content, keep the content and print the cell border
+					displayCellBorder = append(displayCellBorder, true)
+				}
+			}
+
+			// This would print alignment
+			// Default alignment  would use multiple configuration
+			switch t.columnsAlign[y] {
+			case ALIGN_CENTER: //
+				fmt.Fprintf(writer, "%s", Pad(str, SPACE, t.cs[y]))
+			case ALIGN_RIGHT:
+				fmt.Fprintf(writer, "%s", PadLeft(str, SPACE, t.cs[y]))
+			case ALIGN_LEFT:
+				fmt.Fprintf(writer, "%s", PadRight(str, SPACE, t.cs[y]))
+			default:
+				if decimal.MatchString(strings.TrimSpace(str)) || percent.MatchString(strings.TrimSpace(str)) {
+					fmt.Fprintf(writer, "%s", PadLeft(str, SPACE, t.cs[y]))
+				} else {
+					fmt.Fprintf(writer, "%s", PadRight(str, SPACE, t.cs[y]))
+				}
+			}
+			fmt.Fprintf(writer, SPACE)
+		}
+		// Check if border is set
+		// Replace with space if not set
+		fmt.Fprint(writer, ConditionString(t.borders.Left, t.pColumn, SPACE))
+		fmt.Fprint(writer, t.newLine)
+	}
+
+	//The new previous line is the current one
+	previousLine = make([]string, total)
+	for y := 0; y < total; y++ {
+		previousLine[y] = strings.Join(columns[y], " ") //Store the full line for multi-lines cells
+	}
+	//Returns the newly added line and wether or not a border should be displayed above.
+	return previousLine, displayCellBorder
+}
+
+func (t *Table) parseDimension(str string, colKey, rowKey int) []string {
+	var (
+		raw      []string
+		maxWidth int
+	)
+
+	raw = getLines(str)
+	maxWidth = 0
+	for _, line := range raw {
+		if w := DisplayWidth(line); w > maxWidth {
+			maxWidth = w
+		}
+	}
+
+	// If wrapping, ensure that all paragraphs in the cell fit in the
+	// specified width.
+	if t.autoWrap {
+		// If there's a maximum allowed width for wrapping, use that.
+		if maxWidth > t.mW {
+			maxWidth = t.mW
+		}
+
+		// In the process of doing so, we need to recompute maxWidth. This
+		// is because perhaps a word in the cell is longer than the
+		// allowed maximum width in t.mW.
+		newMaxWidth := maxWidth
+		newRaw := make([]string, 0, len(raw))
+
+		if t.reflowText {
+			// Make a single paragraph of everything.
+			raw = []string{strings.Join(raw, " ")}
+		}
+		for i, para := range raw {
+			paraLines, _ := WrapString(para, maxWidth)
+			for _, line := range paraLines {
+				if w := DisplayWidth(line); w > newMaxWidth {
+					newMaxWidth = w
+				}
+			}
+			if i > 0 {
+				newRaw = append(newRaw, " ")
+			}
+			newRaw = append(newRaw, paraLines...)
+		}
+		raw = newRaw
+		maxWidth = newMaxWidth
+	}
+
+	// Store the new known maximum width.
+	v, ok := t.cs[colKey]
+	if !ok || v < maxWidth || v == 0 {
+		t.cs[colKey] = maxWidth
+	}
+
+	// Remember the number of lines for the row printer.
+	h := len(raw)
+	v, ok = t.rs[rowKey]
+
+	if !ok || v < h || v == 0 {
+		t.rs[rowKey] = h
+	}
+	//fmt.Printf("Raw %+v %d\n", raw, len(raw))
+	return raw
+}
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/table_with_color.go b/backend/vendor/github.com/olekukonko/tablewriter/table_with_color.go
new file mode 100644
index 00000000..5a4a53ec
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/table_with_color.go
@@ -0,0 +1,134 @@
+package tablewriter
+
+import (
+	"fmt"
+	"strconv"
+	"strings"
+)
+
+const ESC = "\033"
+const SEP = ";"
+
+const (
+	BgBlackColor int = iota + 40
+	BgRedColor
+	BgGreenColor
+	BgYellowColor
+	BgBlueColor
+	BgMagentaColor
+	BgCyanColor
+	BgWhiteColor
+)
+
+const (
+	FgBlackColor int = iota + 30
+	FgRedColor
+	FgGreenColor
+	FgYellowColor
+	FgBlueColor
+	FgMagentaColor
+	FgCyanColor
+	FgWhiteColor
+)
+
+const (
+	BgHiBlackColor int = iota + 100
+	BgHiRedColor
+	BgHiGreenColor
+	BgHiYellowColor
+	BgHiBlueColor
+	BgHiMagentaColor
+	BgHiCyanColor
+	BgHiWhiteColor
+)
+
+const (
+	FgHiBlackColor int = iota + 90
+	FgHiRedColor
+	FgHiGreenColor
+	FgHiYellowColor
+	FgHiBlueColor
+	FgHiMagentaColor
+	FgHiCyanColor
+	FgHiWhiteColor
+)
+
+const (
+	Normal          = 0
+	Bold            = 1
+	UnderlineSingle = 4
+	Italic
+)
+
+type Colors []int
+
+func startFormat(seq string) string {
+	return fmt.Sprintf("%s[%sm", ESC, seq)
+}
+
+func stopFormat() string {
+	return fmt.Sprintf("%s[%dm", ESC, Normal)
+}
+
+// Making the SGR (Select Graphic Rendition) sequence.
+func makeSequence(codes []int) string {
+	codesInString := []string{}
+	for _, code := range codes {
+		codesInString = append(codesInString, strconv.Itoa(code))
+	}
+	return strings.Join(codesInString, SEP)
+}
+
+// Adding ANSI escape  sequences before and after string
+func format(s string, codes interface{}) string {
+	var seq string
+
+	switch v := codes.(type) {
+
+	case string:
+		seq = v
+	case []int:
+		seq = makeSequence(v)
+	default:
+		return s
+	}
+
+	if len(seq) == 0 {
+		return s
+	}
+	return startFormat(seq) + s + stopFormat()
+}
+
+// Adding header colors (ANSI codes)
+func (t *Table) SetHeaderColor(colors ...Colors) {
+	if t.colSize != len(colors) {
+		panic("Number of header colors must be equal to number of headers.")
+	}
+	for i := 0; i < len(colors); i++ {
+		t.headerParams = append(t.headerParams, makeSequence(colors[i]))
+	}
+}
+
+// Adding column colors (ANSI codes)
+func (t *Table) SetColumnColor(colors ...Colors) {
+	if t.colSize != len(colors) {
+		panic("Number of column colors must be equal to number of headers.")
+	}
+	for i := 0; i < len(colors); i++ {
+		t.columnsParams = append(t.columnsParams, makeSequence(colors[i]))
+	}
+}
+
+// Adding column colors (ANSI codes)
+func (t *Table) SetFooterColor(colors ...Colors) {
+	if len(t.footers) != len(colors) {
+		panic("Number of footer colors must be equal to number of footer.")
+	}
+	for i := 0; i < len(colors); i++ {
+		t.footerParams = append(t.footerParams, makeSequence(colors[i]))
+	}
+}
+
+func Color(colors ...int) []int {
+	return colors
+}
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/util.go b/backend/vendor/github.com/olekukonko/tablewriter/util.go
new file mode 100644
index 00000000..9e8f0cbb
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/util.go
@@ -0,0 +1,93 @@
+// Copyright 2014 Oleku Konko All rights reserved.
+// Use of this source code is governed by a MIT
+// license that can be found in the LICENSE file.
+
+// This module is a Table Writer  API for the Go Programming Language.
+// The protocols were written in pure Go and works on windows and unix systems
+
+package tablewriter
+
+import (
+	"math"
+	"regexp"
+	"strings"
+
+	"github.com/mattn/go-runewidth"
+)
+
+var ansi = regexp.MustCompile("\033\\[(?:[0-9]{1,3}(?:;[0-9]{1,3})*)?[m|K]")
+
+func DisplayWidth(str string) int {
+	return runewidth.StringWidth(ansi.ReplaceAllLiteralString(str, ""))
+}
+
+// Simple Condition for string
+// Returns value based on condition
+func ConditionString(cond bool, valid, inValid string) string {
+	if cond {
+		return valid
+	}
+	return inValid
+}
+
+func isNumOrSpace(r rune) bool {
+	return ('0' <= r && r <= '9') || r == ' '
+}
+
+// Format Table Header
+// Replace _ , . and spaces
+func Title(name string) string {
+	origLen := len(name)
+	rs := []rune(name)
+	for i, r := range rs {
+		switch r {
+		case '_':
+			rs[i] = ' '
+		case '.':
+			// ignore floating number 0.0
+			if (i != 0 && !isNumOrSpace(rs[i-1])) || (i != len(rs)-1 && !isNumOrSpace(rs[i+1])) {
+				rs[i] = ' '
+			}
+		}
+	}
+	name = string(rs)
+	name = strings.TrimSpace(name)
+	if len(name) == 0 && origLen > 0 {
+		// Keep at least one character. This is important to preserve
+		// empty lines in multi-line headers/footers.
+		name = " "
+	}
+	return strings.ToUpper(name)
+}
+
+// Pad String
+// Attempts to play string in the center
+func Pad(s, pad string, width int) string {
+	gap := width - DisplayWidth(s)
+	if gap > 0 {
+		gapLeft := int(math.Ceil(float64(gap / 2)))
+		gapRight := gap - gapLeft
+		return strings.Repeat(string(pad), gapLeft) + s + strings.Repeat(string(pad), gapRight)
+	}
+	return s
+}
+
+// Pad String Right position
+// This would pace string at the left side fo the screen
+func PadRight(s, pad string, width int) string {
+	gap := width - DisplayWidth(s)
+	if gap > 0 {
+		return s + strings.Repeat(string(pad), gap)
+	}
+	return s
+}
+
+// Pad String Left position
+// This would pace string at the right side fo the screen
+func PadLeft(s, pad string, width int) string {
+	gap := width - DisplayWidth(s)
+	if gap > 0 {
+		return strings.Repeat(string(pad), gap) + s
+	}
+	return s
+}
diff --git a/backend/vendor/github.com/olekukonko/tablewriter/wrap.go b/backend/vendor/github.com/olekukonko/tablewriter/wrap.go
new file mode 100644
index 00000000..a092ee1f
--- /dev/null
+++ b/backend/vendor/github.com/olekukonko/tablewriter/wrap.go
@@ -0,0 +1,99 @@
+// Copyright 2014 Oleku Konko All rights reserved.
+// Use of this source code is governed by a MIT
+// license that can be found in the LICENSE file.
+
+// This module is a Table Writer  API for the Go Programming Language.
+// The protocols were written in pure Go and works on windows and unix systems
+
+package tablewriter
+
+import (
+	"math"
+	"strings"
+
+	"github.com/mattn/go-runewidth"
+)
+
+var (
+	nl = "\n"
+	sp = " "
+)
+
+const defaultPenalty = 1e5
+
+// Wrap wraps s into a paragraph of lines of length lim, with minimal
+// raggedness.
+func WrapString(s string, lim int) ([]string, int) {
+	words := strings.Split(strings.Replace(s, nl, sp, -1), sp)
+	var lines []string
+	max := 0
+	for _, v := range words {
+		max = runewidth.StringWidth(v)
+		if max > lim {
+			lim = max
+		}
+	}
+	for _, line := range WrapWords(words, 1, lim, defaultPenalty) {
+		lines = append(lines, strings.Join(line, sp))
+	}
+	return lines, lim
+}
+
+// WrapWords is the low-level line-breaking algorithm, useful if you need more
+// control over the details of the text wrapping process. For most uses,
+// WrapString will be sufficient and more convenient.
+//
+// WrapWords splits a list of words into lines with minimal "raggedness",
+// treating each rune as one unit, accounting for spc units between adjacent
+// words on each line, and attempting to limit lines to lim units. Raggedness
+// is the total error over all lines, where error is the square of the
+// difference of the length of the line and lim. Too-long lines (which only
+// happen when a single word is longer than lim units) have pen penalty units
+// added to the error.
+func WrapWords(words []string, spc, lim, pen int) [][]string {
+	n := len(words)
+
+	length := make([][]int, n)
+	for i := 0; i < n; i++ {
+		length[i] = make([]int, n)
+		length[i][i] = runewidth.StringWidth(words[i])
+		for j := i + 1; j < n; j++ {
+			length[i][j] = length[i][j-1] + spc + runewidth.StringWidth(words[j])
+		}
+	}
+	nbrk := make([]int, n)
+	cost := make([]int, n)
+	for i := range cost {
+		cost[i] = math.MaxInt32
+	}
+	for i := n - 1; i >= 0; i-- {
+		if length[i][n-1] <= lim {
+			cost[i] = 0
+			nbrk[i] = n
+		} else {
+			for j := i + 1; j < n; j++ {
+				d := lim - length[i][j-1]
+				c := d*d + cost[j]
+				if length[i][j-1] > lim {
+					c += pen // too-long lines get a worse penalty
+				}
+				if c < cost[i] {
+					cost[i] = c
+					nbrk[i] = j
+				}
+			}
+		}
+	}
+	var lines [][]string
+	i := 0
+	for i < n {
+		lines = append(lines, words[i:nbrk[i]])
+		i = nbrk[i]
+	}
+	return lines
+}
+
+// getLines decomposes a multiline string into a slice of strings.
+func getLines(s string) []string {
+	return strings.Split(s, nl)
+}
diff --git a/backend/vendor/github.com/shurcooL/sanitized_anchor_name/.travis.yml b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/.travis.yml
new file mode 100644
index 00000000..93b1fcdb
--- /dev/null
+++ b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/.travis.yml
@@ -0,0 +1,16 @@
+sudo: false
+language: go
+go:
+  - 1.x
+  - master
+matrix:
+  allow_failures:
+    - go: master
+  fast_finish: true
+install:
+  - # Do nothing. This is needed to prevent default install action "go get -t -v ./..." from happening here (we want it to happen inside script step).
+script:
+  - go get -t -v ./...
+  - diff -u <(echo -n) <(gofmt -d -s .)
+  - go tool vet .
+  - go test -v -race ./...
diff --git a/backend/vendor/github.com/shurcooL/sanitized_anchor_name/LICENSE b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/LICENSE
new file mode 100644
index 00000000..c35c17af
--- /dev/null
+++ b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2015 Dmitri Shuralyov
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/backend/vendor/github.com/shurcooL/sanitized_anchor_name/README.md b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/README.md
new file mode 100644
index 00000000..670bf0fe
--- /dev/null
+++ b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/README.md
@@ -0,0 +1,36 @@
+sanitized_anchor_name
+=====================
+
+[![Build Status](https://travis-ci.org/shurcooL/sanitized_anchor_name.svg?branch=master)](https://travis-ci.org/shurcooL/sanitized_anchor_name) [![GoDoc](https://godoc.org/github.com/shurcooL/sanitized_anchor_name?status.svg)](https://godoc.org/github.com/shurcooL/sanitized_anchor_name)
+
+Package sanitized_anchor_name provides a func to create sanitized anchor names.
+
+Its logic can be reused by multiple packages to create interoperable anchor names
+and links to those anchors.
+
+At this time, it does not try to ensure that generated anchor names
+are unique, that responsibility falls on the caller.
+
+Installation
+------------
+
+```bash
+go get -u github.com/shurcooL/sanitized_anchor_name
+```
+
+Example
+-------
+
+```Go
+anchorName := sanitized_anchor_name.Create("This is a header")
+
+fmt.Println(anchorName)
+
+// Output:
+// this-is-a-header
+```
+
+License
+-------
+
+-	[MIT License](LICENSE)
diff --git a/backend/vendor/github.com/shurcooL/sanitized_anchor_name/go.mod b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/go.mod
new file mode 100644
index 00000000..1e255347
--- /dev/null
+++ b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/go.mod
@@ -0,0 +1 @@
+module github.com/shurcooL/sanitized_anchor_name
diff --git a/backend/vendor/github.com/shurcooL/sanitized_anchor_name/main.go b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/main.go
new file mode 100644
index 00000000..6a77d124
--- /dev/null
+++ b/backend/vendor/github.com/shurcooL/sanitized_anchor_name/main.go
@@ -0,0 +1,29 @@
+// Package sanitized_anchor_name provides a func to create sanitized anchor names.
+//
+// Its logic can be reused by multiple packages to create interoperable anchor names
+// and links to those anchors.
+//
+// At this time, it does not try to ensure that generated anchor names
+// are unique, that responsibility falls on the caller.
+package sanitized_anchor_name // import "github.com/shurcooL/sanitized_anchor_name"
+
+import "unicode"
+
+// Create returns a sanitized anchor name for the given text.
+func Create(text string) string {
+	var anchorName []rune
+	var futureDash = false
+	for _, r := range text {
+		switch {
+		case unicode.IsLetter(r) || unicode.IsNumber(r):
+			if futureDash && len(anchorName) > 0 {
+				anchorName = append(anchorName, '-')
+			}
+			futureDash = false
+			anchorName = append(anchorName, unicode.ToLower(r))
+		default:
+			futureDash = true
+		}
+	}
+	return string(anchorName)
+}
diff --git a/backend/vendor/github.com/ssor/bom/.travis.yml b/backend/vendor/github.com/ssor/bom/.travis.yml
new file mode 100644
index 00000000..6c7f48ef
--- /dev/null
+++ b/backend/vendor/github.com/ssor/bom/.travis.yml
@@ -0,0 +1,14 @@
+language: go
+go:
+  - tip
+  - 1.8
+  - 1.7
+  - 1.6
+  - 1.5
+  - 1.4
+  - 1.3
+  - 1.2
+notifications:
+  email:
+    on_success: change
+    on_failure: always
diff --git a/backend/vendor/github.com/ssor/bom/LICENSE b/backend/vendor/github.com/ssor/bom/LICENSE
new file mode 100644
index 00000000..374f6855
--- /dev/null
+++ b/backend/vendor/github.com/ssor/bom/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2017 Asher
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/backend/vendor/github.com/ssor/bom/README.md b/backend/vendor/github.com/ssor/bom/README.md
new file mode 100644
index 00000000..2dcc289f
--- /dev/null
+++ b/backend/vendor/github.com/ssor/bom/README.md
@@ -0,0 +1,23 @@
+# bom
+small tools for cleaning bom from byte array or reader
+
+
+## Installation
+
+```sh
+$ go get github.com/ssor/bom
+```
+
+## How to Use
+
+
+```
+	bs := []byte{bom0, bom1, bom2, 0x11}
+	result := CleanBom(bs)
+```
+
+```
+	bs := []byte{bom0, bom1, bom2, 0x11}
+	result := NewReaderWithoutBom(bytes.NewReader(bs))
+
+```
\ No newline at end of file
diff --git a/backend/vendor/github.com/ssor/bom/bom.go b/backend/vendor/github.com/ssor/bom/bom.go
new file mode 100644
index 00000000..907ea98d
--- /dev/null
+++ b/backend/vendor/github.com/ssor/bom/bom.go
@@ -0,0 +1,34 @@
+package bom
+
+import (
+	"bytes"
+	"io"
+	"io/ioutil"
+)
+
+const (
+	bom0 = 0xef
+	bom1 = 0xbb
+	bom2 = 0xbf
+)
+
+// CleanBom returns b with the 3 byte BOM stripped off the front if it is present.
+// If the BOM is not present, then b is returned.
+func CleanBom(b []byte) []byte {
+	if len(b) >= 3 &&
+		b[0] == bom0 &&
+		b[1] == bom1 &&
+		b[2] == bom2 {
+		return b[3:]
+	}
+	return b
+}
+
+// NewReaderWithoutBom returns an io.Reader that will skip over initial UTF-8 byte order marks.
+func NewReaderWithoutBom(r io.Reader) (io.Reader, error) {
+	bs, err := ioutil.ReadAll(r)
+	if err != nil {
+		return nil, err
+	}
+	return bytes.NewReader(CleanBom(bs)), nil
+}
diff --git a/backend/vendor/golang.org/x/crypto/AUTHORS b/backend/vendor/golang.org/x/crypto/AUTHORS
new file mode 100644
index 00000000..2b00ddba
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/AUTHORS
@@ -0,0 +1,3 @@
+# This source code refers to The Go Authors for copyright purposes.
+# The master list of authors is in the main Go distribution,
+# visible at https://tip.golang.org/AUTHORS.
diff --git a/backend/vendor/golang.org/x/crypto/CONTRIBUTORS b/backend/vendor/golang.org/x/crypto/CONTRIBUTORS
new file mode 100644
index 00000000..1fbd3e97
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/CONTRIBUTORS
@@ -0,0 +1,3 @@
+# This source code was written by the Go contributors.
+# The master list of contributors is in the main Go distribution,
+# visible at https://tip.golang.org/CONTRIBUTORS.
diff --git a/backend/vendor/golang.org/x/crypto/LICENSE b/backend/vendor/golang.org/x/crypto/LICENSE
new file mode 100644
index 00000000..6a66aea5
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/LICENSE
@@ -0,0 +1,27 @@
+Copyright (c) 2009 The Go Authors. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+   * Redistributions in binary form must reproduce the above
+copyright notice, this list of conditions and the following disclaimer
+in the documentation and/or other materials provided with the
+distribution.
+   * Neither the name of Google Inc. nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/backend/vendor/golang.org/x/crypto/PATENTS b/backend/vendor/golang.org/x/crypto/PATENTS
new file mode 100644
index 00000000..73309904
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/PATENTS
@@ -0,0 +1,22 @@
+Additional IP Rights Grant (Patents)
+
+"This implementation" means the copyrightable works distributed by
+Google as part of the Go project.
+
+Google hereby grants to You a perpetual, worldwide, non-exclusive,
+no-charge, royalty-free, irrevocable (except as stated in this section)
+patent license to make, have made, use, offer to sell, sell, import,
+transfer and otherwise run, modify and propagate the contents of this
+implementation of Go, where such license applies only to those patent
+claims, both currently owned or controlled by Google and acquired in
+the future, licensable by Google that are necessarily infringed by this
+implementation of Go.  This grant does not include claims that would be
+infringed only as a consequence of further modification of this
+implementation.  If you or your agent or exclusive licensee institute or
+order or agree to the institution of patent litigation against any
+entity (including a cross-claim or counterclaim in a lawsuit) alleging
+that this implementation of Go or any code incorporated within this
+implementation of Go constitutes direct or contributory patent
+infringement, or inducement of patent infringement, then any patent
+rights granted to you under this License for this implementation of Go
+shall terminate as of the date such litigation is filed.
diff --git a/backend/vendor/golang.org/x/crypto/pbkdf2/pbkdf2.go b/backend/vendor/golang.org/x/crypto/pbkdf2/pbkdf2.go
new file mode 100644
index 00000000..593f6530
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/pbkdf2/pbkdf2.go
@@ -0,0 +1,77 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+/*
+Package pbkdf2 implements the key derivation function PBKDF2 as defined in RFC
+2898 / PKCS #5 v2.0.
+
+A key derivation function is useful when encrypting data based on a password
+or any other not-fully-random data. It uses a pseudorandom function to derive
+a secure encryption key based on the password.
+
+While v2.0 of the standard defines only one pseudorandom function to use,
+HMAC-SHA1, the drafted v2.1 specification allows use of all five FIPS Approved
+Hash Functions SHA-1, SHA-224, SHA-256, SHA-384 and SHA-512 for HMAC. To
+choose, you can pass the `New` functions from the different SHA packages to
+pbkdf2.Key.
+*/
+package pbkdf2 // import "golang.org/x/crypto/pbkdf2"
+
+import (
+	"crypto/hmac"
+	"hash"
+)
+
+// Key derives a key from the password, salt and iteration count, returning a
+// []byte of length keylen that can be used as cryptographic key. The key is
+// derived based on the method described as PBKDF2 with the HMAC variant using
+// the supplied hash function.
+//
+// For example, to use a HMAC-SHA-1 based PBKDF2 key derivation function, you
+// can get a derived key for e.g. AES-256 (which needs a 32-byte key) by
+// doing:
+//
+// 	dk := pbkdf2.Key([]byte("some password"), salt, 4096, 32, sha1.New)
+//
+// Remember to get a good random salt. At least 8 bytes is recommended by the
+// RFC.
+//
+// Using a higher iteration count will increase the cost of an exhaustive
+// search but will also make derivation proportionally slower.
+func Key(password, salt []byte, iter, keyLen int, h func() hash.Hash) []byte {
+	prf := hmac.New(h, password)
+	hashLen := prf.Size()
+	numBlocks := (keyLen + hashLen - 1) / hashLen
+
+	var buf [4]byte
+	dk := make([]byte, 0, numBlocks*hashLen)
+	U := make([]byte, hashLen)
+	for block := 1; block <= numBlocks; block++ {
+		// N.B.: || means concatenation, ^ means XOR
+		// for each block T_i = U_1 ^ U_2 ^ ... ^ U_iter
+		// U_1 = PRF(password, salt || uint(i))
+		prf.Reset()
+		prf.Write(salt)
+		buf[0] = byte(block >> 24)
+		buf[1] = byte(block >> 16)
+		buf[2] = byte(block >> 8)
+		buf[3] = byte(block)
+		prf.Write(buf[:4])
+		dk = prf.Sum(dk)
+		T := dk[len(dk)-hashLen:]
+		copy(U, T)
+
+		// U_n = PRF(password, U_(n-1))
+		for n := 2; n <= iter; n++ {
+			prf.Reset()
+			prf.Write(U)
+			U = U[:0]
+			U = prf.Sum(U)
+			for x := range U {
+				T[x] ^= U[x]
+			}
+		}
+	}
+	return dk[:keyLen]
+}
diff --git a/backend/vendor/golang.org/x/crypto/scrypt/scrypt.go b/backend/vendor/golang.org/x/crypto/scrypt/scrypt.go
new file mode 100644
index 00000000..2f81fe41
--- /dev/null
+++ b/backend/vendor/golang.org/x/crypto/scrypt/scrypt.go
@@ -0,0 +1,213 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package scrypt implements the scrypt key derivation function as defined in
+// Colin Percival's paper "Stronger Key Derivation via Sequential Memory-Hard
+// Functions" (https://www.tarsnap.com/scrypt/scrypt.pdf).
+package scrypt // import "golang.org/x/crypto/scrypt"
+
+import (
+	"crypto/sha256"
+	"errors"
+	"math/bits"
+
+	"golang.org/x/crypto/pbkdf2"
+)
+
+const maxInt = int(^uint(0) >> 1)
+
+// blockCopy copies n numbers from src into dst.
+func blockCopy(dst, src []uint32, n int) {
+	copy(dst, src[:n])
+}
+
+// blockXOR XORs numbers from dst with n numbers from src.
+func blockXOR(dst, src []uint32, n int) {
+	for i, v := range src[:n] {
+		dst[i] ^= v
+	}
+}
+
+// salsaXOR applies Salsa20/8 to the XOR of 16 numbers from tmp and in,
+// and puts the result into both tmp and out.
+func salsaXOR(tmp *[16]uint32, in, out []uint32) {
+	w0 := tmp[0] ^ in[0]
+	w1 := tmp[1] ^ in[1]
+	w2 := tmp[2] ^ in[2]
+	w3 := tmp[3] ^ in[3]
+	w4 := tmp[4] ^ in[4]
+	w5 := tmp[5] ^ in[5]
+	w6 := tmp[6] ^ in[6]
+	w7 := tmp[7] ^ in[7]
+	w8 := tmp[8] ^ in[8]
+	w9 := tmp[9] ^ in[9]
+	w10 := tmp[10] ^ in[10]
+	w11 := tmp[11] ^ in[11]
+	w12 := tmp[12] ^ in[12]
+	w13 := tmp[13] ^ in[13]
+	w14 := tmp[14] ^ in[14]
+	w15 := tmp[15] ^ in[15]
+
+	x0, x1, x2, x3, x4, x5, x6, x7, x8 := w0, w1, w2, w3, w4, w5, w6, w7, w8
+	x9, x10, x11, x12, x13, x14, x15 := w9, w10, w11, w12, w13, w14, w15
+
+	for i := 0; i < 8; i += 2 {
+		x4 ^= bits.RotateLeft32(x0+x12, 7)
+		x8 ^= bits.RotateLeft32(x4+x0, 9)
+		x12 ^= bits.RotateLeft32(x8+x4, 13)
+		x0 ^= bits.RotateLeft32(x12+x8, 18)
+
+		x9 ^= bits.RotateLeft32(x5+x1, 7)
+		x13 ^= bits.RotateLeft32(x9+x5, 9)
+		x1 ^= bits.RotateLeft32(x13+x9, 13)
+		x5 ^= bits.RotateLeft32(x1+x13, 18)
+
+		x14 ^= bits.RotateLeft32(x10+x6, 7)
+		x2 ^= bits.RotateLeft32(x14+x10, 9)
+		x6 ^= bits.RotateLeft32(x2+x14, 13)
+		x10 ^= bits.RotateLeft32(x6+x2, 18)
+
+		x3 ^= bits.RotateLeft32(x15+x11, 7)
+		x7 ^= bits.RotateLeft32(x3+x15, 9)
+		x11 ^= bits.RotateLeft32(x7+x3, 13)
+		x15 ^= bits.RotateLeft32(x11+x7, 18)
+
+		x1 ^= bits.RotateLeft32(x0+x3, 7)
+		x2 ^= bits.RotateLeft32(x1+x0, 9)
+		x3 ^= bits.RotateLeft32(x2+x1, 13)
+		x0 ^= bits.RotateLeft32(x3+x2, 18)
+
+		x6 ^= bits.RotateLeft32(x5+x4, 7)
+		x7 ^= bits.RotateLeft32(x6+x5, 9)
+		x4 ^= bits.RotateLeft32(x7+x6, 13)
+		x5 ^= bits.RotateLeft32(x4+x7, 18)
+
+		x11 ^= bits.RotateLeft32(x10+x9, 7)
+		x8 ^= bits.RotateLeft32(x11+x10, 9)
+		x9 ^= bits.RotateLeft32(x8+x11, 13)
+		x10 ^= bits.RotateLeft32(x9+x8, 18)
+
+		x12 ^= bits.RotateLeft32(x15+x14, 7)
+		x13 ^= bits.RotateLeft32(x12+x15, 9)
+		x14 ^= bits.RotateLeft32(x13+x12, 13)
+		x15 ^= bits.RotateLeft32(x14+x13, 18)
+	}
+	x0 += w0
+	x1 += w1
+	x2 += w2
+	x3 += w3
+	x4 += w4
+	x5 += w5
+	x6 += w6
+	x7 += w7
+	x8 += w8
+	x9 += w9
+	x10 += w10
+	x11 += w11
+	x12 += w12
+	x13 += w13
+	x14 += w14
+	x15 += w15
+
+	out[0], tmp[0] = x0, x0
+	out[1], tmp[1] = x1, x1
+	out[2], tmp[2] = x2, x2
+	out[3], tmp[3] = x3, x3
+	out[4], tmp[4] = x4, x4
+	out[5], tmp[5] = x5, x5
+	out[6], tmp[6] = x6, x6
+	out[7], tmp[7] = x7, x7
+	out[8], tmp[8] = x8, x8
+	out[9], tmp[9] = x9, x9
+	out[10], tmp[10] = x10, x10
+	out[11], tmp[11] = x11, x11
+	out[12], tmp[12] = x12, x12
+	out[13], tmp[13] = x13, x13
+	out[14], tmp[14] = x14, x14
+	out[15], tmp[15] = x15, x15
+}
+
+func blockMix(tmp *[16]uint32, in, out []uint32, r int) {
+	blockCopy(tmp[:], in[(2*r-1)*16:], 16)
+	for i := 0; i < 2*r; i += 2 {
+		salsaXOR(tmp, in[i*16:], out[i*8:])
+		salsaXOR(tmp, in[i*16+16:], out[i*8+r*16:])
+	}
+}
+
+func integer(b []uint32, r int) uint64 {
+	j := (2*r - 1) * 16
+	return uint64(b[j]) | uint64(b[j+1])<<32
+}
+
+func smix(b []byte, r, N int, v, xy []uint32) {
+	var tmp [16]uint32
+	x := xy
+	y := xy[32*r:]
+
+	j := 0
+	for i := 0; i < 32*r; i++ {
+		x[i] = uint32(b[j]) | uint32(b[j+1])<<8 | uint32(b[j+2])<<16 | uint32(b[j+3])<<24
+		j += 4
+	}
+	for i := 0; i < N; i += 2 {
+		blockCopy(v[i*(32*r):], x, 32*r)
+		blockMix(&tmp, x, y, r)
+
+		blockCopy(v[(i+1)*(32*r):], y, 32*r)
+		blockMix(&tmp, y, x, r)
+	}
+	for i := 0; i < N; i += 2 {
+		j := int(integer(x, r) & uint64(N-1))
+		blockXOR(x, v[j*(32*r):], 32*r)
+		blockMix(&tmp, x, y, r)
+
+		j = int(integer(y, r) & uint64(N-1))
+		blockXOR(y, v[j*(32*r):], 32*r)
+		blockMix(&tmp, y, x, r)
+	}
+	j = 0
+	for _, v := range x[:32*r] {
+		b[j+0] = byte(v >> 0)
+		b[j+1] = byte(v >> 8)
+		b[j+2] = byte(v >> 16)
+		b[j+3] = byte(v >> 24)
+		j += 4
+	}
+}
+
+// Key derives a key from the password, salt, and cost parameters, returning
+// a byte slice of length keyLen that can be used as cryptographic key.
+//
+// N is a CPU/memory cost parameter, which must be a power of two greater than 1.
+// r and p must satisfy r * p < 2³⁰. If the parameters do not satisfy the
+// limits, the function returns a nil byte slice and an error.
+//
+// For example, you can get a derived key for e.g. AES-256 (which needs a
+// 32-byte key) by doing:
+//
+//      dk, err := scrypt.Key([]byte("some password"), salt, 32768, 8, 1, 32)
+//
+// The recommended parameters for interactive logins as of 2017 are N=32768, r=8
+// and p=1. The parameters N, r, and p should be increased as memory latency and
+// CPU parallelism increases; consider setting N to the highest power of 2 you
+// can derive within 100 milliseconds. Remember to get a good random salt.
+func Key(password, salt []byte, N, r, p, keyLen int) ([]byte, error) {
+	if N <= 1 || N&(N-1) != 0 {
+		return nil, errors.New("scrypt: N must be > 1 and a power of 2")
+	}
+	if uint64(r)*uint64(p) >= 1<<30 || r > maxInt/128/p || r > maxInt/256 || N > maxInt/128/r {
+		return nil, errors.New("scrypt: parameters are too large")
+	}
+
+	xy := make([]uint32, 64*r)
+	v := make([]uint32, 32*N*r)
+	b := pbkdf2.Key(password, salt, 1, p*128*r, sha256.New)
+
+	for i := 0; i < p; i++ {
+		smix(b[i*128*r:], r, N, v, xy)
+	}
+
+	return pbkdf2.Key(password, b, 1, keyLen, sha256.New), nil
+}
diff --git a/backend/vendor/golang.org/x/net/AUTHORS b/backend/vendor/golang.org/x/net/AUTHORS
new file mode 100644
index 00000000..15167cd7
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/AUTHORS
@@ -0,0 +1,3 @@
+# This source code refers to The Go Authors for copyright purposes.
+# The master list of authors is in the main Go distribution,
+# visible at http://tip.golang.org/AUTHORS.
diff --git a/backend/vendor/golang.org/x/net/CONTRIBUTORS b/backend/vendor/golang.org/x/net/CONTRIBUTORS
new file mode 100644
index 00000000..1c4577e9
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/CONTRIBUTORS
@@ -0,0 +1,3 @@
+# This source code was written by the Go contributors.
+# The master list of contributors is in the main Go distribution,
+# visible at http://tip.golang.org/CONTRIBUTORS.
diff --git a/backend/vendor/golang.org/x/net/LICENSE b/backend/vendor/golang.org/x/net/LICENSE
new file mode 100644
index 00000000..6a66aea5
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/LICENSE
@@ -0,0 +1,27 @@
+Copyright (c) 2009 The Go Authors. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+   * Redistributions in binary form must reproduce the above
+copyright notice, this list of conditions and the following disclaimer
+in the documentation and/or other materials provided with the
+distribution.
+   * Neither the name of Google Inc. nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/backend/vendor/golang.org/x/net/PATENTS b/backend/vendor/golang.org/x/net/PATENTS
new file mode 100644
index 00000000..73309904
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/PATENTS
@@ -0,0 +1,22 @@
+Additional IP Rights Grant (Patents)
+
+"This implementation" means the copyrightable works distributed by
+Google as part of the Go project.
+
+Google hereby grants to You a perpetual, worldwide, non-exclusive,
+no-charge, royalty-free, irrevocable (except as stated in this section)
+patent license to make, have made, use, offer to sell, sell, import,
+transfer and otherwise run, modify and propagate the contents of this
+implementation of Go, where such license applies only to those patent
+claims, both currently owned or controlled by Google and acquired in
+the future, licensable by Google that are necessarily infringed by this
+implementation of Go.  This grant does not include claims that would be
+infringed only as a consequence of further modification of this
+implementation.  If you or your agent or exclusive licensee institute or
+order or agree to the institution of patent litigation against any
+entity (including a cross-claim or counterclaim in a lawsuit) alleging
+that this implementation of Go or any code incorporated within this
+implementation of Go constitutes direct or contributory patent
+infringement, or inducement of patent infringement, then any patent
+rights granted to you under this License for this implementation of Go
+shall terminate as of the date such litigation is filed.
diff --git a/backend/vendor/golang.org/x/net/html/atom/atom.go b/backend/vendor/golang.org/x/net/html/atom/atom.go
new file mode 100644
index 00000000..cd0a8ac1
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/atom/atom.go
@@ -0,0 +1,78 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package atom provides integer codes (also known as atoms) for a fixed set of
+// frequently occurring HTML strings: tag names and attribute keys such as "p"
+// and "id".
+//
+// Sharing an atom's name between all elements with the same tag can result in
+// fewer string allocations when tokenizing and parsing HTML. Integer
+// comparisons are also generally faster than string comparisons.
+//
+// The value of an atom's particular code is not guaranteed to stay the same
+// between versions of this package. Neither is any ordering guaranteed:
+// whether atom.H1 < atom.H2 may also change. The codes are not guaranteed to
+// be dense. The only guarantees are that e.g. looking up "div" will yield
+// atom.Div, calling atom.Div.String will return "div", and atom.Div != 0.
+package atom // import "golang.org/x/net/html/atom"
+
+// Atom is an integer code for a string. The zero value maps to "".
+type Atom uint32
+
+// String returns the atom's name.
+func (a Atom) String() string {
+	start := uint32(a >> 8)
+	n := uint32(a & 0xff)
+	if start+n > uint32(len(atomText)) {
+		return ""
+	}
+	return atomText[start : start+n]
+}
+
+func (a Atom) string() string {
+	return atomText[a>>8 : a>>8+a&0xff]
+}
+
+// fnv computes the FNV hash with an arbitrary starting value h.
+func fnv(h uint32, s []byte) uint32 {
+	for i := range s {
+		h ^= uint32(s[i])
+		h *= 16777619
+	}
+	return h
+}
+
+func match(s string, t []byte) bool {
+	for i, c := range t {
+		if s[i] != c {
+			return false
+		}
+	}
+	return true
+}
+
+// Lookup returns the atom whose name is s. It returns zero if there is no
+// such atom. The lookup is case sensitive.
+func Lookup(s []byte) Atom {
+	if len(s) == 0 || len(s) > maxAtomLen {
+		return 0
+	}
+	h := fnv(hash0, s)
+	if a := table[h&uint32(len(table)-1)]; int(a&0xff) == len(s) && match(a.string(), s) {
+		return a
+	}
+	if a := table[(h>>16)&uint32(len(table)-1)]; int(a&0xff) == len(s) && match(a.string(), s) {
+		return a
+	}
+	return 0
+}
+
+// String returns a string whose contents are equal to s. In that sense, it is
+// equivalent to string(s) but may be more efficient.
+func String(s []byte) string {
+	if a := Lookup(s); a != 0 {
+		return a.String()
+	}
+	return string(s)
+}
diff --git a/backend/vendor/golang.org/x/net/html/atom/table.go b/backend/vendor/golang.org/x/net/html/atom/table.go
new file mode 100644
index 00000000..2a938864
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/atom/table.go
@@ -0,0 +1,783 @@
+// Code generated by go generate gen.go; DO NOT EDIT.
+
+//go:generate go run gen.go
+
+package atom
+
+const (
+	A                         Atom = 0x1
+	Abbr                      Atom = 0x4
+	Accept                    Atom = 0x1a06
+	AcceptCharset             Atom = 0x1a0e
+	Accesskey                 Atom = 0x2c09
+	Acronym                   Atom = 0xaa07
+	Action                    Atom = 0x27206
+	Address                   Atom = 0x6f307
+	Align                     Atom = 0xb105
+	Allowfullscreen           Atom = 0x2080f
+	Allowpaymentrequest       Atom = 0xc113
+	Allowusermedia            Atom = 0xdd0e
+	Alt                       Atom = 0xf303
+	Annotation                Atom = 0x1c90a
+	AnnotationXml             Atom = 0x1c90e
+	Applet                    Atom = 0x31906
+	Area                      Atom = 0x35604
+	Article                   Atom = 0x3fc07
+	As                        Atom = 0x3c02
+	Aside                     Atom = 0x10705
+	Async                     Atom = 0xff05
+	Audio                     Atom = 0x11505
+	Autocomplete              Atom = 0x2780c
+	Autofocus                 Atom = 0x12109
+	Autoplay                  Atom = 0x13c08
+	B                         Atom = 0x101
+	Base                      Atom = 0x3b04
+	Basefont                  Atom = 0x3b08
+	Bdi                       Atom = 0xba03
+	Bdo                       Atom = 0x14b03
+	Bgsound                   Atom = 0x15e07
+	Big                       Atom = 0x17003
+	Blink                     Atom = 0x17305
+	Blockquote                Atom = 0x1870a
+	Body                      Atom = 0x2804
+	Br                        Atom = 0x202
+	Button                    Atom = 0x19106
+	Canvas                    Atom = 0x10306
+	Caption                   Atom = 0x23107
+	Center                    Atom = 0x22006
+	Challenge                 Atom = 0x29b09
+	Charset                   Atom = 0x2107
+	Checked                   Atom = 0x47907
+	Cite                      Atom = 0x19c04
+	Class                     Atom = 0x56405
+	Code                      Atom = 0x5c504
+	Col                       Atom = 0x1ab03
+	Colgroup                  Atom = 0x1ab08
+	Color                     Atom = 0x1bf05
+	Cols                      Atom = 0x1c404
+	Colspan                   Atom = 0x1c407
+	Command                   Atom = 0x1d707
+	Content                   Atom = 0x58b07
+	Contenteditable           Atom = 0x58b0f
+	Contextmenu               Atom = 0x3800b
+	Controls                  Atom = 0x1de08
+	Coords                    Atom = 0x1ea06
+	Crossorigin               Atom = 0x1fb0b
+	Data                      Atom = 0x4a504
+	Datalist                  Atom = 0x4a508
+	Datetime                  Atom = 0x2b808
+	Dd                        Atom = 0x2d702
+	Default                   Atom = 0x10a07
+	Defer                     Atom = 0x5c705
+	Del                       Atom = 0x45203
+	Desc                      Atom = 0x56104
+	Details                   Atom = 0x7207
+	Dfn                       Atom = 0x8703
+	Dialog                    Atom = 0xbb06
+	Dir                       Atom = 0x9303
+	Dirname                   Atom = 0x9307
+	Disabled                  Atom = 0x16408
+	Div                       Atom = 0x16b03
+	Dl                        Atom = 0x5e602
+	Download                  Atom = 0x46308
+	Draggable                 Atom = 0x17a09
+	Dropzone                  Atom = 0x40508
+	Dt                        Atom = 0x64b02
+	Em                        Atom = 0x6e02
+	Embed                     Atom = 0x6e05
+	Enctype                   Atom = 0x28d07
+	Face                      Atom = 0x21e04
+	Fieldset                  Atom = 0x22608
+	Figcaption                Atom = 0x22e0a
+	Figure                    Atom = 0x24806
+	Font                      Atom = 0x3f04
+	Footer                    Atom = 0xf606
+	For                       Atom = 0x25403
+	ForeignObject             Atom = 0x2540d
+	Foreignobject             Atom = 0x2610d
+	Form                      Atom = 0x26e04
+	Formaction                Atom = 0x26e0a
+	Formenctype               Atom = 0x2890b
+	Formmethod                Atom = 0x2a40a
+	Formnovalidate            Atom = 0x2ae0e
+	Formtarget                Atom = 0x2c00a
+	Frame                     Atom = 0x8b05
+	Frameset                  Atom = 0x8b08
+	H1                        Atom = 0x15c02
+	H2                        Atom = 0x2de02
+	H3                        Atom = 0x30d02
+	H4                        Atom = 0x34502
+	H5                        Atom = 0x34f02
+	H6                        Atom = 0x64d02
+	Head                      Atom = 0x33104
+	Header                    Atom = 0x33106
+	Headers                   Atom = 0x33107
+	Height                    Atom = 0x5206
+	Hgroup                    Atom = 0x2ca06
+	Hidden                    Atom = 0x2d506
+	High                      Atom = 0x2db04
+	Hr                        Atom = 0x15702
+	Href                      Atom = 0x2e004
+	Hreflang                  Atom = 0x2e008
+	Html                      Atom = 0x5604
+	HttpEquiv                 Atom = 0x2e80a
+	I                         Atom = 0x601
+	Icon                      Atom = 0x58a04
+	Id                        Atom = 0x10902
+	Iframe                    Atom = 0x2fc06
+	Image                     Atom = 0x30205
+	Img                       Atom = 0x30703
+	Input                     Atom = 0x44b05
+	Inputmode                 Atom = 0x44b09
+	Ins                       Atom = 0x20403
+	Integrity                 Atom = 0x23f09
+	Is                        Atom = 0x16502
+	Isindex                   Atom = 0x30f07
+	Ismap                     Atom = 0x31605
+	Itemid                    Atom = 0x38b06
+	Itemprop                  Atom = 0x19d08
+	Itemref                   Atom = 0x3cd07
+	Itemscope                 Atom = 0x67109
+	Itemtype                  Atom = 0x31f08
+	Kbd                       Atom = 0xb903
+	Keygen                    Atom = 0x3206
+	Keytype                   Atom = 0xd607
+	Kind                      Atom = 0x17704
+	Label                     Atom = 0x5905
+	Lang                      Atom = 0x2e404
+	Legend                    Atom = 0x18106
+	Li                        Atom = 0xb202
+	Link                      Atom = 0x17404
+	List                      Atom = 0x4a904
+	Listing                   Atom = 0x4a907
+	Loop                      Atom = 0x5d04
+	Low                       Atom = 0xc303
+	Main                      Atom = 0x1004
+	Malignmark                Atom = 0xb00a
+	Manifest                  Atom = 0x6d708
+	Map                       Atom = 0x31803
+	Mark                      Atom = 0xb604
+	Marquee                   Atom = 0x32707
+	Math                      Atom = 0x32e04
+	Max                       Atom = 0x33d03
+	Maxlength                 Atom = 0x33d09
+	Media                     Atom = 0xe605
+	Mediagroup                Atom = 0xe60a
+	Menu                      Atom = 0x38704
+	Menuitem                  Atom = 0x38708
+	Meta                      Atom = 0x4b804
+	Meter                     Atom = 0x9805
+	Method                    Atom = 0x2a806
+	Mglyph                    Atom = 0x30806
+	Mi                        Atom = 0x34702
+	Min                       Atom = 0x34703
+	Minlength                 Atom = 0x34709
+	Mn                        Atom = 0x2b102
+	Mo                        Atom = 0xa402
+	Ms                        Atom = 0x67402
+	Mtext                     Atom = 0x35105
+	Multiple                  Atom = 0x35f08
+	Muted                     Atom = 0x36705
+	Name                      Atom = 0x9604
+	Nav                       Atom = 0x1303
+	Nobr                      Atom = 0x3704
+	Noembed                   Atom = 0x6c07
+	Noframes                  Atom = 0x8908
+	Nomodule                  Atom = 0xa208
+	Nonce                     Atom = 0x1a605
+	Noscript                  Atom = 0x21608
+	Novalidate                Atom = 0x2b20a
+	Object                    Atom = 0x26806
+	Ol                        Atom = 0x13702
+	Onabort                   Atom = 0x19507
+	Onafterprint              Atom = 0x2360c
+	Onautocomplete            Atom = 0x2760e
+	Onautocompleteerror       Atom = 0x27613
+	Onauxclick                Atom = 0x61f0a
+	Onbeforeprint             Atom = 0x69e0d
+	Onbeforeunload            Atom = 0x6e70e
+	Onblur                    Atom = 0x56d06
+	Oncancel                  Atom = 0x11908
+	Oncanplay                 Atom = 0x14d09
+	Oncanplaythrough          Atom = 0x14d10
+	Onchange                  Atom = 0x41b08
+	Onclick                   Atom = 0x2f507
+	Onclose                   Atom = 0x36c07
+	Oncontextmenu             Atom = 0x37e0d
+	Oncopy                    Atom = 0x39106
+	Oncuechange               Atom = 0x3970b
+	Oncut                     Atom = 0x3a205
+	Ondblclick                Atom = 0x3a70a
+	Ondrag                    Atom = 0x3b106
+	Ondragend                 Atom = 0x3b109
+	Ondragenter               Atom = 0x3ba0b
+	Ondragexit                Atom = 0x3c50a
+	Ondragleave               Atom = 0x3df0b
+	Ondragover                Atom = 0x3ea0a
+	Ondragstart               Atom = 0x3f40b
+	Ondrop                    Atom = 0x40306
+	Ondurationchange          Atom = 0x41310
+	Onemptied                 Atom = 0x40a09
+	Onended                   Atom = 0x42307
+	Onerror                   Atom = 0x42a07
+	Onfocus                   Atom = 0x43107
+	Onhashchange              Atom = 0x43d0c
+	Oninput                   Atom = 0x44907
+	Oninvalid                 Atom = 0x45509
+	Onkeydown                 Atom = 0x45e09
+	Onkeypress                Atom = 0x46b0a
+	Onkeyup                   Atom = 0x48007
+	Onlanguagechange          Atom = 0x48d10
+	Onload                    Atom = 0x49d06
+	Onloadeddata              Atom = 0x49d0c
+	Onloadedmetadata          Atom = 0x4b010
+	Onloadend                 Atom = 0x4c609
+	Onloadstart               Atom = 0x4cf0b
+	Onmessage                 Atom = 0x4da09
+	Onmessageerror            Atom = 0x4da0e
+	Onmousedown               Atom = 0x4e80b
+	Onmouseenter              Atom = 0x4f30c
+	Onmouseleave              Atom = 0x4ff0c
+	Onmousemove               Atom = 0x50b0b
+	Onmouseout                Atom = 0x5160a
+	Onmouseover               Atom = 0x5230b
+	Onmouseup                 Atom = 0x52e09
+	Onmousewheel              Atom = 0x53c0c
+	Onoffline                 Atom = 0x54809
+	Ononline                  Atom = 0x55108
+	Onpagehide                Atom = 0x5590a
+	Onpageshow                Atom = 0x5730a
+	Onpaste                   Atom = 0x57f07
+	Onpause                   Atom = 0x59a07
+	Onplay                    Atom = 0x5a406
+	Onplaying                 Atom = 0x5a409
+	Onpopstate                Atom = 0x5ad0a
+	Onprogress                Atom = 0x5b70a
+	Onratechange              Atom = 0x5cc0c
+	Onrejectionhandled        Atom = 0x5d812
+	Onreset                   Atom = 0x5ea07
+	Onresize                  Atom = 0x5f108
+	Onscroll                  Atom = 0x60008
+	Onsecuritypolicyviolation Atom = 0x60819
+	Onseeked                  Atom = 0x62908
+	Onseeking                 Atom = 0x63109
+	Onselect                  Atom = 0x63a08
+	Onshow                    Atom = 0x64406
+	Onsort                    Atom = 0x64f06
+	Onstalled                 Atom = 0x65909
+	Onstorage                 Atom = 0x66209
+	Onsubmit                  Atom = 0x66b08
+	Onsuspend                 Atom = 0x67b09
+	Ontimeupdate              Atom = 0x400c
+	Ontoggle                  Atom = 0x68408
+	Onunhandledrejection      Atom = 0x68c14
+	Onunload                  Atom = 0x6ab08
+	Onvolumechange            Atom = 0x6b30e
+	Onwaiting                 Atom = 0x6c109
+	Onwheel                   Atom = 0x6ca07
+	Open                      Atom = 0x1a304
+	Optgroup                  Atom = 0x5f08
+	Optimum                   Atom = 0x6d107
+	Option                    Atom = 0x6e306
+	Output                    Atom = 0x51d06
+	P                         Atom = 0xc01
+	Param                     Atom = 0xc05
+	Pattern                   Atom = 0x6607
+	Picture                   Atom = 0x7b07
+	Ping                      Atom = 0xef04
+	Placeholder               Atom = 0x1310b
+	Plaintext                 Atom = 0x1b209
+	Playsinline               Atom = 0x1400b
+	Poster                    Atom = 0x2cf06
+	Pre                       Atom = 0x47003
+	Preload                   Atom = 0x48607
+	Progress                  Atom = 0x5b908
+	Prompt                    Atom = 0x53606
+	Public                    Atom = 0x58606
+	Q                         Atom = 0xcf01
+	Radiogroup                Atom = 0x30a
+	Rb                        Atom = 0x3a02
+	Readonly                  Atom = 0x35708
+	Referrerpolicy            Atom = 0x3d10e
+	Rel                       Atom = 0x48703
+	Required                  Atom = 0x24c08
+	Reversed                  Atom = 0x8008
+	Rows                      Atom = 0x9c04
+	Rowspan                   Atom = 0x9c07
+	Rp                        Atom = 0x23c02
+	Rt                        Atom = 0x19a02
+	Rtc                       Atom = 0x19a03
+	Ruby                      Atom = 0xfb04
+	S                         Atom = 0x2501
+	Samp                      Atom = 0x7804
+	Sandbox                   Atom = 0x12907
+	Scope                     Atom = 0x67505
+	Scoped                    Atom = 0x67506
+	Script                    Atom = 0x21806
+	Seamless                  Atom = 0x37108
+	Section                   Atom = 0x56807
+	Select                    Atom = 0x63c06
+	Selected                  Atom = 0x63c08
+	Shape                     Atom = 0x1e505
+	Size                      Atom = 0x5f504
+	Sizes                     Atom = 0x5f505
+	Slot                      Atom = 0x1ef04
+	Small                     Atom = 0x20605
+	Sortable                  Atom = 0x65108
+	Sorted                    Atom = 0x33706
+	Source                    Atom = 0x37806
+	Spacer                    Atom = 0x43706
+	Span                      Atom = 0x9f04
+	Spellcheck                Atom = 0x4740a
+	Src                       Atom = 0x5c003
+	Srcdoc                    Atom = 0x5c006
+	Srclang                   Atom = 0x5f907
+	Srcset                    Atom = 0x6f906
+	Start                     Atom = 0x3fa05
+	Step                      Atom = 0x58304
+	Strike                    Atom = 0xd206
+	Strong                    Atom = 0x6dd06
+	Style                     Atom = 0x6ff05
+	Sub                       Atom = 0x66d03
+	Summary                   Atom = 0x70407
+	Sup                       Atom = 0x70b03
+	Svg                       Atom = 0x70e03
+	System                    Atom = 0x71106
+	Tabindex                  Atom = 0x4be08
+	Table                     Atom = 0x59505
+	Target                    Atom = 0x2c406
+	Tbody                     Atom = 0x2705
+	Td                        Atom = 0x9202
+	Template                  Atom = 0x71408
+	Textarea                  Atom = 0x35208
+	Tfoot                     Atom = 0xf505
+	Th                        Atom = 0x15602
+	Thead                     Atom = 0x33005
+	Time                      Atom = 0x4204
+	Title                     Atom = 0x11005
+	Tr                        Atom = 0xcc02
+	Track                     Atom = 0x1ba05
+	Translate                 Atom = 0x1f209
+	Tt                        Atom = 0x6802
+	Type                      Atom = 0xd904
+	Typemustmatch             Atom = 0x2900d
+	U                         Atom = 0xb01
+	Ul                        Atom = 0xa702
+	Updateviacache            Atom = 0x460e
+	Usemap                    Atom = 0x59e06
+	Value                     Atom = 0x1505
+	Var                       Atom = 0x16d03
+	Video                     Atom = 0x2f105
+	Wbr                       Atom = 0x57c03
+	Width                     Atom = 0x64905
+	Workertype                Atom = 0x71c0a
+	Wrap                      Atom = 0x72604
+	Xmp                       Atom = 0x12f03
+)
+
+const hash0 = 0x81cdf10e
+
+const maxAtomLen = 25
+
+var table = [1 << 9]Atom{
+	0x1:   0xe60a,  // mediagroup
+	0x2:   0x2e404, // lang
+	0x4:   0x2c09,  // accesskey
+	0x5:   0x8b08,  // frameset
+	0x7:   0x63a08, // onselect
+	0x8:   0x71106, // system
+	0xa:   0x64905, // width
+	0xc:   0x2890b, // formenctype
+	0xd:   0x13702, // ol
+	0xe:   0x3970b, // oncuechange
+	0x10:  0x14b03, // bdo
+	0x11:  0x11505, // audio
+	0x12:  0x17a09, // draggable
+	0x14:  0x2f105, // video
+	0x15:  0x2b102, // mn
+	0x16:  0x38704, // menu
+	0x17:  0x2cf06, // poster
+	0x19:  0xf606,  // footer
+	0x1a:  0x2a806, // method
+	0x1b:  0x2b808, // datetime
+	0x1c:  0x19507, // onabort
+	0x1d:  0x460e,  // updateviacache
+	0x1e:  0xff05,  // async
+	0x1f:  0x49d06, // onload
+	0x21:  0x11908, // oncancel
+	0x22:  0x62908, // onseeked
+	0x23:  0x30205, // image
+	0x24:  0x5d812, // onrejectionhandled
+	0x26:  0x17404, // link
+	0x27:  0x51d06, // output
+	0x28:  0x33104, // head
+	0x29:  0x4ff0c, // onmouseleave
+	0x2a:  0x57f07, // onpaste
+	0x2b:  0x5a409, // onplaying
+	0x2c:  0x1c407, // colspan
+	0x2f:  0x1bf05, // color
+	0x30:  0x5f504, // size
+	0x31:  0x2e80a, // http-equiv
+	0x33:  0x601,   // i
+	0x34:  0x5590a, // onpagehide
+	0x35:  0x68c14, // onunhandledrejection
+	0x37:  0x42a07, // onerror
+	0x3a:  0x3b08,  // basefont
+	0x3f:  0x1303,  // nav
+	0x40:  0x17704, // kind
+	0x41:  0x35708, // readonly
+	0x42:  0x30806, // mglyph
+	0x44:  0xb202,  // li
+	0x46:  0x2d506, // hidden
+	0x47:  0x70e03, // svg
+	0x48:  0x58304, // step
+	0x49:  0x23f09, // integrity
+	0x4a:  0x58606, // public
+	0x4c:  0x1ab03, // col
+	0x4d:  0x1870a, // blockquote
+	0x4e:  0x34f02, // h5
+	0x50:  0x5b908, // progress
+	0x51:  0x5f505, // sizes
+	0x52:  0x34502, // h4
+	0x56:  0x33005, // thead
+	0x57:  0xd607,  // keytype
+	0x58:  0x5b70a, // onprogress
+	0x59:  0x44b09, // inputmode
+	0x5a:  0x3b109, // ondragend
+	0x5d:  0x3a205, // oncut
+	0x5e:  0x43706, // spacer
+	0x5f:  0x1ab08, // colgroup
+	0x62:  0x16502, // is
+	0x65:  0x3c02,  // as
+	0x66:  0x54809, // onoffline
+	0x67:  0x33706, // sorted
+	0x69:  0x48d10, // onlanguagechange
+	0x6c:  0x43d0c, // onhashchange
+	0x6d:  0x9604,  // name
+	0x6e:  0xf505,  // tfoot
+	0x6f:  0x56104, // desc
+	0x70:  0x33d03, // max
+	0x72:  0x1ea06, // coords
+	0x73:  0x30d02, // h3
+	0x74:  0x6e70e, // onbeforeunload
+	0x75:  0x9c04,  // rows
+	0x76:  0x63c06, // select
+	0x77:  0x9805,  // meter
+	0x78:  0x38b06, // itemid
+	0x79:  0x53c0c, // onmousewheel
+	0x7a:  0x5c006, // srcdoc
+	0x7d:  0x1ba05, // track
+	0x7f:  0x31f08, // itemtype
+	0x82:  0xa402,  // mo
+	0x83:  0x41b08, // onchange
+	0x84:  0x33107, // headers
+	0x85:  0x5cc0c, // onratechange
+	0x86:  0x60819, // onsecuritypolicyviolation
+	0x88:  0x4a508, // datalist
+	0x89:  0x4e80b, // onmousedown
+	0x8a:  0x1ef04, // slot
+	0x8b:  0x4b010, // onloadedmetadata
+	0x8c:  0x1a06,  // accept
+	0x8d:  0x26806, // object
+	0x91:  0x6b30e, // onvolumechange
+	0x92:  0x2107,  // charset
+	0x93:  0x27613, // onautocompleteerror
+	0x94:  0xc113,  // allowpaymentrequest
+	0x95:  0x2804,  // body
+	0x96:  0x10a07, // default
+	0x97:  0x63c08, // selected
+	0x98:  0x21e04, // face
+	0x99:  0x1e505, // shape
+	0x9b:  0x68408, // ontoggle
+	0x9e:  0x64b02, // dt
+	0x9f:  0xb604,  // mark
+	0xa1:  0xb01,   // u
+	0xa4:  0x6ab08, // onunload
+	0xa5:  0x5d04,  // loop
+	0xa6:  0x16408, // disabled
+	0xaa:  0x42307, // onended
+	0xab:  0xb00a,  // malignmark
+	0xad:  0x67b09, // onsuspend
+	0xae:  0x35105, // mtext
+	0xaf:  0x64f06, // onsort
+	0xb0:  0x19d08, // itemprop
+	0xb3:  0x67109, // itemscope
+	0xb4:  0x17305, // blink
+	0xb6:  0x3b106, // ondrag
+	0xb7:  0xa702,  // ul
+	0xb8:  0x26e04, // form
+	0xb9:  0x12907, // sandbox
+	0xba:  0x8b05,  // frame
+	0xbb:  0x1505,  // value
+	0xbc:  0x66209, // onstorage
+	0xbf:  0xaa07,  // acronym
+	0xc0:  0x19a02, // rt
+	0xc2:  0x202,   // br
+	0xc3:  0x22608, // fieldset
+	0xc4:  0x2900d, // typemustmatch
+	0xc5:  0xa208,  // nomodule
+	0xc6:  0x6c07,  // noembed
+	0xc7:  0x69e0d, // onbeforeprint
+	0xc8:  0x19106, // button
+	0xc9:  0x2f507, // onclick
+	0xca:  0x70407, // summary
+	0xcd:  0xfb04,  // ruby
+	0xce:  0x56405, // class
+	0xcf:  0x3f40b, // ondragstart
+	0xd0:  0x23107, // caption
+	0xd4:  0xdd0e,  // allowusermedia
+	0xd5:  0x4cf0b, // onloadstart
+	0xd9:  0x16b03, // div
+	0xda:  0x4a904, // list
+	0xdb:  0x32e04, // math
+	0xdc:  0x44b05, // input
+	0xdf:  0x3ea0a, // ondragover
+	0xe0:  0x2de02, // h2
+	0xe2:  0x1b209, // plaintext
+	0xe4:  0x4f30c, // onmouseenter
+	0xe7:  0x47907, // checked
+	0xe8:  0x47003, // pre
+	0xea:  0x35f08, // multiple
+	0xeb:  0xba03,  // bdi
+	0xec:  0x33d09, // maxlength
+	0xed:  0xcf01,  // q
+	0xee:  0x61f0a, // onauxclick
+	0xf0:  0x57c03, // wbr
+	0xf2:  0x3b04,  // base
+	0xf3:  0x6e306, // option
+	0xf5:  0x41310, // ondurationchange
+	0xf7:  0x8908,  // noframes
+	0xf9:  0x40508, // dropzone
+	0xfb:  0x67505, // scope
+	0xfc:  0x8008,  // reversed
+	0xfd:  0x3ba0b, // ondragenter
+	0xfe:  0x3fa05, // start
+	0xff:  0x12f03, // xmp
+	0x100: 0x5f907, // srclang
+	0x101: 0x30703, // img
+	0x104: 0x101,   // b
+	0x105: 0x25403, // for
+	0x106: 0x10705, // aside
+	0x107: 0x44907, // oninput
+	0x108: 0x35604, // area
+	0x109: 0x2a40a, // formmethod
+	0x10a: 0x72604, // wrap
+	0x10c: 0x23c02, // rp
+	0x10d: 0x46b0a, // onkeypress
+	0x10e: 0x6802,  // tt
+	0x110: 0x34702, // mi
+	0x111: 0x36705, // muted
+	0x112: 0xf303,  // alt
+	0x113: 0x5c504, // code
+	0x114: 0x6e02,  // em
+	0x115: 0x3c50a, // ondragexit
+	0x117: 0x9f04,  // span
+	0x119: 0x6d708, // manifest
+	0x11a: 0x38708, // menuitem
+	0x11b: 0x58b07, // content
+	0x11d: 0x6c109, // onwaiting
+	0x11f: 0x4c609, // onloadend
+	0x121: 0x37e0d, // oncontextmenu
+	0x123: 0x56d06, // onblur
+	0x124: 0x3fc07, // article
+	0x125: 0x9303,  // dir
+	0x126: 0xef04,  // ping
+	0x127: 0x24c08, // required
+	0x128: 0x45509, // oninvalid
+	0x129: 0xb105,  // align
+	0x12b: 0x58a04, // icon
+	0x12c: 0x64d02, // h6
+	0x12d: 0x1c404, // cols
+	0x12e: 0x22e0a, // figcaption
+	0x12f: 0x45e09, // onkeydown
+	0x130: 0x66b08, // onsubmit
+	0x131: 0x14d09, // oncanplay
+	0x132: 0x70b03, // sup
+	0x133: 0xc01,   // p
+	0x135: 0x40a09, // onemptied
+	0x136: 0x39106, // oncopy
+	0x137: 0x19c04, // cite
+	0x138: 0x3a70a, // ondblclick
+	0x13a: 0x50b0b, // onmousemove
+	0x13c: 0x66d03, // sub
+	0x13d: 0x48703, // rel
+	0x13e: 0x5f08,  // optgroup
+	0x142: 0x9c07,  // rowspan
+	0x143: 0x37806, // source
+	0x144: 0x21608, // noscript
+	0x145: 0x1a304, // open
+	0x146: 0x20403, // ins
+	0x147: 0x2540d, // foreignObject
+	0x148: 0x5ad0a, // onpopstate
+	0x14a: 0x28d07, // enctype
+	0x14b: 0x2760e, // onautocomplete
+	0x14c: 0x35208, // textarea
+	0x14e: 0x2780c, // autocomplete
+	0x14f: 0x15702, // hr
+	0x150: 0x1de08, // controls
+	0x151: 0x10902, // id
+	0x153: 0x2360c, // onafterprint
+	0x155: 0x2610d, // foreignobject
+	0x156: 0x32707, // marquee
+	0x157: 0x59a07, // onpause
+	0x158: 0x5e602, // dl
+	0x159: 0x5206,  // height
+	0x15a: 0x34703, // min
+	0x15b: 0x9307,  // dirname
+	0x15c: 0x1f209, // translate
+	0x15d: 0x5604,  // html
+	0x15e: 0x34709, // minlength
+	0x15f: 0x48607, // preload
+	0x160: 0x71408, // template
+	0x161: 0x3df0b, // ondragleave
+	0x162: 0x3a02,  // rb
+	0x164: 0x5c003, // src
+	0x165: 0x6dd06, // strong
+	0x167: 0x7804,  // samp
+	0x168: 0x6f307, // address
+	0x169: 0x55108, // ononline
+	0x16b: 0x1310b, // placeholder
+	0x16c: 0x2c406, // target
+	0x16d: 0x20605, // small
+	0x16e: 0x6ca07, // onwheel
+	0x16f: 0x1c90a, // annotation
+	0x170: 0x4740a, // spellcheck
+	0x171: 0x7207,  // details
+	0x172: 0x10306, // canvas
+	0x173: 0x12109, // autofocus
+	0x174: 0xc05,   // param
+	0x176: 0x46308, // download
+	0x177: 0x45203, // del
+	0x178: 0x36c07, // onclose
+	0x179: 0xb903,  // kbd
+	0x17a: 0x31906, // applet
+	0x17b: 0x2e004, // href
+	0x17c: 0x5f108, // onresize
+	0x17e: 0x49d0c, // onloadeddata
+	0x180: 0xcc02,  // tr
+	0x181: 0x2c00a, // formtarget
+	0x182: 0x11005, // title
+	0x183: 0x6ff05, // style
+	0x184: 0xd206,  // strike
+	0x185: 0x59e06, // usemap
+	0x186: 0x2fc06, // iframe
+	0x187: 0x1004,  // main
+	0x189: 0x7b07,  // picture
+	0x18c: 0x31605, // ismap
+	0x18e: 0x4a504, // data
+	0x18f: 0x5905,  // label
+	0x191: 0x3d10e, // referrerpolicy
+	0x192: 0x15602, // th
+	0x194: 0x53606, // prompt
+	0x195: 0x56807, // section
+	0x197: 0x6d107, // optimum
+	0x198: 0x2db04, // high
+	0x199: 0x15c02, // h1
+	0x19a: 0x65909, // onstalled
+	0x19b: 0x16d03, // var
+	0x19c: 0x4204,  // time
+	0x19e: 0x67402, // ms
+	0x19f: 0x33106, // header
+	0x1a0: 0x4da09, // onmessage
+	0x1a1: 0x1a605, // nonce
+	0x1a2: 0x26e0a, // formaction
+	0x1a3: 0x22006, // center
+	0x1a4: 0x3704,  // nobr
+	0x1a5: 0x59505, // table
+	0x1a6: 0x4a907, // listing
+	0x1a7: 0x18106, // legend
+	0x1a9: 0x29b09, // challenge
+	0x1aa: 0x24806, // figure
+	0x1ab: 0xe605,  // media
+	0x1ae: 0xd904,  // type
+	0x1af: 0x3f04,  // font
+	0x1b0: 0x4da0e, // onmessageerror
+	0x1b1: 0x37108, // seamless
+	0x1b2: 0x8703,  // dfn
+	0x1b3: 0x5c705, // defer
+	0x1b4: 0xc303,  // low
+	0x1b5: 0x19a03, // rtc
+	0x1b6: 0x5230b, // onmouseover
+	0x1b7: 0x2b20a, // novalidate
+	0x1b8: 0x71c0a, // workertype
+	0x1ba: 0x3cd07, // itemref
+	0x1bd: 0x1,     // a
+	0x1be: 0x31803, // map
+	0x1bf: 0x400c,  // ontimeupdate
+	0x1c0: 0x15e07, // bgsound
+	0x1c1: 0x3206,  // keygen
+	0x1c2: 0x2705,  // tbody
+	0x1c5: 0x64406, // onshow
+	0x1c7: 0x2501,  // s
+	0x1c8: 0x6607,  // pattern
+	0x1cc: 0x14d10, // oncanplaythrough
+	0x1ce: 0x2d702, // dd
+	0x1cf: 0x6f906, // srcset
+	0x1d0: 0x17003, // big
+	0x1d2: 0x65108, // sortable
+	0x1d3: 0x48007, // onkeyup
+	0x1d5: 0x5a406, // onplay
+	0x1d7: 0x4b804, // meta
+	0x1d8: 0x40306, // ondrop
+	0x1da: 0x60008, // onscroll
+	0x1db: 0x1fb0b, // crossorigin
+	0x1dc: 0x5730a, // onpageshow
+	0x1dd: 0x4,     // abbr
+	0x1de: 0x9202,  // td
+	0x1df: 0x58b0f, // contenteditable
+	0x1e0: 0x27206, // action
+	0x1e1: 0x1400b, // playsinline
+	0x1e2: 0x43107, // onfocus
+	0x1e3: 0x2e008, // hreflang
+	0x1e5: 0x5160a, // onmouseout
+	0x1e6: 0x5ea07, // onreset
+	0x1e7: 0x13c08, // autoplay
+	0x1e8: 0x63109, // onseeking
+	0x1ea: 0x67506, // scoped
+	0x1ec: 0x30a,   // radiogroup
+	0x1ee: 0x3800b, // contextmenu
+	0x1ef: 0x52e09, // onmouseup
+	0x1f1: 0x2ca06, // hgroup
+	0x1f2: 0x2080f, // allowfullscreen
+	0x1f3: 0x4be08, // tabindex
+	0x1f6: 0x30f07, // isindex
+	0x1f7: 0x1a0e,  // accept-charset
+	0x1f8: 0x2ae0e, // formnovalidate
+	0x1fb: 0x1c90e, // annotation-xml
+	0x1fc: 0x6e05,  // embed
+	0x1fd: 0x21806, // script
+	0x1fe: 0xbb06,  // dialog
+	0x1ff: 0x1d707, // command
+}
+
+const atomText = "abbradiogrouparamainavalueaccept-charsetbodyaccesskeygenobrb" +
+	"asefontimeupdateviacacheightmlabelooptgroupatternoembedetail" +
+	"sampictureversedfnoframesetdirnameterowspanomoduleacronymali" +
+	"gnmarkbdialogallowpaymentrequestrikeytypeallowusermediagroup" +
+	"ingaltfooterubyasyncanvasidefaultitleaudioncancelautofocusan" +
+	"dboxmplaceholderautoplaysinlinebdoncanplaythrough1bgsoundisa" +
+	"bledivarbigblinkindraggablegendblockquotebuttonabortcitempro" +
+	"penoncecolgrouplaintextrackcolorcolspannotation-xmlcommandco" +
+	"ntrolshapecoordslotranslatecrossoriginsmallowfullscreenoscri" +
+	"ptfacenterfieldsetfigcaptionafterprintegrityfigurequiredfore" +
+	"ignObjectforeignobjectformactionautocompleteerrorformenctype" +
+	"mustmatchallengeformmethodformnovalidatetimeformtargethgroup" +
+	"osterhiddenhigh2hreflanghttp-equivideonclickiframeimageimgly" +
+	"ph3isindexismappletitemtypemarqueematheadersortedmaxlength4m" +
+	"inlength5mtextareadonlymultiplemutedoncloseamlessourceoncont" +
+	"extmenuitemidoncopyoncuechangeoncutondblclickondragendondrag" +
+	"enterondragexitemreferrerpolicyondragleaveondragoverondragst" +
+	"articleondropzonemptiedondurationchangeonendedonerroronfocus" +
+	"paceronhashchangeoninputmodeloninvalidonkeydownloadonkeypres" +
+	"spellcheckedonkeyupreloadonlanguagechangeonloadeddatalisting" +
+	"onloadedmetadatabindexonloadendonloadstartonmessageerroronmo" +
+	"usedownonmouseenteronmouseleaveonmousemoveonmouseoutputonmou" +
+	"seoveronmouseupromptonmousewheelonofflineononlineonpagehides" +
+	"classectionbluronpageshowbronpastepublicontenteditableonpaus" +
+	"emaponplayingonpopstateonprogressrcdocodeferonratechangeonre" +
+	"jectionhandledonresetonresizesrclangonscrollonsecuritypolicy" +
+	"violationauxclickonseekedonseekingonselectedonshowidth6onsor" +
+	"tableonstalledonstorageonsubmitemscopedonsuspendontoggleonun" +
+	"handledrejectionbeforeprintonunloadonvolumechangeonwaitingon" +
+	"wheeloptimumanifestrongoptionbeforeunloaddressrcsetstylesumm" +
+	"arysupsvgsystemplateworkertypewrap"
diff --git a/backend/vendor/golang.org/x/net/html/const.go b/backend/vendor/golang.org/x/net/html/const.go
new file mode 100644
index 00000000..a3a918f0
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/const.go
@@ -0,0 +1,112 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+// Section 12.2.4.2 of the HTML5 specification says "The following elements
+// have varying levels of special parsing rules".
+// https://html.spec.whatwg.org/multipage/syntax.html#the-stack-of-open-elements
+var isSpecialElementMap = map[string]bool{
+	"address":    true,
+	"applet":     true,
+	"area":       true,
+	"article":    true,
+	"aside":      true,
+	"base":       true,
+	"basefont":   true,
+	"bgsound":    true,
+	"blockquote": true,
+	"body":       true,
+	"br":         true,
+	"button":     true,
+	"caption":    true,
+	"center":     true,
+	"col":        true,
+	"colgroup":   true,
+	"dd":         true,
+	"details":    true,
+	"dir":        true,
+	"div":        true,
+	"dl":         true,
+	"dt":         true,
+	"embed":      true,
+	"fieldset":   true,
+	"figcaption": true,
+	"figure":     true,
+	"footer":     true,
+	"form":       true,
+	"frame":      true,
+	"frameset":   true,
+	"h1":         true,
+	"h2":         true,
+	"h3":         true,
+	"h4":         true,
+	"h5":         true,
+	"h6":         true,
+	"head":       true,
+	"header":     true,
+	"hgroup":     true,
+	"hr":         true,
+	"html":       true,
+	"iframe":     true,
+	"img":        true,
+	"input":      true,
+	"isindex":    true, // The 'isindex' element has been removed, but keep it for backwards compatibility.
+	"keygen":     true,
+	"li":         true,
+	"link":       true,
+	"listing":    true,
+	"main":       true,
+	"marquee":    true,
+	"menu":       true,
+	"meta":       true,
+	"nav":        true,
+	"noembed":    true,
+	"noframes":   true,
+	"noscript":   true,
+	"object":     true,
+	"ol":         true,
+	"p":          true,
+	"param":      true,
+	"plaintext":  true,
+	"pre":        true,
+	"script":     true,
+	"section":    true,
+	"select":     true,
+	"source":     true,
+	"style":      true,
+	"summary":    true,
+	"table":      true,
+	"tbody":      true,
+	"td":         true,
+	"template":   true,
+	"textarea":   true,
+	"tfoot":      true,
+	"th":         true,
+	"thead":      true,
+	"title":      true,
+	"tr":         true,
+	"track":      true,
+	"ul":         true,
+	"wbr":        true,
+	"xmp":        true,
+}
+
+func isSpecialElement(element *Node) bool {
+	switch element.Namespace {
+	case "", "html":
+		return isSpecialElementMap[element.Data]
+	case "math":
+		switch element.Data {
+		case "mi", "mo", "mn", "ms", "mtext", "annotation-xml":
+			return true
+		}
+	case "svg":
+		switch element.Data {
+		case "foreignObject", "desc", "title":
+			return true
+		}
+	}
+	return false
+}
diff --git a/backend/vendor/golang.org/x/net/html/doc.go b/backend/vendor/golang.org/x/net/html/doc.go
new file mode 100644
index 00000000..822ed42a
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/doc.go
@@ -0,0 +1,106 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+/*
+Package html implements an HTML5-compliant tokenizer and parser.
+
+Tokenization is done by creating a Tokenizer for an io.Reader r. It is the
+caller's responsibility to ensure that r provides UTF-8 encoded HTML.
+
+	z := html.NewTokenizer(r)
+
+Given a Tokenizer z, the HTML is tokenized by repeatedly calling z.Next(),
+which parses the next token and returns its type, or an error:
+
+	for {
+		tt := z.Next()
+		if tt == html.ErrorToken {
+			// ...
+			return ...
+		}
+		// Process the current token.
+	}
+
+There are two APIs for retrieving the current token. The high-level API is to
+call Token; the low-level API is to call Text or TagName / TagAttr. Both APIs
+allow optionally calling Raw after Next but before Token, Text, TagName, or
+TagAttr. In EBNF notation, the valid call sequence per token is:
+
+	Next {Raw} [ Token | Text | TagName {TagAttr} ]
+
+Token returns an independent data structure that completely describes a token.
+Entities (such as "&lt;") are unescaped, tag names and attribute keys are
+lower-cased, and attributes are collected into a []Attribute. For example:
+
+	for {
+		if z.Next() == html.ErrorToken {
+			// Returning io.EOF indicates success.
+			return z.Err()
+		}
+		emitToken(z.Token())
+	}
+
+The low-level API performs fewer allocations and copies, but the contents of
+the []byte values returned by Text, TagName and TagAttr may change on the next
+call to Next. For example, to extract an HTML page's anchor text:
+
+	depth := 0
+	for {
+		tt := z.Next()
+		switch tt {
+		case html.ErrorToken:
+			return z.Err()
+		case html.TextToken:
+			if depth > 0 {
+				// emitBytes should copy the []byte it receives,
+				// if it doesn't process it immediately.
+				emitBytes(z.Text())
+			}
+		case html.StartTagToken, html.EndTagToken:
+			tn, _ := z.TagName()
+			if len(tn) == 1 && tn[0] == 'a' {
+				if tt == html.StartTagToken {
+					depth++
+				} else {
+					depth--
+				}
+			}
+		}
+	}
+
+Parsing is done by calling Parse with an io.Reader, which returns the root of
+the parse tree (the document element) as a *Node. It is the caller's
+responsibility to ensure that the Reader provides UTF-8 encoded HTML. For
+example, to process each anchor node in depth-first order:
+
+	doc, err := html.Parse(r)
+	if err != nil {
+		// ...
+	}
+	var f func(*html.Node)
+	f = func(n *html.Node) {
+		if n.Type == html.ElementNode && n.Data == "a" {
+			// Do something with n...
+		}
+		for c := n.FirstChild; c != nil; c = c.NextSibling {
+			f(c)
+		}
+	}
+	f(doc)
+
+The relevant specifications include:
+https://html.spec.whatwg.org/multipage/syntax.html and
+https://html.spec.whatwg.org/multipage/syntax.html#tokenization
+*/
+package html // import "golang.org/x/net/html"
+
+// The tokenization algorithm implemented by this package is not a line-by-line
+// transliteration of the relatively verbose state-machine in the WHATWG
+// specification. A more direct approach is used instead, where the program
+// counter implies the state, such as whether it is tokenizing a tag or a text
+// node. Specification compliance is verified by checking expected and actual
+// outputs over a test suite rather than aiming for algorithmic fidelity.
+
+// TODO(nigeltao): Does a DOM API belong in this package or a separate one?
+// TODO(nigeltao): How does parsing interact with a JavaScript engine?
diff --git a/backend/vendor/golang.org/x/net/html/doctype.go b/backend/vendor/golang.org/x/net/html/doctype.go
new file mode 100644
index 00000000..c484e5a9
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/doctype.go
@@ -0,0 +1,156 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"strings"
+)
+
+// parseDoctype parses the data from a DoctypeToken into a name,
+// public identifier, and system identifier. It returns a Node whose Type
+// is DoctypeNode, whose Data is the name, and which has attributes
+// named "system" and "public" for the two identifiers if they were present.
+// quirks is whether the document should be parsed in "quirks mode".
+func parseDoctype(s string) (n *Node, quirks bool) {
+	n = &Node{Type: DoctypeNode}
+
+	// Find the name.
+	space := strings.IndexAny(s, whitespace)
+	if space == -1 {
+		space = len(s)
+	}
+	n.Data = s[:space]
+	// The comparison to "html" is case-sensitive.
+	if n.Data != "html" {
+		quirks = true
+	}
+	n.Data = strings.ToLower(n.Data)
+	s = strings.TrimLeft(s[space:], whitespace)
+
+	if len(s) < 6 {
+		// It can't start with "PUBLIC" or "SYSTEM".
+		// Ignore the rest of the string.
+		return n, quirks || s != ""
+	}
+
+	key := strings.ToLower(s[:6])
+	s = s[6:]
+	for key == "public" || key == "system" {
+		s = strings.TrimLeft(s, whitespace)
+		if s == "" {
+			break
+		}
+		quote := s[0]
+		if quote != '"' && quote != '\'' {
+			break
+		}
+		s = s[1:]
+		q := strings.IndexRune(s, rune(quote))
+		var id string
+		if q == -1 {
+			id = s
+			s = ""
+		} else {
+			id = s[:q]
+			s = s[q+1:]
+		}
+		n.Attr = append(n.Attr, Attribute{Key: key, Val: id})
+		if key == "public" {
+			key = "system"
+		} else {
+			key = ""
+		}
+	}
+
+	if key != "" || s != "" {
+		quirks = true
+	} else if len(n.Attr) > 0 {
+		if n.Attr[0].Key == "public" {
+			public := strings.ToLower(n.Attr[0].Val)
+			switch public {
+			case "-//w3o//dtd w3 html strict 3.0//en//", "-/w3d/dtd html 4.0 transitional/en", "html":
+				quirks = true
+			default:
+				for _, q := range quirkyIDs {
+					if strings.HasPrefix(public, q) {
+						quirks = true
+						break
+					}
+				}
+			}
+			// The following two public IDs only cause quirks mode if there is no system ID.
+			if len(n.Attr) == 1 && (strings.HasPrefix(public, "-//w3c//dtd html 4.01 frameset//") ||
+				strings.HasPrefix(public, "-//w3c//dtd html 4.01 transitional//")) {
+				quirks = true
+			}
+		}
+		if lastAttr := n.Attr[len(n.Attr)-1]; lastAttr.Key == "system" &&
+			strings.ToLower(lastAttr.Val) == "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd" {
+			quirks = true
+		}
+	}
+
+	return n, quirks
+}
+
+// quirkyIDs is a list of public doctype identifiers that cause a document
+// to be interpreted in quirks mode. The identifiers should be in lower case.
+var quirkyIDs = []string{
+	"+//silmaril//dtd html pro v0r11 19970101//",
+	"-//advasoft ltd//dtd html 3.0 aswedit + extensions//",
+	"-//as//dtd html 3.0 aswedit + extensions//",
+	"-//ietf//dtd html 2.0 level 1//",
+	"-//ietf//dtd html 2.0 level 2//",
+	"-//ietf//dtd html 2.0 strict level 1//",
+	"-//ietf//dtd html 2.0 strict level 2//",
+	"-//ietf//dtd html 2.0 strict//",
+	"-//ietf//dtd html 2.0//",
+	"-//ietf//dtd html 2.1e//",
+	"-//ietf//dtd html 3.0//",
+	"-//ietf//dtd html 3.2 final//",
+	"-//ietf//dtd html 3.2//",
+	"-//ietf//dtd html 3//",
+	"-//ietf//dtd html level 0//",
+	"-//ietf//dtd html level 1//",
+	"-//ietf//dtd html level 2//",
+	"-//ietf//dtd html level 3//",
+	"-//ietf//dtd html strict level 0//",
+	"-//ietf//dtd html strict level 1//",
+	"-//ietf//dtd html strict level 2//",
+	"-//ietf//dtd html strict level 3//",
+	"-//ietf//dtd html strict//",
+	"-//ietf//dtd html//",
+	"-//metrius//dtd metrius presentational//",
+	"-//microsoft//dtd internet explorer 2.0 html strict//",
+	"-//microsoft//dtd internet explorer 2.0 html//",
+	"-//microsoft//dtd internet explorer 2.0 tables//",
+	"-//microsoft//dtd internet explorer 3.0 html strict//",
+	"-//microsoft//dtd internet explorer 3.0 html//",
+	"-//microsoft//dtd internet explorer 3.0 tables//",
+	"-//netscape comm. corp.//dtd html//",
+	"-//netscape comm. corp.//dtd strict html//",
+	"-//o'reilly and associates//dtd html 2.0//",
+	"-//o'reilly and associates//dtd html extended 1.0//",
+	"-//o'reilly and associates//dtd html extended relaxed 1.0//",
+	"-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//",
+	"-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//",
+	"-//spyglass//dtd html 2.0 extended//",
+	"-//sq//dtd html 2.0 hotmetal + extensions//",
+	"-//sun microsystems corp.//dtd hotjava html//",
+	"-//sun microsystems corp.//dtd hotjava strict html//",
+	"-//w3c//dtd html 3 1995-03-24//",
+	"-//w3c//dtd html 3.2 draft//",
+	"-//w3c//dtd html 3.2 final//",
+	"-//w3c//dtd html 3.2//",
+	"-//w3c//dtd html 3.2s draft//",
+	"-//w3c//dtd html 4.0 frameset//",
+	"-//w3c//dtd html 4.0 transitional//",
+	"-//w3c//dtd html experimental 19960712//",
+	"-//w3c//dtd html experimental 970421//",
+	"-//w3c//dtd w3 html//",
+	"-//w3o//dtd w3 html 3.0//",
+	"-//webtechs//dtd mozilla html 2.0//",
+	"-//webtechs//dtd mozilla html//",
+}
diff --git a/backend/vendor/golang.org/x/net/html/entity.go b/backend/vendor/golang.org/x/net/html/entity.go
new file mode 100644
index 00000000..b628880a
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/entity.go
@@ -0,0 +1,2253 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+// All entities that do not end with ';' are 6 or fewer bytes long.
+const longestEntityWithoutSemicolon = 6
+
+// entity is a map from HTML entity names to their values. The semicolon matters:
+// https://html.spec.whatwg.org/multipage/syntax.html#named-character-references
+// lists both "amp" and "amp;" as two separate entries.
+//
+// Note that the HTML5 list is larger than the HTML4 list at
+// http://www.w3.org/TR/html4/sgml/entities.html
+var entity = map[string]rune{
+	"AElig;":                           '\U000000C6',
+	"AMP;":                             '\U00000026',
+	"Aacute;":                          '\U000000C1',
+	"Abreve;":                          '\U00000102',
+	"Acirc;":                           '\U000000C2',
+	"Acy;":                             '\U00000410',
+	"Afr;":                             '\U0001D504',
+	"Agrave;":                          '\U000000C0',
+	"Alpha;":                           '\U00000391',
+	"Amacr;":                           '\U00000100',
+	"And;":                             '\U00002A53',
+	"Aogon;":                           '\U00000104',
+	"Aopf;":                            '\U0001D538',
+	"ApplyFunction;":                   '\U00002061',
+	"Aring;":                           '\U000000C5',
+	"Ascr;":                            '\U0001D49C',
+	"Assign;":                          '\U00002254',
+	"Atilde;":                          '\U000000C3',
+	"Auml;":                            '\U000000C4',
+	"Backslash;":                       '\U00002216',
+	"Barv;":                            '\U00002AE7',
+	"Barwed;":                          '\U00002306',
+	"Bcy;":                             '\U00000411',
+	"Because;":                         '\U00002235',
+	"Bernoullis;":                      '\U0000212C',
+	"Beta;":                            '\U00000392',
+	"Bfr;":                             '\U0001D505',
+	"Bopf;":                            '\U0001D539',
+	"Breve;":                           '\U000002D8',
+	"Bscr;":                            '\U0000212C',
+	"Bumpeq;":                          '\U0000224E',
+	"CHcy;":                            '\U00000427',
+	"COPY;":                            '\U000000A9',
+	"Cacute;":                          '\U00000106',
+	"Cap;":                             '\U000022D2',
+	"CapitalDifferentialD;":            '\U00002145',
+	"Cayleys;":                         '\U0000212D',
+	"Ccaron;":                          '\U0000010C',
+	"Ccedil;":                          '\U000000C7',
+	"Ccirc;":                           '\U00000108',
+	"Cconint;":                         '\U00002230',
+	"Cdot;":                            '\U0000010A',
+	"Cedilla;":                         '\U000000B8',
+	"CenterDot;":                       '\U000000B7',
+	"Cfr;":                             '\U0000212D',
+	"Chi;":                             '\U000003A7',
+	"CircleDot;":                       '\U00002299',
+	"CircleMinus;":                     '\U00002296',
+	"CirclePlus;":                      '\U00002295',
+	"CircleTimes;":                     '\U00002297',
+	"ClockwiseContourIntegral;":        '\U00002232',
+	"CloseCurlyDoubleQuote;":           '\U0000201D',
+	"CloseCurlyQuote;":                 '\U00002019',
+	"Colon;":                           '\U00002237',
+	"Colone;":                          '\U00002A74',
+	"Congruent;":                       '\U00002261',
+	"Conint;":                          '\U0000222F',
+	"ContourIntegral;":                 '\U0000222E',
+	"Copf;":                            '\U00002102',
+	"Coproduct;":                       '\U00002210',
+	"CounterClockwiseContourIntegral;": '\U00002233',
+	"Cross;":                           '\U00002A2F',
+	"Cscr;":                            '\U0001D49E',
+	"Cup;":                             '\U000022D3',
+	"CupCap;":                          '\U0000224D',
+	"DD;":                              '\U00002145',
+	"DDotrahd;":                        '\U00002911',
+	"DJcy;":                            '\U00000402',
+	"DScy;":                            '\U00000405',
+	"DZcy;":                            '\U0000040F',
+	"Dagger;":                          '\U00002021',
+	"Darr;":                            '\U000021A1',
+	"Dashv;":                           '\U00002AE4',
+	"Dcaron;":                          '\U0000010E',
+	"Dcy;":                             '\U00000414',
+	"Del;":                             '\U00002207',
+	"Delta;":                           '\U00000394',
+	"Dfr;":                             '\U0001D507',
+	"DiacriticalAcute;":                '\U000000B4',
+	"DiacriticalDot;":                  '\U000002D9',
+	"DiacriticalDoubleAcute;":          '\U000002DD',
+	"DiacriticalGrave;":                '\U00000060',
+	"DiacriticalTilde;":                '\U000002DC',
+	"Diamond;":                         '\U000022C4',
+	"DifferentialD;":                   '\U00002146',
+	"Dopf;":                            '\U0001D53B',
+	"Dot;":                             '\U000000A8',
+	"DotDot;":                          '\U000020DC',
+	"DotEqual;":                        '\U00002250',
+	"DoubleContourIntegral;":           '\U0000222F',
+	"DoubleDot;":                       '\U000000A8',
+	"DoubleDownArrow;":                 '\U000021D3',
+	"DoubleLeftArrow;":                 '\U000021D0',
+	"DoubleLeftRightArrow;":            '\U000021D4',
+	"DoubleLeftTee;":                   '\U00002AE4',
+	"DoubleLongLeftArrow;":             '\U000027F8',
+	"DoubleLongLeftRightArrow;":        '\U000027FA',
+	"DoubleLongRightArrow;":            '\U000027F9',
+	"DoubleRightArrow;":                '\U000021D2',
+	"DoubleRightTee;":                  '\U000022A8',
+	"DoubleUpArrow;":                   '\U000021D1',
+	"DoubleUpDownArrow;":               '\U000021D5',
+	"DoubleVerticalBar;":               '\U00002225',
+	"DownArrow;":                       '\U00002193',
+	"DownArrowBar;":                    '\U00002913',
+	"DownArrowUpArrow;":                '\U000021F5',
+	"DownBreve;":                       '\U00000311',
+	"DownLeftRightVector;":             '\U00002950',
+	"DownLeftTeeVector;":               '\U0000295E',
+	"DownLeftVector;":                  '\U000021BD',
+	"DownLeftVectorBar;":               '\U00002956',
+	"DownRightTeeVector;":              '\U0000295F',
+	"DownRightVector;":                 '\U000021C1',
+	"DownRightVectorBar;":              '\U00002957',
+	"DownTee;":                         '\U000022A4',
+	"DownTeeArrow;":                    '\U000021A7',
+	"Downarrow;":                       '\U000021D3',
+	"Dscr;":                            '\U0001D49F',
+	"Dstrok;":                          '\U00000110',
+	"ENG;":                             '\U0000014A',
+	"ETH;":                             '\U000000D0',
+	"Eacute;":                          '\U000000C9',
+	"Ecaron;":                          '\U0000011A',
+	"Ecirc;":                           '\U000000CA',
+	"Ecy;":                             '\U0000042D',
+	"Edot;":                            '\U00000116',
+	"Efr;":                             '\U0001D508',
+	"Egrave;":                          '\U000000C8',
+	"Element;":                         '\U00002208',
+	"Emacr;":                           '\U00000112',
+	"EmptySmallSquare;":                '\U000025FB',
+	"EmptyVerySmallSquare;":            '\U000025AB',
+	"Eogon;":                           '\U00000118',
+	"Eopf;":                            '\U0001D53C',
+	"Epsilon;":                         '\U00000395',
+	"Equal;":                           '\U00002A75',
+	"EqualTilde;":                      '\U00002242',
+	"Equilibrium;":                     '\U000021CC',
+	"Escr;":                            '\U00002130',
+	"Esim;":                            '\U00002A73',
+	"Eta;":                             '\U00000397',
+	"Euml;":                            '\U000000CB',
+	"Exists;":                          '\U00002203',
+	"ExponentialE;":                    '\U00002147',
+	"Fcy;":                             '\U00000424',
+	"Ffr;":                             '\U0001D509',
+	"FilledSmallSquare;":               '\U000025FC',
+	"FilledVerySmallSquare;":           '\U000025AA',
+	"Fopf;":                            '\U0001D53D',
+	"ForAll;":                          '\U00002200',
+	"Fouriertrf;":                      '\U00002131',
+	"Fscr;":                            '\U00002131',
+	"GJcy;":                            '\U00000403',
+	"GT;":                              '\U0000003E',
+	"Gamma;":                           '\U00000393',
+	"Gammad;":                          '\U000003DC',
+	"Gbreve;":                          '\U0000011E',
+	"Gcedil;":                          '\U00000122',
+	"Gcirc;":                           '\U0000011C',
+	"Gcy;":                             '\U00000413',
+	"Gdot;":                            '\U00000120',
+	"Gfr;":                             '\U0001D50A',
+	"Gg;":                              '\U000022D9',
+	"Gopf;":                            '\U0001D53E',
+	"GreaterEqual;":                    '\U00002265',
+	"GreaterEqualLess;":                '\U000022DB',
+	"GreaterFullEqual;":                '\U00002267',
+	"GreaterGreater;":                  '\U00002AA2',
+	"GreaterLess;":                     '\U00002277',
+	"GreaterSlantEqual;":               '\U00002A7E',
+	"GreaterTilde;":                    '\U00002273',
+	"Gscr;":                            '\U0001D4A2',
+	"Gt;":                              '\U0000226B',
+	"HARDcy;":                          '\U0000042A',
+	"Hacek;":                           '\U000002C7',
+	"Hat;":                             '\U0000005E',
+	"Hcirc;":                           '\U00000124',
+	"Hfr;":                             '\U0000210C',
+	"HilbertSpace;":                    '\U0000210B',
+	"Hopf;":                            '\U0000210D',
+	"HorizontalLine;":                  '\U00002500',
+	"Hscr;":                            '\U0000210B',
+	"Hstrok;":                          '\U00000126',
+	"HumpDownHump;":                    '\U0000224E',
+	"HumpEqual;":                       '\U0000224F',
+	"IEcy;":                            '\U00000415',
+	"IJlig;":                           '\U00000132',
+	"IOcy;":                            '\U00000401',
+	"Iacute;":                          '\U000000CD',
+	"Icirc;":                           '\U000000CE',
+	"Icy;":                             '\U00000418',
+	"Idot;":                            '\U00000130',
+	"Ifr;":                             '\U00002111',
+	"Igrave;":                          '\U000000CC',
+	"Im;":                              '\U00002111',
+	"Imacr;":                           '\U0000012A',
+	"ImaginaryI;":                      '\U00002148',
+	"Implies;":                         '\U000021D2',
+	"Int;":                             '\U0000222C',
+	"Integral;":                        '\U0000222B',
+	"Intersection;":                    '\U000022C2',
+	"InvisibleComma;":                  '\U00002063',
+	"InvisibleTimes;":                  '\U00002062',
+	"Iogon;":                           '\U0000012E',
+	"Iopf;":                            '\U0001D540',
+	"Iota;":                            '\U00000399',
+	"Iscr;":                            '\U00002110',
+	"Itilde;":                          '\U00000128',
+	"Iukcy;":                           '\U00000406',
+	"Iuml;":                            '\U000000CF',
+	"Jcirc;":                           '\U00000134',
+	"Jcy;":                             '\U00000419',
+	"Jfr;":                             '\U0001D50D',
+	"Jopf;":                            '\U0001D541',
+	"Jscr;":                            '\U0001D4A5',
+	"Jsercy;":                          '\U00000408',
+	"Jukcy;":                           '\U00000404',
+	"KHcy;":                            '\U00000425',
+	"KJcy;":                            '\U0000040C',
+	"Kappa;":                           '\U0000039A',
+	"Kcedil;":                          '\U00000136',
+	"Kcy;":                             '\U0000041A',
+	"Kfr;":                             '\U0001D50E',
+	"Kopf;":                            '\U0001D542',
+	"Kscr;":                            '\U0001D4A6',
+	"LJcy;":                            '\U00000409',
+	"LT;":                              '\U0000003C',
+	"Lacute;":                          '\U00000139',
+	"Lambda;":                          '\U0000039B',
+	"Lang;":                            '\U000027EA',
+	"Laplacetrf;":                      '\U00002112',
+	"Larr;":                            '\U0000219E',
+	"Lcaron;":                          '\U0000013D',
+	"Lcedil;":                          '\U0000013B',
+	"Lcy;":                             '\U0000041B',
+	"LeftAngleBracket;":                '\U000027E8',
+	"LeftArrow;":                       '\U00002190',
+	"LeftArrowBar;":                    '\U000021E4',
+	"LeftArrowRightArrow;":             '\U000021C6',
+	"LeftCeiling;":                     '\U00002308',
+	"LeftDoubleBracket;":               '\U000027E6',
+	"LeftDownTeeVector;":               '\U00002961',
+	"LeftDownVector;":                  '\U000021C3',
+	"LeftDownVectorBar;":               '\U00002959',
+	"LeftFloor;":                       '\U0000230A',
+	"LeftRightArrow;":                  '\U00002194',
+	"LeftRightVector;":                 '\U0000294E',
+	"LeftTee;":                         '\U000022A3',
+	"LeftTeeArrow;":                    '\U000021A4',
+	"LeftTeeVector;":                   '\U0000295A',
+	"LeftTriangle;":                    '\U000022B2',
+	"LeftTriangleBar;":                 '\U000029CF',
+	"LeftTriangleEqual;":               '\U000022B4',
+	"LeftUpDownVector;":                '\U00002951',
+	"LeftUpTeeVector;":                 '\U00002960',
+	"LeftUpVector;":                    '\U000021BF',
+	"LeftUpVectorBar;":                 '\U00002958',
+	"LeftVector;":                      '\U000021BC',
+	"LeftVectorBar;":                   '\U00002952',
+	"Leftarrow;":                       '\U000021D0',
+	"Leftrightarrow;":                  '\U000021D4',
+	"LessEqualGreater;":                '\U000022DA',
+	"LessFullEqual;":                   '\U00002266',
+	"LessGreater;":                     '\U00002276',
+	"LessLess;":                        '\U00002AA1',
+	"LessSlantEqual;":                  '\U00002A7D',
+	"LessTilde;":                       '\U00002272',
+	"Lfr;":                             '\U0001D50F',
+	"Ll;":                              '\U000022D8',
+	"Lleftarrow;":                      '\U000021DA',
+	"Lmidot;":                          '\U0000013F',
+	"LongLeftArrow;":                   '\U000027F5',
+	"LongLeftRightArrow;":              '\U000027F7',
+	"LongRightArrow;":                  '\U000027F6',
+	"Longleftarrow;":                   '\U000027F8',
+	"Longleftrightarrow;":              '\U000027FA',
+	"Longrightarrow;":                  '\U000027F9',
+	"Lopf;":                            '\U0001D543',
+	"LowerLeftArrow;":                  '\U00002199',
+	"LowerRightArrow;":                 '\U00002198',
+	"Lscr;":                            '\U00002112',
+	"Lsh;":                             '\U000021B0',
+	"Lstrok;":                          '\U00000141',
+	"Lt;":                              '\U0000226A',
+	"Map;":                             '\U00002905',
+	"Mcy;":                             '\U0000041C',
+	"MediumSpace;":                     '\U0000205F',
+	"Mellintrf;":                       '\U00002133',
+	"Mfr;":                             '\U0001D510',
+	"MinusPlus;":                       '\U00002213',
+	"Mopf;":                            '\U0001D544',
+	"Mscr;":                            '\U00002133',
+	"Mu;":                              '\U0000039C',
+	"NJcy;":                            '\U0000040A',
+	"Nacute;":                          '\U00000143',
+	"Ncaron;":                          '\U00000147',
+	"Ncedil;":                          '\U00000145',
+	"Ncy;":                             '\U0000041D',
+	"NegativeMediumSpace;":             '\U0000200B',
+	"NegativeThickSpace;":              '\U0000200B',
+	"NegativeThinSpace;":               '\U0000200B',
+	"NegativeVeryThinSpace;":           '\U0000200B',
+	"NestedGreaterGreater;":            '\U0000226B',
+	"NestedLessLess;":                  '\U0000226A',
+	"NewLine;":                         '\U0000000A',
+	"Nfr;":                             '\U0001D511',
+	"NoBreak;":                         '\U00002060',
+	"NonBreakingSpace;":                '\U000000A0',
+	"Nopf;":                            '\U00002115',
+	"Not;":                             '\U00002AEC',
+	"NotCongruent;":                    '\U00002262',
+	"NotCupCap;":                       '\U0000226D',
+	"NotDoubleVerticalBar;":            '\U00002226',
+	"NotElement;":                      '\U00002209',
+	"NotEqual;":                        '\U00002260',
+	"NotExists;":                       '\U00002204',
+	"NotGreater;":                      '\U0000226F',
+	"NotGreaterEqual;":                 '\U00002271',
+	"NotGreaterLess;":                  '\U00002279',
+	"NotGreaterTilde;":                 '\U00002275',
+	"NotLeftTriangle;":                 '\U000022EA',
+	"NotLeftTriangleEqual;":            '\U000022EC',
+	"NotLess;":                         '\U0000226E',
+	"NotLessEqual;":                    '\U00002270',
+	"NotLessGreater;":                  '\U00002278',
+	"NotLessTilde;":                    '\U00002274',
+	"NotPrecedes;":                     '\U00002280',
+	"NotPrecedesSlantEqual;":           '\U000022E0',
+	"NotReverseElement;":               '\U0000220C',
+	"NotRightTriangle;":                '\U000022EB',
+	"NotRightTriangleEqual;":           '\U000022ED',
+	"NotSquareSubsetEqual;":            '\U000022E2',
+	"NotSquareSupersetEqual;":          '\U000022E3',
+	"NotSubsetEqual;":                  '\U00002288',
+	"NotSucceeds;":                     '\U00002281',
+	"NotSucceedsSlantEqual;":           '\U000022E1',
+	"NotSupersetEqual;":                '\U00002289',
+	"NotTilde;":                        '\U00002241',
+	"NotTildeEqual;":                   '\U00002244',
+	"NotTildeFullEqual;":               '\U00002247',
+	"NotTildeTilde;":                   '\U00002249',
+	"NotVerticalBar;":                  '\U00002224',
+	"Nscr;":                            '\U0001D4A9',
+	"Ntilde;":                          '\U000000D1',
+	"Nu;":                              '\U0000039D',
+	"OElig;":                           '\U00000152',
+	"Oacute;":                          '\U000000D3',
+	"Ocirc;":                           '\U000000D4',
+	"Ocy;":                             '\U0000041E',
+	"Odblac;":                          '\U00000150',
+	"Ofr;":                             '\U0001D512',
+	"Ograve;":                          '\U000000D2',
+	"Omacr;":                           '\U0000014C',
+	"Omega;":                           '\U000003A9',
+	"Omicron;":                         '\U0000039F',
+	"Oopf;":                            '\U0001D546',
+	"OpenCurlyDoubleQuote;":            '\U0000201C',
+	"OpenCurlyQuote;":                  '\U00002018',
+	"Or;":                              '\U00002A54',
+	"Oscr;":                            '\U0001D4AA',
+	"Oslash;":                          '\U000000D8',
+	"Otilde;":                          '\U000000D5',
+	"Otimes;":                          '\U00002A37',
+	"Ouml;":                            '\U000000D6',
+	"OverBar;":                         '\U0000203E',
+	"OverBrace;":                       '\U000023DE',
+	"OverBracket;":                     '\U000023B4',
+	"OverParenthesis;":                 '\U000023DC',
+	"PartialD;":                        '\U00002202',
+	"Pcy;":                             '\U0000041F',
+	"Pfr;":                             '\U0001D513',
+	"Phi;":                             '\U000003A6',
+	"Pi;":                              '\U000003A0',
+	"PlusMinus;":                       '\U000000B1',
+	"Poincareplane;":                   '\U0000210C',
+	"Popf;":                            '\U00002119',
+	"Pr;":                              '\U00002ABB',
+	"Precedes;":                        '\U0000227A',
+	"PrecedesEqual;":                   '\U00002AAF',
+	"PrecedesSlantEqual;":              '\U0000227C',
+	"PrecedesTilde;":                   '\U0000227E',
+	"Prime;":                           '\U00002033',
+	"Product;":                         '\U0000220F',
+	"Proportion;":                      '\U00002237',
+	"Proportional;":                    '\U0000221D',
+	"Pscr;":                            '\U0001D4AB',
+	"Psi;":                             '\U000003A8',
+	"QUOT;":                            '\U00000022',
+	"Qfr;":                             '\U0001D514',
+	"Qopf;":                            '\U0000211A',
+	"Qscr;":                            '\U0001D4AC',
+	"RBarr;":                           '\U00002910',
+	"REG;":                             '\U000000AE',
+	"Racute;":                          '\U00000154',
+	"Rang;":                            '\U000027EB',
+	"Rarr;":                            '\U000021A0',
+	"Rarrtl;":                          '\U00002916',
+	"Rcaron;":                          '\U00000158',
+	"Rcedil;":                          '\U00000156',
+	"Rcy;":                             '\U00000420',
+	"Re;":                              '\U0000211C',
+	"ReverseElement;":                  '\U0000220B',
+	"ReverseEquilibrium;":              '\U000021CB',
+	"ReverseUpEquilibrium;":            '\U0000296F',
+	"Rfr;":                             '\U0000211C',
+	"Rho;":                             '\U000003A1',
+	"RightAngleBracket;":               '\U000027E9',
+	"RightArrow;":                      '\U00002192',
+	"RightArrowBar;":                   '\U000021E5',
+	"RightArrowLeftArrow;":             '\U000021C4',
+	"RightCeiling;":                    '\U00002309',
+	"RightDoubleBracket;":              '\U000027E7',
+	"RightDownTeeVector;":              '\U0000295D',
+	"RightDownVector;":                 '\U000021C2',
+	"RightDownVectorBar;":              '\U00002955',
+	"RightFloor;":                      '\U0000230B',
+	"RightTee;":                        '\U000022A2',
+	"RightTeeArrow;":                   '\U000021A6',
+	"RightTeeVector;":                  '\U0000295B',
+	"RightTriangle;":                   '\U000022B3',
+	"RightTriangleBar;":                '\U000029D0',
+	"RightTriangleEqual;":              '\U000022B5',
+	"RightUpDownVector;":               '\U0000294F',
+	"RightUpTeeVector;":                '\U0000295C',
+	"RightUpVector;":                   '\U000021BE',
+	"RightUpVectorBar;":                '\U00002954',
+	"RightVector;":                     '\U000021C0',
+	"RightVectorBar;":                  '\U00002953',
+	"Rightarrow;":                      '\U000021D2',
+	"Ropf;":                            '\U0000211D',
+	"RoundImplies;":                    '\U00002970',
+	"Rrightarrow;":                     '\U000021DB',
+	"Rscr;":                            '\U0000211B',
+	"Rsh;":                             '\U000021B1',
+	"RuleDelayed;":                     '\U000029F4',
+	"SHCHcy;":                          '\U00000429',
+	"SHcy;":                            '\U00000428',
+	"SOFTcy;":                          '\U0000042C',
+	"Sacute;":                          '\U0000015A',
+	"Sc;":                              '\U00002ABC',
+	"Scaron;":                          '\U00000160',
+	"Scedil;":                          '\U0000015E',
+	"Scirc;":                           '\U0000015C',
+	"Scy;":                             '\U00000421',
+	"Sfr;":                             '\U0001D516',
+	"ShortDownArrow;":                  '\U00002193',
+	"ShortLeftArrow;":                  '\U00002190',
+	"ShortRightArrow;":                 '\U00002192',
+	"ShortUpArrow;":                    '\U00002191',
+	"Sigma;":                           '\U000003A3',
+	"SmallCircle;":                     '\U00002218',
+	"Sopf;":                            '\U0001D54A',
+	"Sqrt;":                            '\U0000221A',
+	"Square;":                          '\U000025A1',
+	"SquareIntersection;":              '\U00002293',
+	"SquareSubset;":                    '\U0000228F',
+	"SquareSubsetEqual;":               '\U00002291',
+	"SquareSuperset;":                  '\U00002290',
+	"SquareSupersetEqual;":             '\U00002292',
+	"SquareUnion;":                     '\U00002294',
+	"Sscr;":                            '\U0001D4AE',
+	"Star;":                            '\U000022C6',
+	"Sub;":                             '\U000022D0',
+	"Subset;":                          '\U000022D0',
+	"SubsetEqual;":                     '\U00002286',
+	"Succeeds;":                        '\U0000227B',
+	"SucceedsEqual;":                   '\U00002AB0',
+	"SucceedsSlantEqual;":              '\U0000227D',
+	"SucceedsTilde;":                   '\U0000227F',
+	"SuchThat;":                        '\U0000220B',
+	"Sum;":                             '\U00002211',
+	"Sup;":                             '\U000022D1',
+	"Superset;":                        '\U00002283',
+	"SupersetEqual;":                   '\U00002287',
+	"Supset;":                          '\U000022D1',
+	"THORN;":                           '\U000000DE',
+	"TRADE;":                           '\U00002122',
+	"TSHcy;":                           '\U0000040B',
+	"TScy;":                            '\U00000426',
+	"Tab;":                             '\U00000009',
+	"Tau;":                             '\U000003A4',
+	"Tcaron;":                          '\U00000164',
+	"Tcedil;":                          '\U00000162',
+	"Tcy;":                             '\U00000422',
+	"Tfr;":                             '\U0001D517',
+	"Therefore;":                       '\U00002234',
+	"Theta;":                           '\U00000398',
+	"ThinSpace;":                       '\U00002009',
+	"Tilde;":                           '\U0000223C',
+	"TildeEqual;":                      '\U00002243',
+	"TildeFullEqual;":                  '\U00002245',
+	"TildeTilde;":                      '\U00002248',
+	"Topf;":                            '\U0001D54B',
+	"TripleDot;":                       '\U000020DB',
+	"Tscr;":                            '\U0001D4AF',
+	"Tstrok;":                          '\U00000166',
+	"Uacute;":                          '\U000000DA',
+	"Uarr;":                            '\U0000219F',
+	"Uarrocir;":                        '\U00002949',
+	"Ubrcy;":                           '\U0000040E',
+	"Ubreve;":                          '\U0000016C',
+	"Ucirc;":                           '\U000000DB',
+	"Ucy;":                             '\U00000423',
+	"Udblac;":                          '\U00000170',
+	"Ufr;":                             '\U0001D518',
+	"Ugrave;":                          '\U000000D9',
+	"Umacr;":                           '\U0000016A',
+	"UnderBar;":                        '\U0000005F',
+	"UnderBrace;":                      '\U000023DF',
+	"UnderBracket;":                    '\U000023B5',
+	"UnderParenthesis;":                '\U000023DD',
+	"Union;":                           '\U000022C3',
+	"UnionPlus;":                       '\U0000228E',
+	"Uogon;":                           '\U00000172',
+	"Uopf;":                            '\U0001D54C',
+	"UpArrow;":                         '\U00002191',
+	"UpArrowBar;":                      '\U00002912',
+	"UpArrowDownArrow;":                '\U000021C5',
+	"UpDownArrow;":                     '\U00002195',
+	"UpEquilibrium;":                   '\U0000296E',
+	"UpTee;":                           '\U000022A5',
+	"UpTeeArrow;":                      '\U000021A5',
+	"Uparrow;":                         '\U000021D1',
+	"Updownarrow;":                     '\U000021D5',
+	"UpperLeftArrow;":                  '\U00002196',
+	"UpperRightArrow;":                 '\U00002197',
+	"Upsi;":                            '\U000003D2',
+	"Upsilon;":                         '\U000003A5',
+	"Uring;":                           '\U0000016E',
+	"Uscr;":                            '\U0001D4B0',
+	"Utilde;":                          '\U00000168',
+	"Uuml;":                            '\U000000DC',
+	"VDash;":                           '\U000022AB',
+	"Vbar;":                            '\U00002AEB',
+	"Vcy;":                             '\U00000412',
+	"Vdash;":                           '\U000022A9',
+	"Vdashl;":                          '\U00002AE6',
+	"Vee;":                             '\U000022C1',
+	"Verbar;":                          '\U00002016',
+	"Vert;":                            '\U00002016',
+	"VerticalBar;":                     '\U00002223',
+	"VerticalLine;":                    '\U0000007C',
+	"VerticalSeparator;":               '\U00002758',
+	"VerticalTilde;":                   '\U00002240',
+	"VeryThinSpace;":                   '\U0000200A',
+	"Vfr;":                             '\U0001D519',
+	"Vopf;":                            '\U0001D54D',
+	"Vscr;":                            '\U0001D4B1',
+	"Vvdash;":                          '\U000022AA',
+	"Wcirc;":                           '\U00000174',
+	"Wedge;":                           '\U000022C0',
+	"Wfr;":                             '\U0001D51A',
+	"Wopf;":                            '\U0001D54E',
+	"Wscr;":                            '\U0001D4B2',
+	"Xfr;":                             '\U0001D51B',
+	"Xi;":                              '\U0000039E',
+	"Xopf;":                            '\U0001D54F',
+	"Xscr;":                            '\U0001D4B3',
+	"YAcy;":                            '\U0000042F',
+	"YIcy;":                            '\U00000407',
+	"YUcy;":                            '\U0000042E',
+	"Yacute;":                          '\U000000DD',
+	"Ycirc;":                           '\U00000176',
+	"Ycy;":                             '\U0000042B',
+	"Yfr;":                             '\U0001D51C',
+	"Yopf;":                            '\U0001D550',
+	"Yscr;":                            '\U0001D4B4',
+	"Yuml;":                            '\U00000178',
+	"ZHcy;":                            '\U00000416',
+	"Zacute;":                          '\U00000179',
+	"Zcaron;":                          '\U0000017D',
+	"Zcy;":                             '\U00000417',
+	"Zdot;":                            '\U0000017B',
+	"ZeroWidthSpace;":                  '\U0000200B',
+	"Zeta;":                            '\U00000396',
+	"Zfr;":                             '\U00002128',
+	"Zopf;":                            '\U00002124',
+	"Zscr;":                            '\U0001D4B5',
+	"aacute;":                          '\U000000E1',
+	"abreve;":                          '\U00000103',
+	"ac;":                              '\U0000223E',
+	"acd;":                             '\U0000223F',
+	"acirc;":                           '\U000000E2',
+	"acute;":                           '\U000000B4',
+	"acy;":                             '\U00000430',
+	"aelig;":                           '\U000000E6',
+	"af;":                              '\U00002061',
+	"afr;":                             '\U0001D51E',
+	"agrave;":                          '\U000000E0',
+	"alefsym;":                         '\U00002135',
+	"aleph;":                           '\U00002135',
+	"alpha;":                           '\U000003B1',
+	"amacr;":                           '\U00000101',
+	"amalg;":                           '\U00002A3F',
+	"amp;":                             '\U00000026',
+	"and;":                             '\U00002227',
+	"andand;":                          '\U00002A55',
+	"andd;":                            '\U00002A5C',
+	"andslope;":                        '\U00002A58',
+	"andv;":                            '\U00002A5A',
+	"ang;":                             '\U00002220',
+	"ange;":                            '\U000029A4',
+	"angle;":                           '\U00002220',
+	"angmsd;":                          '\U00002221',
+	"angmsdaa;":                        '\U000029A8',
+	"angmsdab;":                        '\U000029A9',
+	"angmsdac;":                        '\U000029AA',
+	"angmsdad;":                        '\U000029AB',
+	"angmsdae;":                        '\U000029AC',
+	"angmsdaf;":                        '\U000029AD',
+	"angmsdag;":                        '\U000029AE',
+	"angmsdah;":                        '\U000029AF',
+	"angrt;":                           '\U0000221F',
+	"angrtvb;":                         '\U000022BE',
+	"angrtvbd;":                        '\U0000299D',
+	"angsph;":                          '\U00002222',
+	"angst;":                           '\U000000C5',
+	"angzarr;":                         '\U0000237C',
+	"aogon;":                           '\U00000105',
+	"aopf;":                            '\U0001D552',
+	"ap;":                              '\U00002248',
+	"apE;":                             '\U00002A70',
+	"apacir;":                          '\U00002A6F',
+	"ape;":                             '\U0000224A',
+	"apid;":                            '\U0000224B',
+	"apos;":                            '\U00000027',
+	"approx;":                          '\U00002248',
+	"approxeq;":                        '\U0000224A',
+	"aring;":                           '\U000000E5',
+	"ascr;":                            '\U0001D4B6',
+	"ast;":                             '\U0000002A',
+	"asymp;":                           '\U00002248',
+	"asympeq;":                         '\U0000224D',
+	"atilde;":                          '\U000000E3',
+	"auml;":                            '\U000000E4',
+	"awconint;":                        '\U00002233',
+	"awint;":                           '\U00002A11',
+	"bNot;":                            '\U00002AED',
+	"backcong;":                        '\U0000224C',
+	"backepsilon;":                     '\U000003F6',
+	"backprime;":                       '\U00002035',
+	"backsim;":                         '\U0000223D',
+	"backsimeq;":                       '\U000022CD',
+	"barvee;":                          '\U000022BD',
+	"barwed;":                          '\U00002305',
+	"barwedge;":                        '\U00002305',
+	"bbrk;":                            '\U000023B5',
+	"bbrktbrk;":                        '\U000023B6',
+	"bcong;":                           '\U0000224C',
+	"bcy;":                             '\U00000431',
+	"bdquo;":                           '\U0000201E',
+	"becaus;":                          '\U00002235',
+	"because;":                         '\U00002235',
+	"bemptyv;":                         '\U000029B0',
+	"bepsi;":                           '\U000003F6',
+	"bernou;":                          '\U0000212C',
+	"beta;":                            '\U000003B2',
+	"beth;":                            '\U00002136',
+	"between;":                         '\U0000226C',
+	"bfr;":                             '\U0001D51F',
+	"bigcap;":                          '\U000022C2',
+	"bigcirc;":                         '\U000025EF',
+	"bigcup;":                          '\U000022C3',
+	"bigodot;":                         '\U00002A00',
+	"bigoplus;":                        '\U00002A01',
+	"bigotimes;":                       '\U00002A02',
+	"bigsqcup;":                        '\U00002A06',
+	"bigstar;":                         '\U00002605',
+	"bigtriangledown;":                 '\U000025BD',
+	"bigtriangleup;":                   '\U000025B3',
+	"biguplus;":                        '\U00002A04',
+	"bigvee;":                          '\U000022C1',
+	"bigwedge;":                        '\U000022C0',
+	"bkarow;":                          '\U0000290D',
+	"blacklozenge;":                    '\U000029EB',
+	"blacksquare;":                     '\U000025AA',
+	"blacktriangle;":                   '\U000025B4',
+	"blacktriangledown;":               '\U000025BE',
+	"blacktriangleleft;":               '\U000025C2',
+	"blacktriangleright;":              '\U000025B8',
+	"blank;":                           '\U00002423',
+	"blk12;":                           '\U00002592',
+	"blk14;":                           '\U00002591',
+	"blk34;":                           '\U00002593',
+	"block;":                           '\U00002588',
+	"bnot;":                            '\U00002310',
+	"bopf;":                            '\U0001D553',
+	"bot;":                             '\U000022A5',
+	"bottom;":                          '\U000022A5',
+	"bowtie;":                          '\U000022C8',
+	"boxDL;":                           '\U00002557',
+	"boxDR;":                           '\U00002554',
+	"boxDl;":                           '\U00002556',
+	"boxDr;":                           '\U00002553',
+	"boxH;":                            '\U00002550',
+	"boxHD;":                           '\U00002566',
+	"boxHU;":                           '\U00002569',
+	"boxHd;":                           '\U00002564',
+	"boxHu;":                           '\U00002567',
+	"boxUL;":                           '\U0000255D',
+	"boxUR;":                           '\U0000255A',
+	"boxUl;":                           '\U0000255C',
+	"boxUr;":                           '\U00002559',
+	"boxV;":                            '\U00002551',
+	"boxVH;":                           '\U0000256C',
+	"boxVL;":                           '\U00002563',
+	"boxVR;":                           '\U00002560',
+	"boxVh;":                           '\U0000256B',
+	"boxVl;":                           '\U00002562',
+	"boxVr;":                           '\U0000255F',
+	"boxbox;":                          '\U000029C9',
+	"boxdL;":                           '\U00002555',
+	"boxdR;":                           '\U00002552',
+	"boxdl;":                           '\U00002510',
+	"boxdr;":                           '\U0000250C',
+	"boxh;":                            '\U00002500',
+	"boxhD;":                           '\U00002565',
+	"boxhU;":                           '\U00002568',
+	"boxhd;":                           '\U0000252C',
+	"boxhu;":                           '\U00002534',
+	"boxminus;":                        '\U0000229F',
+	"boxplus;":                         '\U0000229E',
+	"boxtimes;":                        '\U000022A0',
+	"boxuL;":                           '\U0000255B',
+	"boxuR;":                           '\U00002558',
+	"boxul;":                           '\U00002518',
+	"boxur;":                           '\U00002514',
+	"boxv;":                            '\U00002502',
+	"boxvH;":                           '\U0000256A',
+	"boxvL;":                           '\U00002561',
+	"boxvR;":                           '\U0000255E',
+	"boxvh;":                           '\U0000253C',
+	"boxvl;":                           '\U00002524',
+	"boxvr;":                           '\U0000251C',
+	"bprime;":                          '\U00002035',
+	"breve;":                           '\U000002D8',
+	"brvbar;":                          '\U000000A6',
+	"bscr;":                            '\U0001D4B7',
+	"bsemi;":                           '\U0000204F',
+	"bsim;":                            '\U0000223D',
+	"bsime;":                           '\U000022CD',
+	"bsol;":                            '\U0000005C',
+	"bsolb;":                           '\U000029C5',
+	"bsolhsub;":                        '\U000027C8',
+	"bull;":                            '\U00002022',
+	"bullet;":                          '\U00002022',
+	"bump;":                            '\U0000224E',
+	"bumpE;":                           '\U00002AAE',
+	"bumpe;":                           '\U0000224F',
+	"bumpeq;":                          '\U0000224F',
+	"cacute;":                          '\U00000107',
+	"cap;":                             '\U00002229',
+	"capand;":                          '\U00002A44',
+	"capbrcup;":                        '\U00002A49',
+	"capcap;":                          '\U00002A4B',
+	"capcup;":                          '\U00002A47',
+	"capdot;":                          '\U00002A40',
+	"caret;":                           '\U00002041',
+	"caron;":                           '\U000002C7',
+	"ccaps;":                           '\U00002A4D',
+	"ccaron;":                          '\U0000010D',
+	"ccedil;":                          '\U000000E7',
+	"ccirc;":                           '\U00000109',
+	"ccups;":                           '\U00002A4C',
+	"ccupssm;":                         '\U00002A50',
+	"cdot;":                            '\U0000010B',
+	"cedil;":                           '\U000000B8',
+	"cemptyv;":                         '\U000029B2',
+	"cent;":                            '\U000000A2',
+	"centerdot;":                       '\U000000B7',
+	"cfr;":                             '\U0001D520',
+	"chcy;":                            '\U00000447',
+	"check;":                           '\U00002713',
+	"checkmark;":                       '\U00002713',
+	"chi;":                             '\U000003C7',
+	"cir;":                             '\U000025CB',
+	"cirE;":                            '\U000029C3',
+	"circ;":                            '\U000002C6',
+	"circeq;":                          '\U00002257',
+	"circlearrowleft;":                 '\U000021BA',
+	"circlearrowright;":                '\U000021BB',
+	"circledR;":                        '\U000000AE',
+	"circledS;":                        '\U000024C8',
+	"circledast;":                      '\U0000229B',
+	"circledcirc;":                     '\U0000229A',
+	"circleddash;":                     '\U0000229D',
+	"cire;":                            '\U00002257',
+	"cirfnint;":                        '\U00002A10',
+	"cirmid;":                          '\U00002AEF',
+	"cirscir;":                         '\U000029C2',
+	"clubs;":                           '\U00002663',
+	"clubsuit;":                        '\U00002663',
+	"colon;":                           '\U0000003A',
+	"colone;":                          '\U00002254',
+	"coloneq;":                         '\U00002254',
+	"comma;":                           '\U0000002C',
+	"commat;":                          '\U00000040',
+	"comp;":                            '\U00002201',
+	"compfn;":                          '\U00002218',
+	"complement;":                      '\U00002201',
+	"complexes;":                       '\U00002102',
+	"cong;":                            '\U00002245',
+	"congdot;":                         '\U00002A6D',
+	"conint;":                          '\U0000222E',
+	"copf;":                            '\U0001D554',
+	"coprod;":                          '\U00002210',
+	"copy;":                            '\U000000A9',
+	"copysr;":                          '\U00002117',
+	"crarr;":                           '\U000021B5',
+	"cross;":                           '\U00002717',
+	"cscr;":                            '\U0001D4B8',
+	"csub;":                            '\U00002ACF',
+	"csube;":                           '\U00002AD1',
+	"csup;":                            '\U00002AD0',
+	"csupe;":                           '\U00002AD2',
+	"ctdot;":                           '\U000022EF',
+	"cudarrl;":                         '\U00002938',
+	"cudarrr;":                         '\U00002935',
+	"cuepr;":                           '\U000022DE',
+	"cuesc;":                           '\U000022DF',
+	"cularr;":                          '\U000021B6',
+	"cularrp;":                         '\U0000293D',
+	"cup;":                             '\U0000222A',
+	"cupbrcap;":                        '\U00002A48',
+	"cupcap;":                          '\U00002A46',
+	"cupcup;":                          '\U00002A4A',
+	"cupdot;":                          '\U0000228D',
+	"cupor;":                           '\U00002A45',
+	"curarr;":                          '\U000021B7',
+	"curarrm;":                         '\U0000293C',
+	"curlyeqprec;":                     '\U000022DE',
+	"curlyeqsucc;":                     '\U000022DF',
+	"curlyvee;":                        '\U000022CE',
+	"curlywedge;":                      '\U000022CF',
+	"curren;":                          '\U000000A4',
+	"curvearrowleft;":                  '\U000021B6',
+	"curvearrowright;":                 '\U000021B7',
+	"cuvee;":                           '\U000022CE',
+	"cuwed;":                           '\U000022CF',
+	"cwconint;":                        '\U00002232',
+	"cwint;":                           '\U00002231',
+	"cylcty;":                          '\U0000232D',
+	"dArr;":                            '\U000021D3',
+	"dHar;":                            '\U00002965',
+	"dagger;":                          '\U00002020',
+	"daleth;":                          '\U00002138',
+	"darr;":                            '\U00002193',
+	"dash;":                            '\U00002010',
+	"dashv;":                           '\U000022A3',
+	"dbkarow;":                         '\U0000290F',
+	"dblac;":                           '\U000002DD',
+	"dcaron;":                          '\U0000010F',
+	"dcy;":                             '\U00000434',
+	"dd;":                              '\U00002146',
+	"ddagger;":                         '\U00002021',
+	"ddarr;":                           '\U000021CA',
+	"ddotseq;":                         '\U00002A77',
+	"deg;":                             '\U000000B0',
+	"delta;":                           '\U000003B4',
+	"demptyv;":                         '\U000029B1',
+	"dfisht;":                          '\U0000297F',
+	"dfr;":                             '\U0001D521',
+	"dharl;":                           '\U000021C3',
+	"dharr;":                           '\U000021C2',
+	"diam;":                            '\U000022C4',
+	"diamond;":                         '\U000022C4',
+	"diamondsuit;":                     '\U00002666',
+	"diams;":                           '\U00002666',
+	"die;":                             '\U000000A8',
+	"digamma;":                         '\U000003DD',
+	"disin;":                           '\U000022F2',
+	"div;":                             '\U000000F7',
+	"divide;":                          '\U000000F7',
+	"divideontimes;":                   '\U000022C7',
+	"divonx;":                          '\U000022C7',
+	"djcy;":                            '\U00000452',
+	"dlcorn;":                          '\U0000231E',
+	"dlcrop;":                          '\U0000230D',
+	"dollar;":                          '\U00000024',
+	"dopf;":                            '\U0001D555',
+	"dot;":                             '\U000002D9',
+	"doteq;":                           '\U00002250',
+	"doteqdot;":                        '\U00002251',
+	"dotminus;":                        '\U00002238',
+	"dotplus;":                         '\U00002214',
+	"dotsquare;":                       '\U000022A1',
+	"doublebarwedge;":                  '\U00002306',
+	"downarrow;":                       '\U00002193',
+	"downdownarrows;":                  '\U000021CA',
+	"downharpoonleft;":                 '\U000021C3',
+	"downharpoonright;":                '\U000021C2',
+	"drbkarow;":                        '\U00002910',
+	"drcorn;":                          '\U0000231F',
+	"drcrop;":                          '\U0000230C',
+	"dscr;":                            '\U0001D4B9',
+	"dscy;":                            '\U00000455',
+	"dsol;":                            '\U000029F6',
+	"dstrok;":                          '\U00000111',
+	"dtdot;":                           '\U000022F1',
+	"dtri;":                            '\U000025BF',
+	"dtrif;":                           '\U000025BE',
+	"duarr;":                           '\U000021F5',
+	"duhar;":                           '\U0000296F',
+	"dwangle;":                         '\U000029A6',
+	"dzcy;":                            '\U0000045F',
+	"dzigrarr;":                        '\U000027FF',
+	"eDDot;":                           '\U00002A77',
+	"eDot;":                            '\U00002251',
+	"eacute;":                          '\U000000E9',
+	"easter;":                          '\U00002A6E',
+	"ecaron;":                          '\U0000011B',
+	"ecir;":                            '\U00002256',
+	"ecirc;":                           '\U000000EA',
+	"ecolon;":                          '\U00002255',
+	"ecy;":                             '\U0000044D',
+	"edot;":                            '\U00000117',
+	"ee;":                              '\U00002147',
+	"efDot;":                           '\U00002252',
+	"efr;":                             '\U0001D522',
+	"eg;":                              '\U00002A9A',
+	"egrave;":                          '\U000000E8',
+	"egs;":                             '\U00002A96',
+	"egsdot;":                          '\U00002A98',
+	"el;":                              '\U00002A99',
+	"elinters;":                        '\U000023E7',
+	"ell;":                             '\U00002113',
+	"els;":                             '\U00002A95',
+	"elsdot;":                          '\U00002A97',
+	"emacr;":                           '\U00000113',
+	"empty;":                           '\U00002205',
+	"emptyset;":                        '\U00002205',
+	"emptyv;":                          '\U00002205',
+	"emsp;":                            '\U00002003',
+	"emsp13;":                          '\U00002004',
+	"emsp14;":                          '\U00002005',
+	"eng;":                             '\U0000014B',
+	"ensp;":                            '\U00002002',
+	"eogon;":                           '\U00000119',
+	"eopf;":                            '\U0001D556',
+	"epar;":                            '\U000022D5',
+	"eparsl;":                          '\U000029E3',
+	"eplus;":                           '\U00002A71',
+	"epsi;":                            '\U000003B5',
+	"epsilon;":                         '\U000003B5',
+	"epsiv;":                           '\U000003F5',
+	"eqcirc;":                          '\U00002256',
+	"eqcolon;":                         '\U00002255',
+	"eqsim;":                           '\U00002242',
+	"eqslantgtr;":                      '\U00002A96',
+	"eqslantless;":                     '\U00002A95',
+	"equals;":                          '\U0000003D',
+	"equest;":                          '\U0000225F',
+	"equiv;":                           '\U00002261',
+	"equivDD;":                         '\U00002A78',
+	"eqvparsl;":                        '\U000029E5',
+	"erDot;":                           '\U00002253',
+	"erarr;":                           '\U00002971',
+	"escr;":                            '\U0000212F',
+	"esdot;":                           '\U00002250',
+	"esim;":                            '\U00002242',
+	"eta;":                             '\U000003B7',
+	"eth;":                             '\U000000F0',
+	"euml;":                            '\U000000EB',
+	"euro;":                            '\U000020AC',
+	"excl;":                            '\U00000021',
+	"exist;":                           '\U00002203',
+	"expectation;":                     '\U00002130',
+	"exponentiale;":                    '\U00002147',
+	"fallingdotseq;":                   '\U00002252',
+	"fcy;":                             '\U00000444',
+	"female;":                          '\U00002640',
+	"ffilig;":                          '\U0000FB03',
+	"fflig;":                           '\U0000FB00',
+	"ffllig;":                          '\U0000FB04',
+	"ffr;":                             '\U0001D523',
+	"filig;":                           '\U0000FB01',
+	"flat;":                            '\U0000266D',
+	"fllig;":                           '\U0000FB02',
+	"fltns;":                           '\U000025B1',
+	"fnof;":                            '\U00000192',
+	"fopf;":                            '\U0001D557',
+	"forall;":                          '\U00002200',
+	"fork;":                            '\U000022D4',
+	"forkv;":                           '\U00002AD9',
+	"fpartint;":                        '\U00002A0D',
+	"frac12;":                          '\U000000BD',
+	"frac13;":                          '\U00002153',
+	"frac14;":                          '\U000000BC',
+	"frac15;":                          '\U00002155',
+	"frac16;":                          '\U00002159',
+	"frac18;":                          '\U0000215B',
+	"frac23;":                          '\U00002154',
+	"frac25;":                          '\U00002156',
+	"frac34;":                          '\U000000BE',
+	"frac35;":                          '\U00002157',
+	"frac38;":                          '\U0000215C',
+	"frac45;":                          '\U00002158',
+	"frac56;":                          '\U0000215A',
+	"frac58;":                          '\U0000215D',
+	"frac78;":                          '\U0000215E',
+	"frasl;":                           '\U00002044',
+	"frown;":                           '\U00002322',
+	"fscr;":                            '\U0001D4BB',
+	"gE;":                              '\U00002267',
+	"gEl;":                             '\U00002A8C',
+	"gacute;":                          '\U000001F5',
+	"gamma;":                           '\U000003B3',
+	"gammad;":                          '\U000003DD',
+	"gap;":                             '\U00002A86',
+	"gbreve;":                          '\U0000011F',
+	"gcirc;":                           '\U0000011D',
+	"gcy;":                             '\U00000433',
+	"gdot;":                            '\U00000121',
+	"ge;":                              '\U00002265',
+	"gel;":                             '\U000022DB',
+	"geq;":                             '\U00002265',
+	"geqq;":                            '\U00002267',
+	"geqslant;":                        '\U00002A7E',
+	"ges;":                             '\U00002A7E',
+	"gescc;":                           '\U00002AA9',
+	"gesdot;":                          '\U00002A80',
+	"gesdoto;":                         '\U00002A82',
+	"gesdotol;":                        '\U00002A84',
+	"gesles;":                          '\U00002A94',
+	"gfr;":                             '\U0001D524',
+	"gg;":                              '\U0000226B',
+	"ggg;":                             '\U000022D9',
+	"gimel;":                           '\U00002137',
+	"gjcy;":                            '\U00000453',
+	"gl;":                              '\U00002277',
+	"glE;":                             '\U00002A92',
+	"gla;":                             '\U00002AA5',
+	"glj;":                             '\U00002AA4',
+	"gnE;":                             '\U00002269',
+	"gnap;":                            '\U00002A8A',
+	"gnapprox;":                        '\U00002A8A',
+	"gne;":                             '\U00002A88',
+	"gneq;":                            '\U00002A88',
+	"gneqq;":                           '\U00002269',
+	"gnsim;":                           '\U000022E7',
+	"gopf;":                            '\U0001D558',
+	"grave;":                           '\U00000060',
+	"gscr;":                            '\U0000210A',
+	"gsim;":                            '\U00002273',
+	"gsime;":                           '\U00002A8E',
+	"gsiml;":                           '\U00002A90',
+	"gt;":                              '\U0000003E',
+	"gtcc;":                            '\U00002AA7',
+	"gtcir;":                           '\U00002A7A',
+	"gtdot;":                           '\U000022D7',
+	"gtlPar;":                          '\U00002995',
+	"gtquest;":                         '\U00002A7C',
+	"gtrapprox;":                       '\U00002A86',
+	"gtrarr;":                          '\U00002978',
+	"gtrdot;":                          '\U000022D7',
+	"gtreqless;":                       '\U000022DB',
+	"gtreqqless;":                      '\U00002A8C',
+	"gtrless;":                         '\U00002277',
+	"gtrsim;":                          '\U00002273',
+	"hArr;":                            '\U000021D4',
+	"hairsp;":                          '\U0000200A',
+	"half;":                            '\U000000BD',
+	"hamilt;":                          '\U0000210B',
+	"hardcy;":                          '\U0000044A',
+	"harr;":                            '\U00002194',
+	"harrcir;":                         '\U00002948',
+	"harrw;":                           '\U000021AD',
+	"hbar;":                            '\U0000210F',
+	"hcirc;":                           '\U00000125',
+	"hearts;":                          '\U00002665',
+	"heartsuit;":                       '\U00002665',
+	"hellip;":                          '\U00002026',
+	"hercon;":                          '\U000022B9',
+	"hfr;":                             '\U0001D525',
+	"hksearow;":                        '\U00002925',
+	"hkswarow;":                        '\U00002926',
+	"hoarr;":                           '\U000021FF',
+	"homtht;":                          '\U0000223B',
+	"hookleftarrow;":                   '\U000021A9',
+	"hookrightarrow;":                  '\U000021AA',
+	"hopf;":                            '\U0001D559',
+	"horbar;":                          '\U00002015',
+	"hscr;":                            '\U0001D4BD',
+	"hslash;":                          '\U0000210F',
+	"hstrok;":                          '\U00000127',
+	"hybull;":                          '\U00002043',
+	"hyphen;":                          '\U00002010',
+	"iacute;":                          '\U000000ED',
+	"ic;":                              '\U00002063',
+	"icirc;":                           '\U000000EE',
+	"icy;":                             '\U00000438',
+	"iecy;":                            '\U00000435',
+	"iexcl;":                           '\U000000A1',
+	"iff;":                             '\U000021D4',
+	"ifr;":                             '\U0001D526',
+	"igrave;":                          '\U000000EC',
+	"ii;":                              '\U00002148',
+	"iiiint;":                          '\U00002A0C',
+	"iiint;":                           '\U0000222D',
+	"iinfin;":                          '\U000029DC',
+	"iiota;":                           '\U00002129',
+	"ijlig;":                           '\U00000133',
+	"imacr;":                           '\U0000012B',
+	"image;":                           '\U00002111',
+	"imagline;":                        '\U00002110',
+	"imagpart;":                        '\U00002111',
+	"imath;":                           '\U00000131',
+	"imof;":                            '\U000022B7',
+	"imped;":                           '\U000001B5',
+	"in;":                              '\U00002208',
+	"incare;":                          '\U00002105',
+	"infin;":                           '\U0000221E',
+	"infintie;":                        '\U000029DD',
+	"inodot;":                          '\U00000131',
+	"int;":                             '\U0000222B',
+	"intcal;":                          '\U000022BA',
+	"integers;":                        '\U00002124',
+	"intercal;":                        '\U000022BA',
+	"intlarhk;":                        '\U00002A17',
+	"intprod;":                         '\U00002A3C',
+	"iocy;":                            '\U00000451',
+	"iogon;":                           '\U0000012F',
+	"iopf;":                            '\U0001D55A',
+	"iota;":                            '\U000003B9',
+	"iprod;":                           '\U00002A3C',
+	"iquest;":                          '\U000000BF',
+	"iscr;":                            '\U0001D4BE',
+	"isin;":                            '\U00002208',
+	"isinE;":                           '\U000022F9',
+	"isindot;":                         '\U000022F5',
+	"isins;":                           '\U000022F4',
+	"isinsv;":                          '\U000022F3',
+	"isinv;":                           '\U00002208',
+	"it;":                              '\U00002062',
+	"itilde;":                          '\U00000129',
+	"iukcy;":                           '\U00000456',
+	"iuml;":                            '\U000000EF',
+	"jcirc;":                           '\U00000135',
+	"jcy;":                             '\U00000439',
+	"jfr;":                             '\U0001D527',
+	"jmath;":                           '\U00000237',
+	"jopf;":                            '\U0001D55B',
+	"jscr;":                            '\U0001D4BF',
+	"jsercy;":                          '\U00000458',
+	"jukcy;":                           '\U00000454',
+	"kappa;":                           '\U000003BA',
+	"kappav;":                          '\U000003F0',
+	"kcedil;":                          '\U00000137',
+	"kcy;":                             '\U0000043A',
+	"kfr;":                             '\U0001D528',
+	"kgreen;":                          '\U00000138',
+	"khcy;":                            '\U00000445',
+	"kjcy;":                            '\U0000045C',
+	"kopf;":                            '\U0001D55C',
+	"kscr;":                            '\U0001D4C0',
+	"lAarr;":                           '\U000021DA',
+	"lArr;":                            '\U000021D0',
+	"lAtail;":                          '\U0000291B',
+	"lBarr;":                           '\U0000290E',
+	"lE;":                              '\U00002266',
+	"lEg;":                             '\U00002A8B',
+	"lHar;":                            '\U00002962',
+	"lacute;":                          '\U0000013A',
+	"laemptyv;":                        '\U000029B4',
+	"lagran;":                          '\U00002112',
+	"lambda;":                          '\U000003BB',
+	"lang;":                            '\U000027E8',
+	"langd;":                           '\U00002991',
+	"langle;":                          '\U000027E8',
+	"lap;":                             '\U00002A85',
+	"laquo;":                           '\U000000AB',
+	"larr;":                            '\U00002190',
+	"larrb;":                           '\U000021E4',
+	"larrbfs;":                         '\U0000291F',
+	"larrfs;":                          '\U0000291D',
+	"larrhk;":                          '\U000021A9',
+	"larrlp;":                          '\U000021AB',
+	"larrpl;":                          '\U00002939',
+	"larrsim;":                         '\U00002973',
+	"larrtl;":                          '\U000021A2',
+	"lat;":                             '\U00002AAB',
+	"latail;":                          '\U00002919',
+	"late;":                            '\U00002AAD',
+	"lbarr;":                           '\U0000290C',
+	"lbbrk;":                           '\U00002772',
+	"lbrace;":                          '\U0000007B',
+	"lbrack;":                          '\U0000005B',
+	"lbrke;":                           '\U0000298B',
+	"lbrksld;":                         '\U0000298F',
+	"lbrkslu;":                         '\U0000298D',
+	"lcaron;":                          '\U0000013E',
+	"lcedil;":                          '\U0000013C',
+	"lceil;":                           '\U00002308',
+	"lcub;":                            '\U0000007B',
+	"lcy;":                             '\U0000043B',
+	"ldca;":                            '\U00002936',
+	"ldquo;":                           '\U0000201C',
+	"ldquor;":                          '\U0000201E',
+	"ldrdhar;":                         '\U00002967',
+	"ldrushar;":                        '\U0000294B',
+	"ldsh;":                            '\U000021B2',
+	"le;":                              '\U00002264',
+	"leftarrow;":                       '\U00002190',
+	"leftarrowtail;":                   '\U000021A2',
+	"leftharpoondown;":                 '\U000021BD',
+	"leftharpoonup;":                   '\U000021BC',
+	"leftleftarrows;":                  '\U000021C7',
+	"leftrightarrow;":                  '\U00002194',
+	"leftrightarrows;":                 '\U000021C6',
+	"leftrightharpoons;":               '\U000021CB',
+	"leftrightsquigarrow;":             '\U000021AD',
+	"leftthreetimes;":                  '\U000022CB',
+	"leg;":                             '\U000022DA',
+	"leq;":                             '\U00002264',
+	"leqq;":                            '\U00002266',
+	"leqslant;":                        '\U00002A7D',
+	"les;":                             '\U00002A7D',
+	"lescc;":                           '\U00002AA8',
+	"lesdot;":                          '\U00002A7F',
+	"lesdoto;":                         '\U00002A81',
+	"lesdotor;":                        '\U00002A83',
+	"lesges;":                          '\U00002A93',
+	"lessapprox;":                      '\U00002A85',
+	"lessdot;":                         '\U000022D6',
+	"lesseqgtr;":                       '\U000022DA',
+	"lesseqqgtr;":                      '\U00002A8B',
+	"lessgtr;":                         '\U00002276',
+	"lesssim;":                         '\U00002272',
+	"lfisht;":                          '\U0000297C',
+	"lfloor;":                          '\U0000230A',
+	"lfr;":                             '\U0001D529',
+	"lg;":                              '\U00002276',
+	"lgE;":                             '\U00002A91',
+	"lhard;":                           '\U000021BD',
+	"lharu;":                           '\U000021BC',
+	"lharul;":                          '\U0000296A',
+	"lhblk;":                           '\U00002584',
+	"ljcy;":                            '\U00000459',
+	"ll;":                              '\U0000226A',
+	"llarr;":                           '\U000021C7',
+	"llcorner;":                        '\U0000231E',
+	"llhard;":                          '\U0000296B',
+	"lltri;":                           '\U000025FA',
+	"lmidot;":                          '\U00000140',
+	"lmoust;":                          '\U000023B0',
+	"lmoustache;":                      '\U000023B0',
+	"lnE;":                             '\U00002268',
+	"lnap;":                            '\U00002A89',
+	"lnapprox;":                        '\U00002A89',
+	"lne;":                             '\U00002A87',
+	"lneq;":                            '\U00002A87',
+	"lneqq;":                           '\U00002268',
+	"lnsim;":                           '\U000022E6',
+	"loang;":                           '\U000027EC',
+	"loarr;":                           '\U000021FD',
+	"lobrk;":                           '\U000027E6',
+	"longleftarrow;":                   '\U000027F5',
+	"longleftrightarrow;":              '\U000027F7',
+	"longmapsto;":                      '\U000027FC',
+	"longrightarrow;":                  '\U000027F6',
+	"looparrowleft;":                   '\U000021AB',
+	"looparrowright;":                  '\U000021AC',
+	"lopar;":                           '\U00002985',
+	"lopf;":                            '\U0001D55D',
+	"loplus;":                          '\U00002A2D',
+	"lotimes;":                         '\U00002A34',
+	"lowast;":                          '\U00002217',
+	"lowbar;":                          '\U0000005F',
+	"loz;":                             '\U000025CA',
+	"lozenge;":                         '\U000025CA',
+	"lozf;":                            '\U000029EB',
+	"lpar;":                            '\U00000028',
+	"lparlt;":                          '\U00002993',
+	"lrarr;":                           '\U000021C6',
+	"lrcorner;":                        '\U0000231F',
+	"lrhar;":                           '\U000021CB',
+	"lrhard;":                          '\U0000296D',
+	"lrm;":                             '\U0000200E',
+	"lrtri;":                           '\U000022BF',
+	"lsaquo;":                          '\U00002039',
+	"lscr;":                            '\U0001D4C1',
+	"lsh;":                             '\U000021B0',
+	"lsim;":                            '\U00002272',
+	"lsime;":                           '\U00002A8D',
+	"lsimg;":                           '\U00002A8F',
+	"lsqb;":                            '\U0000005B',
+	"lsquo;":                           '\U00002018',
+	"lsquor;":                          '\U0000201A',
+	"lstrok;":                          '\U00000142',
+	"lt;":                              '\U0000003C',
+	"ltcc;":                            '\U00002AA6',
+	"ltcir;":                           '\U00002A79',
+	"ltdot;":                           '\U000022D6',
+	"lthree;":                          '\U000022CB',
+	"ltimes;":                          '\U000022C9',
+	"ltlarr;":                          '\U00002976',
+	"ltquest;":                         '\U00002A7B',
+	"ltrPar;":                          '\U00002996',
+	"ltri;":                            '\U000025C3',
+	"ltrie;":                           '\U000022B4',
+	"ltrif;":                           '\U000025C2',
+	"lurdshar;":                        '\U0000294A',
+	"luruhar;":                         '\U00002966',
+	"mDDot;":                           '\U0000223A',
+	"macr;":                            '\U000000AF',
+	"male;":                            '\U00002642',
+	"malt;":                            '\U00002720',
+	"maltese;":                         '\U00002720',
+	"map;":                             '\U000021A6',
+	"mapsto;":                          '\U000021A6',
+	"mapstodown;":                      '\U000021A7',
+	"mapstoleft;":                      '\U000021A4',
+	"mapstoup;":                        '\U000021A5',
+	"marker;":                          '\U000025AE',
+	"mcomma;":                          '\U00002A29',
+	"mcy;":                             '\U0000043C',
+	"mdash;":                           '\U00002014',
+	"measuredangle;":                   '\U00002221',
+	"mfr;":                             '\U0001D52A',
+	"mho;":                             '\U00002127',
+	"micro;":                           '\U000000B5',
+	"mid;":                             '\U00002223',
+	"midast;":                          '\U0000002A',
+	"midcir;":                          '\U00002AF0',
+	"middot;":                          '\U000000B7',
+	"minus;":                           '\U00002212',
+	"minusb;":                          '\U0000229F',
+	"minusd;":                          '\U00002238',
+	"minusdu;":                         '\U00002A2A',
+	"mlcp;":                            '\U00002ADB',
+	"mldr;":                            '\U00002026',
+	"mnplus;":                          '\U00002213',
+	"models;":                          '\U000022A7',
+	"mopf;":                            '\U0001D55E',
+	"mp;":                              '\U00002213',
+	"mscr;":                            '\U0001D4C2',
+	"mstpos;":                          '\U0000223E',
+	"mu;":                              '\U000003BC',
+	"multimap;":                        '\U000022B8',
+	"mumap;":                           '\U000022B8',
+	"nLeftarrow;":                      '\U000021CD',
+	"nLeftrightarrow;":                 '\U000021CE',
+	"nRightarrow;":                     '\U000021CF',
+	"nVDash;":                          '\U000022AF',
+	"nVdash;":                          '\U000022AE',
+	"nabla;":                           '\U00002207',
+	"nacute;":                          '\U00000144',
+	"nap;":                             '\U00002249',
+	"napos;":                           '\U00000149',
+	"napprox;":                         '\U00002249',
+	"natur;":                           '\U0000266E',
+	"natural;":                         '\U0000266E',
+	"naturals;":                        '\U00002115',
+	"nbsp;":                            '\U000000A0',
+	"ncap;":                            '\U00002A43',
+	"ncaron;":                          '\U00000148',
+	"ncedil;":                          '\U00000146',
+	"ncong;":                           '\U00002247',
+	"ncup;":                            '\U00002A42',
+	"ncy;":                             '\U0000043D',
+	"ndash;":                           '\U00002013',
+	"ne;":                              '\U00002260',
+	"neArr;":                           '\U000021D7',
+	"nearhk;":                          '\U00002924',
+	"nearr;":                           '\U00002197',
+	"nearrow;":                         '\U00002197',
+	"nequiv;":                          '\U00002262',
+	"nesear;":                          '\U00002928',
+	"nexist;":                          '\U00002204',
+	"nexists;":                         '\U00002204',
+	"nfr;":                             '\U0001D52B',
+	"nge;":                             '\U00002271',
+	"ngeq;":                            '\U00002271',
+	"ngsim;":                           '\U00002275',
+	"ngt;":                             '\U0000226F',
+	"ngtr;":                            '\U0000226F',
+	"nhArr;":                           '\U000021CE',
+	"nharr;":                           '\U000021AE',
+	"nhpar;":                           '\U00002AF2',
+	"ni;":                              '\U0000220B',
+	"nis;":                             '\U000022FC',
+	"nisd;":                            '\U000022FA',
+	"niv;":                             '\U0000220B',
+	"njcy;":                            '\U0000045A',
+	"nlArr;":                           '\U000021CD',
+	"nlarr;":                           '\U0000219A',
+	"nldr;":                            '\U00002025',
+	"nle;":                             '\U00002270',
+	"nleftarrow;":                      '\U0000219A',
+	"nleftrightarrow;":                 '\U000021AE',
+	"nleq;":                            '\U00002270',
+	"nless;":                           '\U0000226E',
+	"nlsim;":                           '\U00002274',
+	"nlt;":                             '\U0000226E',
+	"nltri;":                           '\U000022EA',
+	"nltrie;":                          '\U000022EC',
+	"nmid;":                            '\U00002224',
+	"nopf;":                            '\U0001D55F',
+	"not;":                             '\U000000AC',
+	"notin;":                           '\U00002209',
+	"notinva;":                         '\U00002209',
+	"notinvb;":                         '\U000022F7',
+	"notinvc;":                         '\U000022F6',
+	"notni;":                           '\U0000220C',
+	"notniva;":                         '\U0000220C',
+	"notnivb;":                         '\U000022FE',
+	"notnivc;":                         '\U000022FD',
+	"npar;":                            '\U00002226',
+	"nparallel;":                       '\U00002226',
+	"npolint;":                         '\U00002A14',
+	"npr;":                             '\U00002280',
+	"nprcue;":                          '\U000022E0',
+	"nprec;":                           '\U00002280',
+	"nrArr;":                           '\U000021CF',
+	"nrarr;":                           '\U0000219B',
+	"nrightarrow;":                     '\U0000219B',
+	"nrtri;":                           '\U000022EB',
+	"nrtrie;":                          '\U000022ED',
+	"nsc;":                             '\U00002281',
+	"nsccue;":                          '\U000022E1',
+	"nscr;":                            '\U0001D4C3',
+	"nshortmid;":                       '\U00002224',
+	"nshortparallel;":                  '\U00002226',
+	"nsim;":                            '\U00002241',
+	"nsime;":                           '\U00002244',
+	"nsimeq;":                          '\U00002244',
+	"nsmid;":                           '\U00002224',
+	"nspar;":                           '\U00002226',
+	"nsqsube;":                         '\U000022E2',
+	"nsqsupe;":                         '\U000022E3',
+	"nsub;":                            '\U00002284',
+	"nsube;":                           '\U00002288',
+	"nsubseteq;":                       '\U00002288',
+	"nsucc;":                           '\U00002281',
+	"nsup;":                            '\U00002285',
+	"nsupe;":                           '\U00002289',
+	"nsupseteq;":                       '\U00002289',
+	"ntgl;":                            '\U00002279',
+	"ntilde;":                          '\U000000F1',
+	"ntlg;":                            '\U00002278',
+	"ntriangleleft;":                   '\U000022EA',
+	"ntrianglelefteq;":                 '\U000022EC',
+	"ntriangleright;":                  '\U000022EB',
+	"ntrianglerighteq;":                '\U000022ED',
+	"nu;":                              '\U000003BD',
+	"num;":                             '\U00000023',
+	"numero;":                          '\U00002116',
+	"numsp;":                           '\U00002007',
+	"nvDash;":                          '\U000022AD',
+	"nvHarr;":                          '\U00002904',
+	"nvdash;":                          '\U000022AC',
+	"nvinfin;":                         '\U000029DE',
+	"nvlArr;":                          '\U00002902',
+	"nvrArr;":                          '\U00002903',
+	"nwArr;":                           '\U000021D6',
+	"nwarhk;":                          '\U00002923',
+	"nwarr;":                           '\U00002196',
+	"nwarrow;":                         '\U00002196',
+	"nwnear;":                          '\U00002927',
+	"oS;":                              '\U000024C8',
+	"oacute;":                          '\U000000F3',
+	"oast;":                            '\U0000229B',
+	"ocir;":                            '\U0000229A',
+	"ocirc;":                           '\U000000F4',
+	"ocy;":                             '\U0000043E',
+	"odash;":                           '\U0000229D',
+	"odblac;":                          '\U00000151',
+	"odiv;":                            '\U00002A38',
+	"odot;":                            '\U00002299',
+	"odsold;":                          '\U000029BC',
+	"oelig;":                           '\U00000153',
+	"ofcir;":                           '\U000029BF',
+	"ofr;":                             '\U0001D52C',
+	"ogon;":                            '\U000002DB',
+	"ograve;":                          '\U000000F2',
+	"ogt;":                             '\U000029C1',
+	"ohbar;":                           '\U000029B5',
+	"ohm;":                             '\U000003A9',
+	"oint;":                            '\U0000222E',
+	"olarr;":                           '\U000021BA',
+	"olcir;":                           '\U000029BE',
+	"olcross;":                         '\U000029BB',
+	"oline;":                           '\U0000203E',
+	"olt;":                             '\U000029C0',
+	"omacr;":                           '\U0000014D',
+	"omega;":                           '\U000003C9',
+	"omicron;":                         '\U000003BF',
+	"omid;":                            '\U000029B6',
+	"ominus;":                          '\U00002296',
+	"oopf;":                            '\U0001D560',
+	"opar;":                            '\U000029B7',
+	"operp;":                           '\U000029B9',
+	"oplus;":                           '\U00002295',
+	"or;":                              '\U00002228',
+	"orarr;":                           '\U000021BB',
+	"ord;":                             '\U00002A5D',
+	"order;":                           '\U00002134',
+	"orderof;":                         '\U00002134',
+	"ordf;":                            '\U000000AA',
+	"ordm;":                            '\U000000BA',
+	"origof;":                          '\U000022B6',
+	"oror;":                            '\U00002A56',
+	"orslope;":                         '\U00002A57',
+	"orv;":                             '\U00002A5B',
+	"oscr;":                            '\U00002134',
+	"oslash;":                          '\U000000F8',
+	"osol;":                            '\U00002298',
+	"otilde;":                          '\U000000F5',
+	"otimes;":                          '\U00002297',
+	"otimesas;":                        '\U00002A36',
+	"ouml;":                            '\U000000F6',
+	"ovbar;":                           '\U0000233D',
+	"par;":                             '\U00002225',
+	"para;":                            '\U000000B6',
+	"parallel;":                        '\U00002225',
+	"parsim;":                          '\U00002AF3',
+	"parsl;":                           '\U00002AFD',
+	"part;":                            '\U00002202',
+	"pcy;":                             '\U0000043F',
+	"percnt;":                          '\U00000025',
+	"period;":                          '\U0000002E',
+	"permil;":                          '\U00002030',
+	"perp;":                            '\U000022A5',
+	"pertenk;":                         '\U00002031',
+	"pfr;":                             '\U0001D52D',
+	"phi;":                             '\U000003C6',
+	"phiv;":                            '\U000003D5',
+	"phmmat;":                          '\U00002133',
+	"phone;":                           '\U0000260E',
+	"pi;":                              '\U000003C0',
+	"pitchfork;":                       '\U000022D4',
+	"piv;":                             '\U000003D6',
+	"planck;":                          '\U0000210F',
+	"planckh;":                         '\U0000210E',
+	"plankv;":                          '\U0000210F',
+	"plus;":                            '\U0000002B',
+	"plusacir;":                        '\U00002A23',
+	"plusb;":                           '\U0000229E',
+	"pluscir;":                         '\U00002A22',
+	"plusdo;":                          '\U00002214',
+	"plusdu;":                          '\U00002A25',
+	"pluse;":                           '\U00002A72',
+	"plusmn;":                          '\U000000B1',
+	"plussim;":                         '\U00002A26',
+	"plustwo;":                         '\U00002A27',
+	"pm;":                              '\U000000B1',
+	"pointint;":                        '\U00002A15',
+	"popf;":                            '\U0001D561',
+	"pound;":                           '\U000000A3',
+	"pr;":                              '\U0000227A',
+	"prE;":                             '\U00002AB3',
+	"prap;":                            '\U00002AB7',
+	"prcue;":                           '\U0000227C',
+	"pre;":                             '\U00002AAF',
+	"prec;":                            '\U0000227A',
+	"precapprox;":                      '\U00002AB7',
+	"preccurlyeq;":                     '\U0000227C',
+	"preceq;":                          '\U00002AAF',
+	"precnapprox;":                     '\U00002AB9',
+	"precneqq;":                        '\U00002AB5',
+	"precnsim;":                        '\U000022E8',
+	"precsim;":                         '\U0000227E',
+	"prime;":                           '\U00002032',
+	"primes;":                          '\U00002119',
+	"prnE;":                            '\U00002AB5',
+	"prnap;":                           '\U00002AB9',
+	"prnsim;":                          '\U000022E8',
+	"prod;":                            '\U0000220F',
+	"profalar;":                        '\U0000232E',
+	"profline;":                        '\U00002312',
+	"profsurf;":                        '\U00002313',
+	"prop;":                            '\U0000221D',
+	"propto;":                          '\U0000221D',
+	"prsim;":                           '\U0000227E',
+	"prurel;":                          '\U000022B0',
+	"pscr;":                            '\U0001D4C5',
+	"psi;":                             '\U000003C8',
+	"puncsp;":                          '\U00002008',
+	"qfr;":                             '\U0001D52E',
+	"qint;":                            '\U00002A0C',
+	"qopf;":                            '\U0001D562',
+	"qprime;":                          '\U00002057',
+	"qscr;":                            '\U0001D4C6',
+	"quaternions;":                     '\U0000210D',
+	"quatint;":                         '\U00002A16',
+	"quest;":                           '\U0000003F',
+	"questeq;":                         '\U0000225F',
+	"quot;":                            '\U00000022',
+	"rAarr;":                           '\U000021DB',
+	"rArr;":                            '\U000021D2',
+	"rAtail;":                          '\U0000291C',
+	"rBarr;":                           '\U0000290F',
+	"rHar;":                            '\U00002964',
+	"racute;":                          '\U00000155',
+	"radic;":                           '\U0000221A',
+	"raemptyv;":                        '\U000029B3',
+	"rang;":                            '\U000027E9',
+	"rangd;":                           '\U00002992',
+	"range;":                           '\U000029A5',
+	"rangle;":                          '\U000027E9',
+	"raquo;":                           '\U000000BB',
+	"rarr;":                            '\U00002192',
+	"rarrap;":                          '\U00002975',
+	"rarrb;":                           '\U000021E5',
+	"rarrbfs;":                         '\U00002920',
+	"rarrc;":                           '\U00002933',
+	"rarrfs;":                          '\U0000291E',
+	"rarrhk;":                          '\U000021AA',
+	"rarrlp;":                          '\U000021AC',
+	"rarrpl;":                          '\U00002945',
+	"rarrsim;":                         '\U00002974',
+	"rarrtl;":                          '\U000021A3',
+	"rarrw;":                           '\U0000219D',
+	"ratail;":                          '\U0000291A',
+	"ratio;":                           '\U00002236',
+	"rationals;":                       '\U0000211A',
+	"rbarr;":                           '\U0000290D',
+	"rbbrk;":                           '\U00002773',
+	"rbrace;":                          '\U0000007D',
+	"rbrack;":                          '\U0000005D',
+	"rbrke;":                           '\U0000298C',
+	"rbrksld;":                         '\U0000298E',
+	"rbrkslu;":                         '\U00002990',
+	"rcaron;":                          '\U00000159',
+	"rcedil;":                          '\U00000157',
+	"rceil;":                           '\U00002309',
+	"rcub;":                            '\U0000007D',
+	"rcy;":                             '\U00000440',
+	"rdca;":                            '\U00002937',
+	"rdldhar;":                         '\U00002969',
+	"rdquo;":                           '\U0000201D',
+	"rdquor;":                          '\U0000201D',
+	"rdsh;":                            '\U000021B3',
+	"real;":                            '\U0000211C',
+	"realine;":                         '\U0000211B',
+	"realpart;":                        '\U0000211C',
+	"reals;":                           '\U0000211D',
+	"rect;":                            '\U000025AD',
+	"reg;":                             '\U000000AE',
+	"rfisht;":                          '\U0000297D',
+	"rfloor;":                          '\U0000230B',
+	"rfr;":                             '\U0001D52F',
+	"rhard;":                           '\U000021C1',
+	"rharu;":                           '\U000021C0',
+	"rharul;":                          '\U0000296C',
+	"rho;":                             '\U000003C1',
+	"rhov;":                            '\U000003F1',
+	"rightarrow;":                      '\U00002192',
+	"rightarrowtail;":                  '\U000021A3',
+	"rightharpoondown;":                '\U000021C1',
+	"rightharpoonup;":                  '\U000021C0',
+	"rightleftarrows;":                 '\U000021C4',
+	"rightleftharpoons;":               '\U000021CC',
+	"rightrightarrows;":                '\U000021C9',
+	"rightsquigarrow;":                 '\U0000219D',
+	"rightthreetimes;":                 '\U000022CC',
+	"ring;":                            '\U000002DA',
+	"risingdotseq;":                    '\U00002253',
+	"rlarr;":                           '\U000021C4',
+	"rlhar;":                           '\U000021CC',
+	"rlm;":                             '\U0000200F',
+	"rmoust;":                          '\U000023B1',
+	"rmoustache;":                      '\U000023B1',
+	"rnmid;":                           '\U00002AEE',
+	"roang;":                           '\U000027ED',
+	"roarr;":                           '\U000021FE',
+	"robrk;":                           '\U000027E7',
+	"ropar;":                           '\U00002986',
+	"ropf;":                            '\U0001D563',
+	"roplus;":                          '\U00002A2E',
+	"rotimes;":                         '\U00002A35',
+	"rpar;":                            '\U00000029',
+	"rpargt;":                          '\U00002994',
+	"rppolint;":                        '\U00002A12',
+	"rrarr;":                           '\U000021C9',
+	"rsaquo;":                          '\U0000203A',
+	"rscr;":                            '\U0001D4C7',
+	"rsh;":                             '\U000021B1',
+	"rsqb;":                            '\U0000005D',
+	"rsquo;":                           '\U00002019',
+	"rsquor;":                          '\U00002019',
+	"rthree;":                          '\U000022CC',
+	"rtimes;":                          '\U000022CA',
+	"rtri;":                            '\U000025B9',
+	"rtrie;":                           '\U000022B5',
+	"rtrif;":                           '\U000025B8',
+	"rtriltri;":                        '\U000029CE',
+	"ruluhar;":                         '\U00002968',
+	"rx;":                              '\U0000211E',
+	"sacute;":                          '\U0000015B',
+	"sbquo;":                           '\U0000201A',
+	"sc;":                              '\U0000227B',
+	"scE;":                             '\U00002AB4',
+	"scap;":                            '\U00002AB8',
+	"scaron;":                          '\U00000161',
+	"sccue;":                           '\U0000227D',
+	"sce;":                             '\U00002AB0',
+	"scedil;":                          '\U0000015F',
+	"scirc;":                           '\U0000015D',
+	"scnE;":                            '\U00002AB6',
+	"scnap;":                           '\U00002ABA',
+	"scnsim;":                          '\U000022E9',
+	"scpolint;":                        '\U00002A13',
+	"scsim;":                           '\U0000227F',
+	"scy;":                             '\U00000441',
+	"sdot;":                            '\U000022C5',
+	"sdotb;":                           '\U000022A1',
+	"sdote;":                           '\U00002A66',
+	"seArr;":                           '\U000021D8',
+	"searhk;":                          '\U00002925',
+	"searr;":                           '\U00002198',
+	"searrow;":                         '\U00002198',
+	"sect;":                            '\U000000A7',
+	"semi;":                            '\U0000003B',
+	"seswar;":                          '\U00002929',
+	"setminus;":                        '\U00002216',
+	"setmn;":                           '\U00002216',
+	"sext;":                            '\U00002736',
+	"sfr;":                             '\U0001D530',
+	"sfrown;":                          '\U00002322',
+	"sharp;":                           '\U0000266F',
+	"shchcy;":                          '\U00000449',
+	"shcy;":                            '\U00000448',
+	"shortmid;":                        '\U00002223',
+	"shortparallel;":                   '\U00002225',
+	"shy;":                             '\U000000AD',
+	"sigma;":                           '\U000003C3',
+	"sigmaf;":                          '\U000003C2',
+	"sigmav;":                          '\U000003C2',
+	"sim;":                             '\U0000223C',
+	"simdot;":                          '\U00002A6A',
+	"sime;":                            '\U00002243',
+	"simeq;":                           '\U00002243',
+	"simg;":                            '\U00002A9E',
+	"simgE;":                           '\U00002AA0',
+	"siml;":                            '\U00002A9D',
+	"simlE;":                           '\U00002A9F',
+	"simne;":                           '\U00002246',
+	"simplus;":                         '\U00002A24',
+	"simrarr;":                         '\U00002972',
+	"slarr;":                           '\U00002190',
+	"smallsetminus;":                   '\U00002216',
+	"smashp;":                          '\U00002A33',
+	"smeparsl;":                        '\U000029E4',
+	"smid;":                            '\U00002223',
+	"smile;":                           '\U00002323',
+	"smt;":                             '\U00002AAA',
+	"smte;":                            '\U00002AAC',
+	"softcy;":                          '\U0000044C',
+	"sol;":                             '\U0000002F',
+	"solb;":                            '\U000029C4',
+	"solbar;":                          '\U0000233F',
+	"sopf;":                            '\U0001D564',
+	"spades;":                          '\U00002660',
+	"spadesuit;":                       '\U00002660',
+	"spar;":                            '\U00002225',
+	"sqcap;":                           '\U00002293',
+	"sqcup;":                           '\U00002294',
+	"sqsub;":                           '\U0000228F',
+	"sqsube;":                          '\U00002291',
+	"sqsubset;":                        '\U0000228F',
+	"sqsubseteq;":                      '\U00002291',
+	"sqsup;":                           '\U00002290',
+	"sqsupe;":                          '\U00002292',
+	"sqsupset;":                        '\U00002290',
+	"sqsupseteq;":                      '\U00002292',
+	"squ;":                             '\U000025A1',
+	"square;":                          '\U000025A1',
+	"squarf;":                          '\U000025AA',
+	"squf;":                            '\U000025AA',
+	"srarr;":                           '\U00002192',
+	"sscr;":                            '\U0001D4C8',
+	"ssetmn;":                          '\U00002216',
+	"ssmile;":                          '\U00002323',
+	"sstarf;":                          '\U000022C6',
+	"star;":                            '\U00002606',
+	"starf;":                           '\U00002605',
+	"straightepsilon;":                 '\U000003F5',
+	"straightphi;":                     '\U000003D5',
+	"strns;":                           '\U000000AF',
+	"sub;":                             '\U00002282',
+	"subE;":                            '\U00002AC5',
+	"subdot;":                          '\U00002ABD',
+	"sube;":                            '\U00002286',
+	"subedot;":                         '\U00002AC3',
+	"submult;":                         '\U00002AC1',
+	"subnE;":                           '\U00002ACB',
+	"subne;":                           '\U0000228A',
+	"subplus;":                         '\U00002ABF',
+	"subrarr;":                         '\U00002979',
+	"subset;":                          '\U00002282',
+	"subseteq;":                        '\U00002286',
+	"subseteqq;":                       '\U00002AC5',
+	"subsetneq;":                       '\U0000228A',
+	"subsetneqq;":                      '\U00002ACB',
+	"subsim;":                          '\U00002AC7',
+	"subsub;":                          '\U00002AD5',
+	"subsup;":                          '\U00002AD3',
+	"succ;":                            '\U0000227B',
+	"succapprox;":                      '\U00002AB8',
+	"succcurlyeq;":                     '\U0000227D',
+	"succeq;":                          '\U00002AB0',
+	"succnapprox;":                     '\U00002ABA',
+	"succneqq;":                        '\U00002AB6',
+	"succnsim;":                        '\U000022E9',
+	"succsim;":                         '\U0000227F',
+	"sum;":                             '\U00002211',
+	"sung;":                            '\U0000266A',
+	"sup;":                             '\U00002283',
+	"sup1;":                            '\U000000B9',
+	"sup2;":                            '\U000000B2',
+	"sup3;":                            '\U000000B3',
+	"supE;":                            '\U00002AC6',
+	"supdot;":                          '\U00002ABE',
+	"supdsub;":                         '\U00002AD8',
+	"supe;":                            '\U00002287',
+	"supedot;":                         '\U00002AC4',
+	"suphsol;":                         '\U000027C9',
+	"suphsub;":                         '\U00002AD7',
+	"suplarr;":                         '\U0000297B',
+	"supmult;":                         '\U00002AC2',
+	"supnE;":                           '\U00002ACC',
+	"supne;":                           '\U0000228B',
+	"supplus;":                         '\U00002AC0',
+	"supset;":                          '\U00002283',
+	"supseteq;":                        '\U00002287',
+	"supseteqq;":                       '\U00002AC6',
+	"supsetneq;":                       '\U0000228B',
+	"supsetneqq;":                      '\U00002ACC',
+	"supsim;":                          '\U00002AC8',
+	"supsub;":                          '\U00002AD4',
+	"supsup;":                          '\U00002AD6',
+	"swArr;":                           '\U000021D9',
+	"swarhk;":                          '\U00002926',
+	"swarr;":                           '\U00002199',
+	"swarrow;":                         '\U00002199',
+	"swnwar;":                          '\U0000292A',
+	"szlig;":                           '\U000000DF',
+	"target;":                          '\U00002316',
+	"tau;":                             '\U000003C4',
+	"tbrk;":                            '\U000023B4',
+	"tcaron;":                          '\U00000165',
+	"tcedil;":                          '\U00000163',
+	"tcy;":                             '\U00000442',
+	"tdot;":                            '\U000020DB',
+	"telrec;":                          '\U00002315',
+	"tfr;":                             '\U0001D531',
+	"there4;":                          '\U00002234',
+	"therefore;":                       '\U00002234',
+	"theta;":                           '\U000003B8',
+	"thetasym;":                        '\U000003D1',
+	"thetav;":                          '\U000003D1',
+	"thickapprox;":                     '\U00002248',
+	"thicksim;":                        '\U0000223C',
+	"thinsp;":                          '\U00002009',
+	"thkap;":                           '\U00002248',
+	"thksim;":                          '\U0000223C',
+	"thorn;":                           '\U000000FE',
+	"tilde;":                           '\U000002DC',
+	"times;":                           '\U000000D7',
+	"timesb;":                          '\U000022A0',
+	"timesbar;":                        '\U00002A31',
+	"timesd;":                          '\U00002A30',
+	"tint;":                            '\U0000222D',
+	"toea;":                            '\U00002928',
+	"top;":                             '\U000022A4',
+	"topbot;":                          '\U00002336',
+	"topcir;":                          '\U00002AF1',
+	"topf;":                            '\U0001D565',
+	"topfork;":                         '\U00002ADA',
+	"tosa;":                            '\U00002929',
+	"tprime;":                          '\U00002034',
+	"trade;":                           '\U00002122',
+	"triangle;":                        '\U000025B5',
+	"triangledown;":                    '\U000025BF',
+	"triangleleft;":                    '\U000025C3',
+	"trianglelefteq;":                  '\U000022B4',
+	"triangleq;":                       '\U0000225C',
+	"triangleright;":                   '\U000025B9',
+	"trianglerighteq;":                 '\U000022B5',
+	"tridot;":                          '\U000025EC',
+	"trie;":                            '\U0000225C',
+	"triminus;":                        '\U00002A3A',
+	"triplus;":                         '\U00002A39',
+	"trisb;":                           '\U000029CD',
+	"tritime;":                         '\U00002A3B',
+	"trpezium;":                        '\U000023E2',
+	"tscr;":                            '\U0001D4C9',
+	"tscy;":                            '\U00000446',
+	"tshcy;":                           '\U0000045B',
+	"tstrok;":                          '\U00000167',
+	"twixt;":                           '\U0000226C',
+	"twoheadleftarrow;":                '\U0000219E',
+	"twoheadrightarrow;":               '\U000021A0',
+	"uArr;":                            '\U000021D1',
+	"uHar;":                            '\U00002963',
+	"uacute;":                          '\U000000FA',
+	"uarr;":                            '\U00002191',
+	"ubrcy;":                           '\U0000045E',
+	"ubreve;":                          '\U0000016D',
+	"ucirc;":                           '\U000000FB',
+	"ucy;":                             '\U00000443',
+	"udarr;":                           '\U000021C5',
+	"udblac;":                          '\U00000171',
+	"udhar;":                           '\U0000296E',
+	"ufisht;":                          '\U0000297E',
+	"ufr;":                             '\U0001D532',
+	"ugrave;":                          '\U000000F9',
+	"uharl;":                           '\U000021BF',
+	"uharr;":                           '\U000021BE',
+	"uhblk;":                           '\U00002580',
+	"ulcorn;":                          '\U0000231C',
+	"ulcorner;":                        '\U0000231C',
+	"ulcrop;":                          '\U0000230F',
+	"ultri;":                           '\U000025F8',
+	"umacr;":                           '\U0000016B',
+	"uml;":                             '\U000000A8',
+	"uogon;":                           '\U00000173',
+	"uopf;":                            '\U0001D566',
+	"uparrow;":                         '\U00002191',
+	"updownarrow;":                     '\U00002195',
+	"upharpoonleft;":                   '\U000021BF',
+	"upharpoonright;":                  '\U000021BE',
+	"uplus;":                           '\U0000228E',
+	"upsi;":                            '\U000003C5',
+	"upsih;":                           '\U000003D2',
+	"upsilon;":                         '\U000003C5',
+	"upuparrows;":                      '\U000021C8',
+	"urcorn;":                          '\U0000231D',
+	"urcorner;":                        '\U0000231D',
+	"urcrop;":                          '\U0000230E',
+	"uring;":                           '\U0000016F',
+	"urtri;":                           '\U000025F9',
+	"uscr;":                            '\U0001D4CA',
+	"utdot;":                           '\U000022F0',
+	"utilde;":                          '\U00000169',
+	"utri;":                            '\U000025B5',
+	"utrif;":                           '\U000025B4',
+	"uuarr;":                           '\U000021C8',
+	"uuml;":                            '\U000000FC',
+	"uwangle;":                         '\U000029A7',
+	"vArr;":                            '\U000021D5',
+	"vBar;":                            '\U00002AE8',
+	"vBarv;":                           '\U00002AE9',
+	"vDash;":                           '\U000022A8',
+	"vangrt;":                          '\U0000299C',
+	"varepsilon;":                      '\U000003F5',
+	"varkappa;":                        '\U000003F0',
+	"varnothing;":                      '\U00002205',
+	"varphi;":                          '\U000003D5',
+	"varpi;":                           '\U000003D6',
+	"varpropto;":                       '\U0000221D',
+	"varr;":                            '\U00002195',
+	"varrho;":                          '\U000003F1',
+	"varsigma;":                        '\U000003C2',
+	"vartheta;":                        '\U000003D1',
+	"vartriangleleft;":                 '\U000022B2',
+	"vartriangleright;":                '\U000022B3',
+	"vcy;":                             '\U00000432',
+	"vdash;":                           '\U000022A2',
+	"vee;":                             '\U00002228',
+	"veebar;":                          '\U000022BB',
+	"veeeq;":                           '\U0000225A',
+	"vellip;":                          '\U000022EE',
+	"verbar;":                          '\U0000007C',
+	"vert;":                            '\U0000007C',
+	"vfr;":                             '\U0001D533',
+	"vltri;":                           '\U000022B2',
+	"vopf;":                            '\U0001D567',
+	"vprop;":                           '\U0000221D',
+	"vrtri;":                           '\U000022B3',
+	"vscr;":                            '\U0001D4CB',
+	"vzigzag;":                         '\U0000299A',
+	"wcirc;":                           '\U00000175',
+	"wedbar;":                          '\U00002A5F',
+	"wedge;":                           '\U00002227',
+	"wedgeq;":                          '\U00002259',
+	"weierp;":                          '\U00002118',
+	"wfr;":                             '\U0001D534',
+	"wopf;":                            '\U0001D568',
+	"wp;":                              '\U00002118',
+	"wr;":                              '\U00002240',
+	"wreath;":                          '\U00002240',
+	"wscr;":                            '\U0001D4CC',
+	"xcap;":                            '\U000022C2',
+	"xcirc;":                           '\U000025EF',
+	"xcup;":                            '\U000022C3',
+	"xdtri;":                           '\U000025BD',
+	"xfr;":                             '\U0001D535',
+	"xhArr;":                           '\U000027FA',
+	"xharr;":                           '\U000027F7',
+	"xi;":                              '\U000003BE',
+	"xlArr;":                           '\U000027F8',
+	"xlarr;":                           '\U000027F5',
+	"xmap;":                            '\U000027FC',
+	"xnis;":                            '\U000022FB',
+	"xodot;":                           '\U00002A00',
+	"xopf;":                            '\U0001D569',
+	"xoplus;":                          '\U00002A01',
+	"xotime;":                          '\U00002A02',
+	"xrArr;":                           '\U000027F9',
+	"xrarr;":                           '\U000027F6',
+	"xscr;":                            '\U0001D4CD',
+	"xsqcup;":                          '\U00002A06',
+	"xuplus;":                          '\U00002A04',
+	"xutri;":                           '\U000025B3',
+	"xvee;":                            '\U000022C1',
+	"xwedge;":                          '\U000022C0',
+	"yacute;":                          '\U000000FD',
+	"yacy;":                            '\U0000044F',
+	"ycirc;":                           '\U00000177',
+	"ycy;":                             '\U0000044B',
+	"yen;":                             '\U000000A5',
+	"yfr;":                             '\U0001D536',
+	"yicy;":                            '\U00000457',
+	"yopf;":                            '\U0001D56A',
+	"yscr;":                            '\U0001D4CE',
+	"yucy;":                            '\U0000044E',
+	"yuml;":                            '\U000000FF',
+	"zacute;":                          '\U0000017A',
+	"zcaron;":                          '\U0000017E',
+	"zcy;":                             '\U00000437',
+	"zdot;":                            '\U0000017C',
+	"zeetrf;":                          '\U00002128',
+	"zeta;":                            '\U000003B6',
+	"zfr;":                             '\U0001D537',
+	"zhcy;":                            '\U00000436',
+	"zigrarr;":                         '\U000021DD',
+	"zopf;":                            '\U0001D56B',
+	"zscr;":                            '\U0001D4CF',
+	"zwj;":                             '\U0000200D',
+	"zwnj;":                            '\U0000200C',
+	"AElig":                            '\U000000C6',
+	"AMP":                              '\U00000026',
+	"Aacute":                           '\U000000C1',
+	"Acirc":                            '\U000000C2',
+	"Agrave":                           '\U000000C0',
+	"Aring":                            '\U000000C5',
+	"Atilde":                           '\U000000C3',
+	"Auml":                             '\U000000C4',
+	"COPY":                             '\U000000A9',
+	"Ccedil":                           '\U000000C7',
+	"ETH":                              '\U000000D0',
+	"Eacute":                           '\U000000C9',
+	"Ecirc":                            '\U000000CA',
+	"Egrave":                           '\U000000C8',
+	"Euml":                             '\U000000CB',
+	"GT":                               '\U0000003E',
+	"Iacute":                           '\U000000CD',
+	"Icirc":                            '\U000000CE',
+	"Igrave":                           '\U000000CC',
+	"Iuml":                             '\U000000CF',
+	"LT":                               '\U0000003C',
+	"Ntilde":                           '\U000000D1',
+	"Oacute":                           '\U000000D3',
+	"Ocirc":                            '\U000000D4',
+	"Ograve":                           '\U000000D2',
+	"Oslash":                           '\U000000D8',
+	"Otilde":                           '\U000000D5',
+	"Ouml":                             '\U000000D6',
+	"QUOT":                             '\U00000022',
+	"REG":                              '\U000000AE',
+	"THORN":                            '\U000000DE',
+	"Uacute":                           '\U000000DA',
+	"Ucirc":                            '\U000000DB',
+	"Ugrave":                           '\U000000D9',
+	"Uuml":                             '\U000000DC',
+	"Yacute":                           '\U000000DD',
+	"aacute":                           '\U000000E1',
+	"acirc":                            '\U000000E2',
+	"acute":                            '\U000000B4',
+	"aelig":                            '\U000000E6',
+	"agrave":                           '\U000000E0',
+	"amp":                              '\U00000026',
+	"aring":                            '\U000000E5',
+	"atilde":                           '\U000000E3',
+	"auml":                             '\U000000E4',
+	"brvbar":                           '\U000000A6',
+	"ccedil":                           '\U000000E7',
+	"cedil":                            '\U000000B8',
+	"cent":                             '\U000000A2',
+	"copy":                             '\U000000A9',
+	"curren":                           '\U000000A4',
+	"deg":                              '\U000000B0',
+	"divide":                           '\U000000F7',
+	"eacute":                           '\U000000E9',
+	"ecirc":                            '\U000000EA',
+	"egrave":                           '\U000000E8',
+	"eth":                              '\U000000F0',
+	"euml":                             '\U000000EB',
+	"frac12":                           '\U000000BD',
+	"frac14":                           '\U000000BC',
+	"frac34":                           '\U000000BE',
+	"gt":                               '\U0000003E',
+	"iacute":                           '\U000000ED',
+	"icirc":                            '\U000000EE',
+	"iexcl":                            '\U000000A1',
+	"igrave":                           '\U000000EC',
+	"iquest":                           '\U000000BF',
+	"iuml":                             '\U000000EF',
+	"laquo":                            '\U000000AB',
+	"lt":                               '\U0000003C',
+	"macr":                             '\U000000AF',
+	"micro":                            '\U000000B5',
+	"middot":                           '\U000000B7',
+	"nbsp":                             '\U000000A0',
+	"not":                              '\U000000AC',
+	"ntilde":                           '\U000000F1',
+	"oacute":                           '\U000000F3',
+	"ocirc":                            '\U000000F4',
+	"ograve":                           '\U000000F2',
+	"ordf":                             '\U000000AA',
+	"ordm":                             '\U000000BA',
+	"oslash":                           '\U000000F8',
+	"otilde":                           '\U000000F5',
+	"ouml":                             '\U000000F6',
+	"para":                             '\U000000B6',
+	"plusmn":                           '\U000000B1',
+	"pound":                            '\U000000A3',
+	"quot":                             '\U00000022',
+	"raquo":                            '\U000000BB',
+	"reg":                              '\U000000AE',
+	"sect":                             '\U000000A7',
+	"shy":                              '\U000000AD',
+	"sup1":                             '\U000000B9',
+	"sup2":                             '\U000000B2',
+	"sup3":                             '\U000000B3',
+	"szlig":                            '\U000000DF',
+	"thorn":                            '\U000000FE',
+	"times":                            '\U000000D7',
+	"uacute":                           '\U000000FA',
+	"ucirc":                            '\U000000FB',
+	"ugrave":                           '\U000000F9',
+	"uml":                              '\U000000A8',
+	"uuml":                             '\U000000FC',
+	"yacute":                           '\U000000FD',
+	"yen":                              '\U000000A5',
+	"yuml":                             '\U000000FF',
+}
+
+// HTML entities that are two unicode codepoints.
+var entity2 = map[string][2]rune{
+	// TODO(nigeltao): Handle replacements that are wider than their names.
+	// "nLt;":                     {'\u226A', '\u20D2'},
+	// "nGt;":                     {'\u226B', '\u20D2'},
+	"NotEqualTilde;":           {'\u2242', '\u0338'},
+	"NotGreaterFullEqual;":     {'\u2267', '\u0338'},
+	"NotGreaterGreater;":       {'\u226B', '\u0338'},
+	"NotGreaterSlantEqual;":    {'\u2A7E', '\u0338'},
+	"NotHumpDownHump;":         {'\u224E', '\u0338'},
+	"NotHumpEqual;":            {'\u224F', '\u0338'},
+	"NotLeftTriangleBar;":      {'\u29CF', '\u0338'},
+	"NotLessLess;":             {'\u226A', '\u0338'},
+	"NotLessSlantEqual;":       {'\u2A7D', '\u0338'},
+	"NotNestedGreaterGreater;": {'\u2AA2', '\u0338'},
+	"NotNestedLessLess;":       {'\u2AA1', '\u0338'},
+	"NotPrecedesEqual;":        {'\u2AAF', '\u0338'},
+	"NotRightTriangleBar;":     {'\u29D0', '\u0338'},
+	"NotSquareSubset;":         {'\u228F', '\u0338'},
+	"NotSquareSuperset;":       {'\u2290', '\u0338'},
+	"NotSubset;":               {'\u2282', '\u20D2'},
+	"NotSucceedsEqual;":        {'\u2AB0', '\u0338'},
+	"NotSucceedsTilde;":        {'\u227F', '\u0338'},
+	"NotSuperset;":             {'\u2283', '\u20D2'},
+	"ThickSpace;":              {'\u205F', '\u200A'},
+	"acE;":                     {'\u223E', '\u0333'},
+	"bne;":                     {'\u003D', '\u20E5'},
+	"bnequiv;":                 {'\u2261', '\u20E5'},
+	"caps;":                    {'\u2229', '\uFE00'},
+	"cups;":                    {'\u222A', '\uFE00'},
+	"fjlig;":                   {'\u0066', '\u006A'},
+	"gesl;":                    {'\u22DB', '\uFE00'},
+	"gvertneqq;":               {'\u2269', '\uFE00'},
+	"gvnE;":                    {'\u2269', '\uFE00'},
+	"lates;":                   {'\u2AAD', '\uFE00'},
+	"lesg;":                    {'\u22DA', '\uFE00'},
+	"lvertneqq;":               {'\u2268', '\uFE00'},
+	"lvnE;":                    {'\u2268', '\uFE00'},
+	"nGg;":                     {'\u22D9', '\u0338'},
+	"nGtv;":                    {'\u226B', '\u0338'},
+	"nLl;":                     {'\u22D8', '\u0338'},
+	"nLtv;":                    {'\u226A', '\u0338'},
+	"nang;":                    {'\u2220', '\u20D2'},
+	"napE;":                    {'\u2A70', '\u0338'},
+	"napid;":                   {'\u224B', '\u0338'},
+	"nbump;":                   {'\u224E', '\u0338'},
+	"nbumpe;":                  {'\u224F', '\u0338'},
+	"ncongdot;":                {'\u2A6D', '\u0338'},
+	"nedot;":                   {'\u2250', '\u0338'},
+	"nesim;":                   {'\u2242', '\u0338'},
+	"ngE;":                     {'\u2267', '\u0338'},
+	"ngeqq;":                   {'\u2267', '\u0338'},
+	"ngeqslant;":               {'\u2A7E', '\u0338'},
+	"nges;":                    {'\u2A7E', '\u0338'},
+	"nlE;":                     {'\u2266', '\u0338'},
+	"nleqq;":                   {'\u2266', '\u0338'},
+	"nleqslant;":               {'\u2A7D', '\u0338'},
+	"nles;":                    {'\u2A7D', '\u0338'},
+	"notinE;":                  {'\u22F9', '\u0338'},
+	"notindot;":                {'\u22F5', '\u0338'},
+	"nparsl;":                  {'\u2AFD', '\u20E5'},
+	"npart;":                   {'\u2202', '\u0338'},
+	"npre;":                    {'\u2AAF', '\u0338'},
+	"npreceq;":                 {'\u2AAF', '\u0338'},
+	"nrarrc;":                  {'\u2933', '\u0338'},
+	"nrarrw;":                  {'\u219D', '\u0338'},
+	"nsce;":                    {'\u2AB0', '\u0338'},
+	"nsubE;":                   {'\u2AC5', '\u0338'},
+	"nsubset;":                 {'\u2282', '\u20D2'},
+	"nsubseteqq;":              {'\u2AC5', '\u0338'},
+	"nsucceq;":                 {'\u2AB0', '\u0338'},
+	"nsupE;":                   {'\u2AC6', '\u0338'},
+	"nsupset;":                 {'\u2283', '\u20D2'},
+	"nsupseteqq;":              {'\u2AC6', '\u0338'},
+	"nvap;":                    {'\u224D', '\u20D2'},
+	"nvge;":                    {'\u2265', '\u20D2'},
+	"nvgt;":                    {'\u003E', '\u20D2'},
+	"nvle;":                    {'\u2264', '\u20D2'},
+	"nvlt;":                    {'\u003C', '\u20D2'},
+	"nvltrie;":                 {'\u22B4', '\u20D2'},
+	"nvrtrie;":                 {'\u22B5', '\u20D2'},
+	"nvsim;":                   {'\u223C', '\u20D2'},
+	"race;":                    {'\u223D', '\u0331'},
+	"smtes;":                   {'\u2AAC', '\uFE00'},
+	"sqcaps;":                  {'\u2293', '\uFE00'},
+	"sqcups;":                  {'\u2294', '\uFE00'},
+	"varsubsetneq;":            {'\u228A', '\uFE00'},
+	"varsubsetneqq;":           {'\u2ACB', '\uFE00'},
+	"varsupsetneq;":            {'\u228B', '\uFE00'},
+	"varsupsetneqq;":           {'\u2ACC', '\uFE00'},
+	"vnsub;":                   {'\u2282', '\u20D2'},
+	"vnsup;":                   {'\u2283', '\u20D2'},
+	"vsubnE;":                  {'\u2ACB', '\uFE00'},
+	"vsubne;":                  {'\u228A', '\uFE00'},
+	"vsupnE;":                  {'\u2ACC', '\uFE00'},
+	"vsupne;":                  {'\u228B', '\uFE00'},
+}
diff --git a/backend/vendor/golang.org/x/net/html/escape.go b/backend/vendor/golang.org/x/net/html/escape.go
new file mode 100644
index 00000000..d8561396
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/escape.go
@@ -0,0 +1,258 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"bytes"
+	"strings"
+	"unicode/utf8"
+)
+
+// These replacements permit compatibility with old numeric entities that
+// assumed Windows-1252 encoding.
+// https://html.spec.whatwg.org/multipage/syntax.html#consume-a-character-reference
+var replacementTable = [...]rune{
+	'\u20AC', // First entry is what 0x80 should be replaced with.
+	'\u0081',
+	'\u201A',
+	'\u0192',
+	'\u201E',
+	'\u2026',
+	'\u2020',
+	'\u2021',
+	'\u02C6',
+	'\u2030',
+	'\u0160',
+	'\u2039',
+	'\u0152',
+	'\u008D',
+	'\u017D',
+	'\u008F',
+	'\u0090',
+	'\u2018',
+	'\u2019',
+	'\u201C',
+	'\u201D',
+	'\u2022',
+	'\u2013',
+	'\u2014',
+	'\u02DC',
+	'\u2122',
+	'\u0161',
+	'\u203A',
+	'\u0153',
+	'\u009D',
+	'\u017E',
+	'\u0178', // Last entry is 0x9F.
+	// 0x00->'\uFFFD' is handled programmatically.
+	// 0x0D->'\u000D' is a no-op.
+}
+
+// unescapeEntity reads an entity like "&lt;" from b[src:] and writes the
+// corresponding "<" to b[dst:], returning the incremented dst and src cursors.
+// Precondition: b[src] == '&' && dst <= src.
+// attribute should be true if parsing an attribute value.
+func unescapeEntity(b []byte, dst, src int, attribute bool) (dst1, src1 int) {
+	// https://html.spec.whatwg.org/multipage/syntax.html#consume-a-character-reference
+
+	// i starts at 1 because we already know that s[0] == '&'.
+	i, s := 1, b[src:]
+
+	if len(s) <= 1 {
+		b[dst] = b[src]
+		return dst + 1, src + 1
+	}
+
+	if s[i] == '#' {
+		if len(s) <= 3 { // We need to have at least "&#.".
+			b[dst] = b[src]
+			return dst + 1, src + 1
+		}
+		i++
+		c := s[i]
+		hex := false
+		if c == 'x' || c == 'X' {
+			hex = true
+			i++
+		}
+
+		x := '\x00'
+		for i < len(s) {
+			c = s[i]
+			i++
+			if hex {
+				if '0' <= c && c <= '9' {
+					x = 16*x + rune(c) - '0'
+					continue
+				} else if 'a' <= c && c <= 'f' {
+					x = 16*x + rune(c) - 'a' + 10
+					continue
+				} else if 'A' <= c && c <= 'F' {
+					x = 16*x + rune(c) - 'A' + 10
+					continue
+				}
+			} else if '0' <= c && c <= '9' {
+				x = 10*x + rune(c) - '0'
+				continue
+			}
+			if c != ';' {
+				i--
+			}
+			break
+		}
+
+		if i <= 3 { // No characters matched.
+			b[dst] = b[src]
+			return dst + 1, src + 1
+		}
+
+		if 0x80 <= x && x <= 0x9F {
+			// Replace characters from Windows-1252 with UTF-8 equivalents.
+			x = replacementTable[x-0x80]
+		} else if x == 0 || (0xD800 <= x && x <= 0xDFFF) || x > 0x10FFFF {
+			// Replace invalid characters with the replacement character.
+			x = '\uFFFD'
+		}
+
+		return dst + utf8.EncodeRune(b[dst:], x), src + i
+	}
+
+	// Consume the maximum number of characters possible, with the
+	// consumed characters matching one of the named references.
+
+	for i < len(s) {
+		c := s[i]
+		i++
+		// Lower-cased characters are more common in entities, so we check for them first.
+		if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' || '0' <= c && c <= '9' {
+			continue
+		}
+		if c != ';' {
+			i--
+		}
+		break
+	}
+
+	entityName := string(s[1:i])
+	if entityName == "" {
+		// No-op.
+	} else if attribute && entityName[len(entityName)-1] != ';' && len(s) > i && s[i] == '=' {
+		// No-op.
+	} else if x := entity[entityName]; x != 0 {
+		return dst + utf8.EncodeRune(b[dst:], x), src + i
+	} else if x := entity2[entityName]; x[0] != 0 {
+		dst1 := dst + utf8.EncodeRune(b[dst:], x[0])
+		return dst1 + utf8.EncodeRune(b[dst1:], x[1]), src + i
+	} else if !attribute {
+		maxLen := len(entityName) - 1
+		if maxLen > longestEntityWithoutSemicolon {
+			maxLen = longestEntityWithoutSemicolon
+		}
+		for j := maxLen; j > 1; j-- {
+			if x := entity[entityName[:j]]; x != 0 {
+				return dst + utf8.EncodeRune(b[dst:], x), src + j + 1
+			}
+		}
+	}
+
+	dst1, src1 = dst+i, src+i
+	copy(b[dst:dst1], b[src:src1])
+	return dst1, src1
+}
+
+// unescape unescapes b's entities in-place, so that "a&lt;b" becomes "a<b".
+// attribute should be true if parsing an attribute value.
+func unescape(b []byte, attribute bool) []byte {
+	for i, c := range b {
+		if c == '&' {
+			dst, src := unescapeEntity(b, i, i, attribute)
+			for src < len(b) {
+				c := b[src]
+				if c == '&' {
+					dst, src = unescapeEntity(b, dst, src, attribute)
+				} else {
+					b[dst] = c
+					dst, src = dst+1, src+1
+				}
+			}
+			return b[0:dst]
+		}
+	}
+	return b
+}
+
+// lower lower-cases the A-Z bytes in b in-place, so that "aBc" becomes "abc".
+func lower(b []byte) []byte {
+	for i, c := range b {
+		if 'A' <= c && c <= 'Z' {
+			b[i] = c + 'a' - 'A'
+		}
+	}
+	return b
+}
+
+const escapedChars = "&'<>\"\r"
+
+func escape(w writer, s string) error {
+	i := strings.IndexAny(s, escapedChars)
+	for i != -1 {
+		if _, err := w.WriteString(s[:i]); err != nil {
+			return err
+		}
+		var esc string
+		switch s[i] {
+		case '&':
+			esc = "&amp;"
+		case '\'':
+			// "&#39;" is shorter than "&apos;" and apos was not in HTML until HTML5.
+			esc = "&#39;"
+		case '<':
+			esc = "&lt;"
+		case '>':
+			esc = "&gt;"
+		case '"':
+			// "&#34;" is shorter than "&quot;".
+			esc = "&#34;"
+		case '\r':
+			esc = "&#13;"
+		default:
+			panic("unrecognized escape character")
+		}
+		s = s[i+1:]
+		if _, err := w.WriteString(esc); err != nil {
+			return err
+		}
+		i = strings.IndexAny(s, escapedChars)
+	}
+	_, err := w.WriteString(s)
+	return err
+}
+
+// EscapeString escapes special characters like "<" to become "&lt;". It
+// escapes only five such characters: <, >, &, ' and ".
+// UnescapeString(EscapeString(s)) == s always holds, but the converse isn't
+// always true.
+func EscapeString(s string) string {
+	if strings.IndexAny(s, escapedChars) == -1 {
+		return s
+	}
+	var buf bytes.Buffer
+	escape(&buf, s)
+	return buf.String()
+}
+
+// UnescapeString unescapes entities like "&lt;" to become "<". It unescapes a
+// larger range of entities than EscapeString escapes. For example, "&aacute;"
+// unescapes to "á", as does "&#225;" and "&xE1;".
+// UnescapeString(EscapeString(s)) == s always holds, but the converse isn't
+// always true.
+func UnescapeString(s string) string {
+	for _, c := range s {
+		if c == '&' {
+			return string(unescape([]byte(s), false))
+		}
+	}
+	return s
+}
diff --git a/backend/vendor/golang.org/x/net/html/foreign.go b/backend/vendor/golang.org/x/net/html/foreign.go
new file mode 100644
index 00000000..01477a96
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/foreign.go
@@ -0,0 +1,226 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"strings"
+)
+
+func adjustAttributeNames(aa []Attribute, nameMap map[string]string) {
+	for i := range aa {
+		if newName, ok := nameMap[aa[i].Key]; ok {
+			aa[i].Key = newName
+		}
+	}
+}
+
+func adjustForeignAttributes(aa []Attribute) {
+	for i, a := range aa {
+		if a.Key == "" || a.Key[0] != 'x' {
+			continue
+		}
+		switch a.Key {
+		case "xlink:actuate", "xlink:arcrole", "xlink:href", "xlink:role", "xlink:show",
+			"xlink:title", "xlink:type", "xml:base", "xml:lang", "xml:space", "xmlns:xlink":
+			j := strings.Index(a.Key, ":")
+			aa[i].Namespace = a.Key[:j]
+			aa[i].Key = a.Key[j+1:]
+		}
+	}
+}
+
+func htmlIntegrationPoint(n *Node) bool {
+	if n.Type != ElementNode {
+		return false
+	}
+	switch n.Namespace {
+	case "math":
+		if n.Data == "annotation-xml" {
+			for _, a := range n.Attr {
+				if a.Key == "encoding" {
+					val := strings.ToLower(a.Val)
+					if val == "text/html" || val == "application/xhtml+xml" {
+						return true
+					}
+				}
+			}
+		}
+	case "svg":
+		switch n.Data {
+		case "desc", "foreignObject", "title":
+			return true
+		}
+	}
+	return false
+}
+
+func mathMLTextIntegrationPoint(n *Node) bool {
+	if n.Namespace != "math" {
+		return false
+	}
+	switch n.Data {
+	case "mi", "mo", "mn", "ms", "mtext":
+		return true
+	}
+	return false
+}
+
+// Section 12.2.6.5.
+var breakout = map[string]bool{
+	"b":          true,
+	"big":        true,
+	"blockquote": true,
+	"body":       true,
+	"br":         true,
+	"center":     true,
+	"code":       true,
+	"dd":         true,
+	"div":        true,
+	"dl":         true,
+	"dt":         true,
+	"em":         true,
+	"embed":      true,
+	"h1":         true,
+	"h2":         true,
+	"h3":         true,
+	"h4":         true,
+	"h5":         true,
+	"h6":         true,
+	"head":       true,
+	"hr":         true,
+	"i":          true,
+	"img":        true,
+	"li":         true,
+	"listing":    true,
+	"menu":       true,
+	"meta":       true,
+	"nobr":       true,
+	"ol":         true,
+	"p":          true,
+	"pre":        true,
+	"ruby":       true,
+	"s":          true,
+	"small":      true,
+	"span":       true,
+	"strong":     true,
+	"strike":     true,
+	"sub":        true,
+	"sup":        true,
+	"table":      true,
+	"tt":         true,
+	"u":          true,
+	"ul":         true,
+	"var":        true,
+}
+
+// Section 12.2.6.5.
+var svgTagNameAdjustments = map[string]string{
+	"altglyph":            "altGlyph",
+	"altglyphdef":         "altGlyphDef",
+	"altglyphitem":        "altGlyphItem",
+	"animatecolor":        "animateColor",
+	"animatemotion":       "animateMotion",
+	"animatetransform":    "animateTransform",
+	"clippath":            "clipPath",
+	"feblend":             "feBlend",
+	"fecolormatrix":       "feColorMatrix",
+	"fecomponenttransfer": "feComponentTransfer",
+	"fecomposite":         "feComposite",
+	"feconvolvematrix":    "feConvolveMatrix",
+	"fediffuselighting":   "feDiffuseLighting",
+	"fedisplacementmap":   "feDisplacementMap",
+	"fedistantlight":      "feDistantLight",
+	"feflood":             "feFlood",
+	"fefunca":             "feFuncA",
+	"fefuncb":             "feFuncB",
+	"fefuncg":             "feFuncG",
+	"fefuncr":             "feFuncR",
+	"fegaussianblur":      "feGaussianBlur",
+	"feimage":             "feImage",
+	"femerge":             "feMerge",
+	"femergenode":         "feMergeNode",
+	"femorphology":        "feMorphology",
+	"feoffset":            "feOffset",
+	"fepointlight":        "fePointLight",
+	"fespecularlighting":  "feSpecularLighting",
+	"fespotlight":         "feSpotLight",
+	"fetile":              "feTile",
+	"feturbulence":        "feTurbulence",
+	"foreignobject":       "foreignObject",
+	"glyphref":            "glyphRef",
+	"lineargradient":      "linearGradient",
+	"radialgradient":      "radialGradient",
+	"textpath":            "textPath",
+}
+
+// Section 12.2.6.1
+var mathMLAttributeAdjustments = map[string]string{
+	"definitionurl": "definitionURL",
+}
+
+var svgAttributeAdjustments = map[string]string{
+	"attributename":             "attributeName",
+	"attributetype":             "attributeType",
+	"basefrequency":             "baseFrequency",
+	"baseprofile":               "baseProfile",
+	"calcmode":                  "calcMode",
+	"clippathunits":             "clipPathUnits",
+	"contentscripttype":         "contentScriptType",
+	"contentstyletype":          "contentStyleType",
+	"diffuseconstant":           "diffuseConstant",
+	"edgemode":                  "edgeMode",
+	"externalresourcesrequired": "externalResourcesRequired",
+	"filterres":                 "filterRes",
+	"filterunits":               "filterUnits",
+	"glyphref":                  "glyphRef",
+	"gradienttransform":         "gradientTransform",
+	"gradientunits":             "gradientUnits",
+	"kernelmatrix":              "kernelMatrix",
+	"kernelunitlength":          "kernelUnitLength",
+	"keypoints":                 "keyPoints",
+	"keysplines":                "keySplines",
+	"keytimes":                  "keyTimes",
+	"lengthadjust":              "lengthAdjust",
+	"limitingconeangle":         "limitingConeAngle",
+	"markerheight":              "markerHeight",
+	"markerunits":               "markerUnits",
+	"markerwidth":               "markerWidth",
+	"maskcontentunits":          "maskContentUnits",
+	"maskunits":                 "maskUnits",
+	"numoctaves":                "numOctaves",
+	"pathlength":                "pathLength",
+	"patterncontentunits":       "patternContentUnits",
+	"patterntransform":          "patternTransform",
+	"patternunits":              "patternUnits",
+	"pointsatx":                 "pointsAtX",
+	"pointsaty":                 "pointsAtY",
+	"pointsatz":                 "pointsAtZ",
+	"preservealpha":             "preserveAlpha",
+	"preserveaspectratio":       "preserveAspectRatio",
+	"primitiveunits":            "primitiveUnits",
+	"refx":                      "refX",
+	"refy":                      "refY",
+	"repeatcount":               "repeatCount",
+	"repeatdur":                 "repeatDur",
+	"requiredextensions":        "requiredExtensions",
+	"requiredfeatures":          "requiredFeatures",
+	"specularconstant":          "specularConstant",
+	"specularexponent":          "specularExponent",
+	"spreadmethod":              "spreadMethod",
+	"startoffset":               "startOffset",
+	"stddeviation":              "stdDeviation",
+	"stitchtiles":               "stitchTiles",
+	"surfacescale":              "surfaceScale",
+	"systemlanguage":            "systemLanguage",
+	"tablevalues":               "tableValues",
+	"targetx":                   "targetX",
+	"targety":                   "targetY",
+	"textlength":                "textLength",
+	"viewbox":                   "viewBox",
+	"viewtarget":                "viewTarget",
+	"xchannelselector":          "xChannelSelector",
+	"ychannelselector":          "yChannelSelector",
+	"zoomandpan":                "zoomAndPan",
+}
diff --git a/backend/vendor/golang.org/x/net/html/node.go b/backend/vendor/golang.org/x/net/html/node.go
new file mode 100644
index 00000000..633ee15d
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/node.go
@@ -0,0 +1,220 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"golang.org/x/net/html/atom"
+)
+
+// A NodeType is the type of a Node.
+type NodeType uint32
+
+const (
+	ErrorNode NodeType = iota
+	TextNode
+	DocumentNode
+	ElementNode
+	CommentNode
+	DoctypeNode
+	scopeMarkerNode
+)
+
+// Section 12.2.4.3 says "The markers are inserted when entering applet,
+// object, marquee, template, td, th, and caption elements, and are used
+// to prevent formatting from "leaking" into applet, object, marquee,
+// template, td, th, and caption elements".
+var scopeMarker = Node{Type: scopeMarkerNode}
+
+// A Node consists of a NodeType and some Data (tag name for element nodes,
+// content for text) and are part of a tree of Nodes. Element nodes may also
+// have a Namespace and contain a slice of Attributes. Data is unescaped, so
+// that it looks like "a<b" rather than "a&lt;b". For element nodes, DataAtom
+// is the atom for Data, or zero if Data is not a known tag name.
+//
+// An empty Namespace implies a "http://www.w3.org/1999/xhtml" namespace.
+// Similarly, "math" is short for "http://www.w3.org/1998/Math/MathML", and
+// "svg" is short for "http://www.w3.org/2000/svg".
+type Node struct {
+	Parent, FirstChild, LastChild, PrevSibling, NextSibling *Node
+
+	Type      NodeType
+	DataAtom  atom.Atom
+	Data      string
+	Namespace string
+	Attr      []Attribute
+}
+
+// InsertBefore inserts newChild as a child of n, immediately before oldChild
+// in the sequence of n's children. oldChild may be nil, in which case newChild
+// is appended to the end of n's children.
+//
+// It will panic if newChild already has a parent or siblings.
+func (n *Node) InsertBefore(newChild, oldChild *Node) {
+	if newChild.Parent != nil || newChild.PrevSibling != nil || newChild.NextSibling != nil {
+		panic("html: InsertBefore called for an attached child Node")
+	}
+	var prev, next *Node
+	if oldChild != nil {
+		prev, next = oldChild.PrevSibling, oldChild
+	} else {
+		prev = n.LastChild
+	}
+	if prev != nil {
+		prev.NextSibling = newChild
+	} else {
+		n.FirstChild = newChild
+	}
+	if next != nil {
+		next.PrevSibling = newChild
+	} else {
+		n.LastChild = newChild
+	}
+	newChild.Parent = n
+	newChild.PrevSibling = prev
+	newChild.NextSibling = next
+}
+
+// AppendChild adds a node c as a child of n.
+//
+// It will panic if c already has a parent or siblings.
+func (n *Node) AppendChild(c *Node) {
+	if c.Parent != nil || c.PrevSibling != nil || c.NextSibling != nil {
+		panic("html: AppendChild called for an attached child Node")
+	}
+	last := n.LastChild
+	if last != nil {
+		last.NextSibling = c
+	} else {
+		n.FirstChild = c
+	}
+	n.LastChild = c
+	c.Parent = n
+	c.PrevSibling = last
+}
+
+// RemoveChild removes a node c that is a child of n. Afterwards, c will have
+// no parent and no siblings.
+//
+// It will panic if c's parent is not n.
+func (n *Node) RemoveChild(c *Node) {
+	if c.Parent != n {
+		panic("html: RemoveChild called for a non-child Node")
+	}
+	if n.FirstChild == c {
+		n.FirstChild = c.NextSibling
+	}
+	if c.NextSibling != nil {
+		c.NextSibling.PrevSibling = c.PrevSibling
+	}
+	if n.LastChild == c {
+		n.LastChild = c.PrevSibling
+	}
+	if c.PrevSibling != nil {
+		c.PrevSibling.NextSibling = c.NextSibling
+	}
+	c.Parent = nil
+	c.PrevSibling = nil
+	c.NextSibling = nil
+}
+
+// reparentChildren reparents all of src's child nodes to dst.
+func reparentChildren(dst, src *Node) {
+	for {
+		child := src.FirstChild
+		if child == nil {
+			break
+		}
+		src.RemoveChild(child)
+		dst.AppendChild(child)
+	}
+}
+
+// clone returns a new node with the same type, data and attributes.
+// The clone has no parent, no siblings and no children.
+func (n *Node) clone() *Node {
+	m := &Node{
+		Type:     n.Type,
+		DataAtom: n.DataAtom,
+		Data:     n.Data,
+		Attr:     make([]Attribute, len(n.Attr)),
+	}
+	copy(m.Attr, n.Attr)
+	return m
+}
+
+// nodeStack is a stack of nodes.
+type nodeStack []*Node
+
+// pop pops the stack. It will panic if s is empty.
+func (s *nodeStack) pop() *Node {
+	i := len(*s)
+	n := (*s)[i-1]
+	*s = (*s)[:i-1]
+	return n
+}
+
+// top returns the most recently pushed node, or nil if s is empty.
+func (s *nodeStack) top() *Node {
+	if i := len(*s); i > 0 {
+		return (*s)[i-1]
+	}
+	return nil
+}
+
+// index returns the index of the top-most occurrence of n in the stack, or -1
+// if n is not present.
+func (s *nodeStack) index(n *Node) int {
+	for i := len(*s) - 1; i >= 0; i-- {
+		if (*s)[i] == n {
+			return i
+		}
+	}
+	return -1
+}
+
+// contains returns whether a is within s.
+func (s *nodeStack) contains(a atom.Atom) bool {
+	for _, n := range *s {
+		if n.DataAtom == a && n.Namespace == "" {
+			return true
+		}
+	}
+	return false
+}
+
+// insert inserts a node at the given index.
+func (s *nodeStack) insert(i int, n *Node) {
+	(*s) = append(*s, nil)
+	copy((*s)[i+1:], (*s)[i:])
+	(*s)[i] = n
+}
+
+// remove removes a node from the stack. It is a no-op if n is not present.
+func (s *nodeStack) remove(n *Node) {
+	i := s.index(n)
+	if i == -1 {
+		return
+	}
+	copy((*s)[i:], (*s)[i+1:])
+	j := len(*s) - 1
+	(*s)[j] = nil
+	*s = (*s)[:j]
+}
+
+type insertionModeStack []insertionMode
+
+func (s *insertionModeStack) pop() (im insertionMode) {
+	i := len(*s)
+	im = (*s)[i-1]
+	*s = (*s)[:i-1]
+	return im
+}
+
+func (s *insertionModeStack) top() insertionMode {
+	if i := len(*s); i > 0 {
+		return (*s)[i-1]
+	}
+	return nil
+}
diff --git a/backend/vendor/golang.org/x/net/html/parse.go b/backend/vendor/golang.org/x/net/html/parse.go
new file mode 100644
index 00000000..992cff2a
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/parse.go
@@ -0,0 +1,2417 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"errors"
+	"fmt"
+	"io"
+	"strings"
+
+	a "golang.org/x/net/html/atom"
+)
+
+// A parser implements the HTML5 parsing algorithm:
+// https://html.spec.whatwg.org/multipage/syntax.html#tree-construction
+type parser struct {
+	// tokenizer provides the tokens for the parser.
+	tokenizer *Tokenizer
+	// tok is the most recently read token.
+	tok Token
+	// Self-closing tags like <hr/> are treated as start tags, except that
+	// hasSelfClosingToken is set while they are being processed.
+	hasSelfClosingToken bool
+	// doc is the document root element.
+	doc *Node
+	// The stack of open elements (section 12.2.4.2) and active formatting
+	// elements (section 12.2.4.3).
+	oe, afe nodeStack
+	// Element pointers (section 12.2.4.4).
+	head, form *Node
+	// Other parsing state flags (section 12.2.4.5).
+	scripting, framesetOK bool
+	// The stack of template insertion modes
+	templateStack insertionModeStack
+	// im is the current insertion mode.
+	im insertionMode
+	// originalIM is the insertion mode to go back to after completing a text
+	// or inTableText insertion mode.
+	originalIM insertionMode
+	// fosterParenting is whether new elements should be inserted according to
+	// the foster parenting rules (section 12.2.6.1).
+	fosterParenting bool
+	// quirks is whether the parser is operating in "quirks mode."
+	quirks bool
+	// fragment is whether the parser is parsing an HTML fragment.
+	fragment bool
+	// context is the context element when parsing an HTML fragment
+	// (section 12.4).
+	context *Node
+}
+
+func (p *parser) top() *Node {
+	if n := p.oe.top(); n != nil {
+		return n
+	}
+	return p.doc
+}
+
+// Stop tags for use in popUntil. These come from section 12.2.4.2.
+var (
+	defaultScopeStopTags = map[string][]a.Atom{
+		"":     {a.Applet, a.Caption, a.Html, a.Table, a.Td, a.Th, a.Marquee, a.Object, a.Template},
+		"math": {a.AnnotationXml, a.Mi, a.Mn, a.Mo, a.Ms, a.Mtext},
+		"svg":  {a.Desc, a.ForeignObject, a.Title},
+	}
+)
+
+type scope int
+
+const (
+	defaultScope scope = iota
+	listItemScope
+	buttonScope
+	tableScope
+	tableRowScope
+	tableBodyScope
+	selectScope
+)
+
+// popUntil pops the stack of open elements at the highest element whose tag
+// is in matchTags, provided there is no higher element in the scope's stop
+// tags (as defined in section 12.2.4.2). It returns whether or not there was
+// such an element. If there was not, popUntil leaves the stack unchanged.
+//
+// For example, the set of stop tags for table scope is: "html", "table". If
+// the stack was:
+// ["html", "body", "font", "table", "b", "i", "u"]
+// then popUntil(tableScope, "font") would return false, but
+// popUntil(tableScope, "i") would return true and the stack would become:
+// ["html", "body", "font", "table", "b"]
+//
+// If an element's tag is in both the stop tags and matchTags, then the stack
+// will be popped and the function returns true (provided, of course, there was
+// no higher element in the stack that was also in the stop tags). For example,
+// popUntil(tableScope, "table") returns true and leaves:
+// ["html", "body", "font"]
+func (p *parser) popUntil(s scope, matchTags ...a.Atom) bool {
+	if i := p.indexOfElementInScope(s, matchTags...); i != -1 {
+		p.oe = p.oe[:i]
+		return true
+	}
+	return false
+}
+
+// indexOfElementInScope returns the index in p.oe of the highest element whose
+// tag is in matchTags that is in scope. If no matching element is in scope, it
+// returns -1.
+func (p *parser) indexOfElementInScope(s scope, matchTags ...a.Atom) int {
+	for i := len(p.oe) - 1; i >= 0; i-- {
+		tagAtom := p.oe[i].DataAtom
+		if p.oe[i].Namespace == "" {
+			for _, t := range matchTags {
+				if t == tagAtom {
+					return i
+				}
+			}
+			switch s {
+			case defaultScope:
+				// No-op.
+			case listItemScope:
+				if tagAtom == a.Ol || tagAtom == a.Ul {
+					return -1
+				}
+			case buttonScope:
+				if tagAtom == a.Button {
+					return -1
+				}
+			case tableScope:
+				if tagAtom == a.Html || tagAtom == a.Table || tagAtom == a.Template {
+					return -1
+				}
+			case selectScope:
+				if tagAtom != a.Optgroup && tagAtom != a.Option {
+					return -1
+				}
+			default:
+				panic("unreachable")
+			}
+		}
+		switch s {
+		case defaultScope, listItemScope, buttonScope:
+			for _, t := range defaultScopeStopTags[p.oe[i].Namespace] {
+				if t == tagAtom {
+					return -1
+				}
+			}
+		}
+	}
+	return -1
+}
+
+// elementInScope is like popUntil, except that it doesn't modify the stack of
+// open elements.
+func (p *parser) elementInScope(s scope, matchTags ...a.Atom) bool {
+	return p.indexOfElementInScope(s, matchTags...) != -1
+}
+
+// clearStackToContext pops elements off the stack of open elements until a
+// scope-defined element is found.
+func (p *parser) clearStackToContext(s scope) {
+	for i := len(p.oe) - 1; i >= 0; i-- {
+		tagAtom := p.oe[i].DataAtom
+		switch s {
+		case tableScope:
+			if tagAtom == a.Html || tagAtom == a.Table || tagAtom == a.Template {
+				p.oe = p.oe[:i+1]
+				return
+			}
+		case tableRowScope:
+			if tagAtom == a.Html || tagAtom == a.Tr || tagAtom == a.Template {
+				p.oe = p.oe[:i+1]
+				return
+			}
+		case tableBodyScope:
+			if tagAtom == a.Html || tagAtom == a.Tbody || tagAtom == a.Tfoot || tagAtom == a.Thead || tagAtom == a.Template {
+				p.oe = p.oe[:i+1]
+				return
+			}
+		default:
+			panic("unreachable")
+		}
+	}
+}
+
+// generateImpliedEndTags pops nodes off the stack of open elements as long as
+// the top node has a tag name of dd, dt, li, optgroup, option, p, rb, rp, rt or rtc.
+// If exceptions are specified, nodes with that name will not be popped off.
+func (p *parser) generateImpliedEndTags(exceptions ...string) {
+	var i int
+loop:
+	for i = len(p.oe) - 1; i >= 0; i-- {
+		n := p.oe[i]
+		if n.Type == ElementNode {
+			switch n.DataAtom {
+			case a.Dd, a.Dt, a.Li, a.Optgroup, a.Option, a.P, a.Rb, a.Rp, a.Rt, a.Rtc:
+				for _, except := range exceptions {
+					if n.Data == except {
+						break loop
+					}
+				}
+				continue
+			}
+		}
+		break
+	}
+
+	p.oe = p.oe[:i+1]
+}
+
+// addChild adds a child node n to the top element, and pushes n onto the stack
+// of open elements if it is an element node.
+func (p *parser) addChild(n *Node) {
+	if p.shouldFosterParent() {
+		p.fosterParent(n)
+	} else {
+		p.top().AppendChild(n)
+	}
+
+	if n.Type == ElementNode {
+		p.oe = append(p.oe, n)
+	}
+}
+
+// shouldFosterParent returns whether the next node to be added should be
+// foster parented.
+func (p *parser) shouldFosterParent() bool {
+	if p.fosterParenting {
+		switch p.top().DataAtom {
+		case a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+			return true
+		}
+	}
+	return false
+}
+
+// fosterParent adds a child node according to the foster parenting rules.
+// Section 12.2.6.1, "foster parenting".
+func (p *parser) fosterParent(n *Node) {
+	var table, parent, prev, template *Node
+	var i int
+	for i = len(p.oe) - 1; i >= 0; i-- {
+		if p.oe[i].DataAtom == a.Table {
+			table = p.oe[i]
+			break
+		}
+	}
+
+	var j int
+	for j = len(p.oe) - 1; j >= 0; j-- {
+		if p.oe[j].DataAtom == a.Template {
+			template = p.oe[j]
+			break
+		}
+	}
+
+	if template != nil && (table == nil || j > i) {
+		template.AppendChild(n)
+		return
+	}
+
+	if table == nil {
+		// The foster parent is the html element.
+		parent = p.oe[0]
+	} else {
+		parent = table.Parent
+	}
+	if parent == nil {
+		parent = p.oe[i-1]
+	}
+
+	if table != nil {
+		prev = table.PrevSibling
+	} else {
+		prev = parent.LastChild
+	}
+	if prev != nil && prev.Type == TextNode && n.Type == TextNode {
+		prev.Data += n.Data
+		return
+	}
+
+	parent.InsertBefore(n, table)
+}
+
+// addText adds text to the preceding node if it is a text node, or else it
+// calls addChild with a new text node.
+func (p *parser) addText(text string) {
+	if text == "" {
+		return
+	}
+
+	if p.shouldFosterParent() {
+		p.fosterParent(&Node{
+			Type: TextNode,
+			Data: text,
+		})
+		return
+	}
+
+	t := p.top()
+	if n := t.LastChild; n != nil && n.Type == TextNode {
+		n.Data += text
+		return
+	}
+	p.addChild(&Node{
+		Type: TextNode,
+		Data: text,
+	})
+}
+
+// addElement adds a child element based on the current token.
+func (p *parser) addElement() {
+	p.addChild(&Node{
+		Type:     ElementNode,
+		DataAtom: p.tok.DataAtom,
+		Data:     p.tok.Data,
+		Attr:     p.tok.Attr,
+	})
+}
+
+// Section 12.2.4.3.
+func (p *parser) addFormattingElement() {
+	tagAtom, attr := p.tok.DataAtom, p.tok.Attr
+	p.addElement()
+
+	// Implement the Noah's Ark clause, but with three per family instead of two.
+	identicalElements := 0
+findIdenticalElements:
+	for i := len(p.afe) - 1; i >= 0; i-- {
+		n := p.afe[i]
+		if n.Type == scopeMarkerNode {
+			break
+		}
+		if n.Type != ElementNode {
+			continue
+		}
+		if n.Namespace != "" {
+			continue
+		}
+		if n.DataAtom != tagAtom {
+			continue
+		}
+		if len(n.Attr) != len(attr) {
+			continue
+		}
+	compareAttributes:
+		for _, t0 := range n.Attr {
+			for _, t1 := range attr {
+				if t0.Key == t1.Key && t0.Namespace == t1.Namespace && t0.Val == t1.Val {
+					// Found a match for this attribute, continue with the next attribute.
+					continue compareAttributes
+				}
+			}
+			// If we get here, there is no attribute that matches a.
+			// Therefore the element is not identical to the new one.
+			continue findIdenticalElements
+		}
+
+		identicalElements++
+		if identicalElements >= 3 {
+			p.afe.remove(n)
+		}
+	}
+
+	p.afe = append(p.afe, p.top())
+}
+
+// Section 12.2.4.3.
+func (p *parser) clearActiveFormattingElements() {
+	for {
+		n := p.afe.pop()
+		if len(p.afe) == 0 || n.Type == scopeMarkerNode {
+			return
+		}
+	}
+}
+
+// Section 12.2.4.3.
+func (p *parser) reconstructActiveFormattingElements() {
+	n := p.afe.top()
+	if n == nil {
+		return
+	}
+	if n.Type == scopeMarkerNode || p.oe.index(n) != -1 {
+		return
+	}
+	i := len(p.afe) - 1
+	for n.Type != scopeMarkerNode && p.oe.index(n) == -1 {
+		if i == 0 {
+			i = -1
+			break
+		}
+		i--
+		n = p.afe[i]
+	}
+	for {
+		i++
+		clone := p.afe[i].clone()
+		p.addChild(clone)
+		p.afe[i] = clone
+		if i == len(p.afe)-1 {
+			break
+		}
+	}
+}
+
+// Section 12.2.5.
+func (p *parser) acknowledgeSelfClosingTag() {
+	p.hasSelfClosingToken = false
+}
+
+// An insertion mode (section 12.2.4.1) is the state transition function from
+// a particular state in the HTML5 parser's state machine. It updates the
+// parser's fields depending on parser.tok (where ErrorToken means EOF).
+// It returns whether the token was consumed.
+type insertionMode func(*parser) bool
+
+// setOriginalIM sets the insertion mode to return to after completing a text or
+// inTableText insertion mode.
+// Section 12.2.4.1, "using the rules for".
+func (p *parser) setOriginalIM() {
+	if p.originalIM != nil {
+		panic("html: bad parser state: originalIM was set twice")
+	}
+	p.originalIM = p.im
+}
+
+// Section 12.2.4.1, "reset the insertion mode".
+func (p *parser) resetInsertionMode() {
+	for i := len(p.oe) - 1; i >= 0; i-- {
+		n := p.oe[i]
+		last := i == 0
+		if last && p.context != nil {
+			n = p.context
+		}
+
+		switch n.DataAtom {
+		case a.Select:
+			if !last {
+				for ancestor, first := n, p.oe[0]; ancestor != first; {
+					ancestor = p.oe[p.oe.index(ancestor)-1]
+					switch ancestor.DataAtom {
+					case a.Template:
+						p.im = inSelectIM
+						return
+					case a.Table:
+						p.im = inSelectInTableIM
+						return
+					}
+				}
+			}
+			p.im = inSelectIM
+		case a.Td, a.Th:
+			// TODO: remove this divergence from the HTML5 spec.
+			//
+			// See https://bugs.chromium.org/p/chromium/issues/detail?id=829668
+			p.im = inCellIM
+		case a.Tr:
+			p.im = inRowIM
+		case a.Tbody, a.Thead, a.Tfoot:
+			p.im = inTableBodyIM
+		case a.Caption:
+			p.im = inCaptionIM
+		case a.Colgroup:
+			p.im = inColumnGroupIM
+		case a.Table:
+			p.im = inTableIM
+		case a.Template:
+			// TODO: remove this divergence from the HTML5 spec.
+			if n.Namespace != "" {
+				continue
+			}
+			p.im = p.templateStack.top()
+		case a.Head:
+			// TODO: remove this divergence from the HTML5 spec.
+			//
+			// See https://bugs.chromium.org/p/chromium/issues/detail?id=829668
+			p.im = inHeadIM
+		case a.Body:
+			p.im = inBodyIM
+		case a.Frameset:
+			p.im = inFramesetIM
+		case a.Html:
+			if p.head == nil {
+				p.im = beforeHeadIM
+			} else {
+				p.im = afterHeadIM
+			}
+		default:
+			if last {
+				p.im = inBodyIM
+				return
+			}
+			continue
+		}
+		return
+	}
+}
+
+const whitespace = " \t\r\n\f"
+
+// Section 12.2.6.4.1.
+func initialIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		p.tok.Data = strings.TrimLeft(p.tok.Data, whitespace)
+		if len(p.tok.Data) == 0 {
+			// It was all whitespace, so ignore it.
+			return true
+		}
+	case CommentToken:
+		p.doc.AppendChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		n, quirks := parseDoctype(p.tok.Data)
+		p.doc.AppendChild(n)
+		p.quirks = quirks
+		p.im = beforeHTMLIM
+		return true
+	}
+	p.quirks = true
+	p.im = beforeHTMLIM
+	return false
+}
+
+// Section 12.2.6.4.2.
+func beforeHTMLIM(p *parser) bool {
+	switch p.tok.Type {
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	case TextToken:
+		p.tok.Data = strings.TrimLeft(p.tok.Data, whitespace)
+		if len(p.tok.Data) == 0 {
+			// It was all whitespace, so ignore it.
+			return true
+		}
+	case StartTagToken:
+		if p.tok.DataAtom == a.Html {
+			p.addElement()
+			p.im = beforeHeadIM
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Head, a.Body, a.Html, a.Br:
+			p.parseImpliedToken(StartTagToken, a.Html, a.Html.String())
+			return false
+		default:
+			// Ignore the token.
+			return true
+		}
+	case CommentToken:
+		p.doc.AppendChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	}
+	p.parseImpliedToken(StartTagToken, a.Html, a.Html.String())
+	return false
+}
+
+// Section 12.2.6.4.3.
+func beforeHeadIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		p.tok.Data = strings.TrimLeft(p.tok.Data, whitespace)
+		if len(p.tok.Data) == 0 {
+			// It was all whitespace, so ignore it.
+			return true
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Head:
+			p.addElement()
+			p.head = p.top()
+			p.im = inHeadIM
+			return true
+		case a.Html:
+			return inBodyIM(p)
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Head, a.Body, a.Html, a.Br:
+			p.parseImpliedToken(StartTagToken, a.Head, a.Head.String())
+			return false
+		default:
+			// Ignore the token.
+			return true
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	}
+
+	p.parseImpliedToken(StartTagToken, a.Head, a.Head.String())
+	return false
+}
+
+// Section 12.2.6.4.4.
+func inHeadIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) < len(p.tok.Data) {
+			// Add the initial whitespace to the current node.
+			p.addText(p.tok.Data[:len(p.tok.Data)-len(s)])
+			if s == "" {
+				return true
+			}
+			p.tok.Data = s
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Base, a.Basefont, a.Bgsound, a.Command, a.Link, a.Meta:
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+			return true
+		case a.Noscript:
+			p.addElement()
+			if p.scripting {
+				p.setOriginalIM()
+				p.im = textIM
+			} else {
+				p.im = inHeadNoscriptIM
+			}
+			return true
+		case a.Script, a.Title, a.Noframes, a.Style:
+			p.addElement()
+			p.setOriginalIM()
+			p.im = textIM
+			return true
+		case a.Head:
+			// Ignore the token.
+			return true
+		case a.Template:
+			p.addElement()
+			p.afe = append(p.afe, &scopeMarker)
+			p.framesetOK = false
+			p.im = inTemplateIM
+			p.templateStack = append(p.templateStack, inTemplateIM)
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Head:
+			p.oe.pop()
+			p.im = afterHeadIM
+			return true
+		case a.Body, a.Html, a.Br:
+			p.parseImpliedToken(EndTagToken, a.Head, a.Head.String())
+			return false
+		case a.Template:
+			if !p.oe.contains(a.Template) {
+				return true
+			}
+			// TODO: remove this divergence from the HTML5 spec.
+			//
+			// See https://bugs.chromium.org/p/chromium/issues/detail?id=829668
+			p.generateImpliedEndTags()
+			for i := len(p.oe) - 1; i >= 0; i-- {
+				if n := p.oe[i]; n.Namespace == "" && n.DataAtom == a.Template {
+					p.oe = p.oe[:i]
+					break
+				}
+			}
+			p.clearActiveFormattingElements()
+			p.templateStack.pop()
+			p.resetInsertionMode()
+			return true
+		default:
+			// Ignore the token.
+			return true
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	}
+
+	p.parseImpliedToken(EndTagToken, a.Head, a.Head.String())
+	return false
+}
+
+// 12.2.6.4.5.
+func inHeadNoscriptIM(p *parser) bool {
+	switch p.tok.Type {
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Basefont, a.Bgsound, a.Link, a.Meta, a.Noframes, a.Style:
+			return inHeadIM(p)
+		case a.Head, a.Noscript:
+			// Ignore the token.
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Noscript, a.Br:
+		default:
+			// Ignore the token.
+			return true
+		}
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) == 0 {
+			// It was all whitespace.
+			return inHeadIM(p)
+		}
+	case CommentToken:
+		return inHeadIM(p)
+	}
+	p.oe.pop()
+	if p.top().DataAtom != a.Head {
+		panic("html: the new current node will be a head element.")
+	}
+	p.im = inHeadIM
+	if p.tok.DataAtom == a.Noscript {
+		return true
+	}
+	return false
+}
+
+// Section 12.2.6.4.6.
+func afterHeadIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) < len(p.tok.Data) {
+			// Add the initial whitespace to the current node.
+			p.addText(p.tok.Data[:len(p.tok.Data)-len(s)])
+			if s == "" {
+				return true
+			}
+			p.tok.Data = s
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Body:
+			p.addElement()
+			p.framesetOK = false
+			p.im = inBodyIM
+			return true
+		case a.Frameset:
+			p.addElement()
+			p.im = inFramesetIM
+			return true
+		case a.Base, a.Basefont, a.Bgsound, a.Link, a.Meta, a.Noframes, a.Script, a.Style, a.Template, a.Title:
+			p.oe = append(p.oe, p.head)
+			defer p.oe.remove(p.head)
+			return inHeadIM(p)
+		case a.Head:
+			// Ignore the token.
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Body, a.Html, a.Br:
+			// Drop down to creating an implied <body> tag.
+		case a.Template:
+			return inHeadIM(p)
+		default:
+			// Ignore the token.
+			return true
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	}
+
+	p.parseImpliedToken(StartTagToken, a.Body, a.Body.String())
+	p.framesetOK = true
+	return false
+}
+
+// copyAttributes copies attributes of src not found on dst to dst.
+func copyAttributes(dst *Node, src Token) {
+	if len(src.Attr) == 0 {
+		return
+	}
+	attr := map[string]string{}
+	for _, t := range dst.Attr {
+		attr[t.Key] = t.Val
+	}
+	for _, t := range src.Attr {
+		if _, ok := attr[t.Key]; !ok {
+			dst.Attr = append(dst.Attr, t)
+			attr[t.Key] = t.Val
+		}
+	}
+}
+
+// Section 12.2.6.4.7.
+func inBodyIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		d := p.tok.Data
+		switch n := p.oe.top(); n.DataAtom {
+		case a.Pre, a.Listing:
+			if n.FirstChild == nil {
+				// Ignore a newline at the start of a <pre> block.
+				if d != "" && d[0] == '\r' {
+					d = d[1:]
+				}
+				if d != "" && d[0] == '\n' {
+					d = d[1:]
+				}
+			}
+		}
+		d = strings.Replace(d, "\x00", "", -1)
+		if d == "" {
+			return true
+		}
+		p.reconstructActiveFormattingElements()
+		p.addText(d)
+		if p.framesetOK && strings.TrimLeft(d, whitespace) != "" {
+			// There were non-whitespace characters inserted.
+			p.framesetOK = false
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			if p.oe.contains(a.Template) {
+				return true
+			}
+			copyAttributes(p.oe[0], p.tok)
+		case a.Base, a.Basefont, a.Bgsound, a.Command, a.Link, a.Meta, a.Noframes, a.Script, a.Style, a.Template, a.Title:
+			return inHeadIM(p)
+		case a.Body:
+			if p.oe.contains(a.Template) {
+				return true
+			}
+			if len(p.oe) >= 2 {
+				body := p.oe[1]
+				if body.Type == ElementNode && body.DataAtom == a.Body {
+					p.framesetOK = false
+					copyAttributes(body, p.tok)
+				}
+			}
+		case a.Frameset:
+			if !p.framesetOK || len(p.oe) < 2 || p.oe[1].DataAtom != a.Body {
+				// Ignore the token.
+				return true
+			}
+			body := p.oe[1]
+			if body.Parent != nil {
+				body.Parent.RemoveChild(body)
+			}
+			p.oe = p.oe[:1]
+			p.addElement()
+			p.im = inFramesetIM
+			return true
+		case a.Address, a.Article, a.Aside, a.Blockquote, a.Center, a.Details, a.Dir, a.Div, a.Dl, a.Fieldset, a.Figcaption, a.Figure, a.Footer, a.Header, a.Hgroup, a.Menu, a.Nav, a.Ol, a.P, a.Section, a.Summary, a.Ul:
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+		case a.H1, a.H2, a.H3, a.H4, a.H5, a.H6:
+			p.popUntil(buttonScope, a.P)
+			switch n := p.top(); n.DataAtom {
+			case a.H1, a.H2, a.H3, a.H4, a.H5, a.H6:
+				p.oe.pop()
+			}
+			p.addElement()
+		case a.Pre, a.Listing:
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+			// The newline, if any, will be dealt with by the TextToken case.
+			p.framesetOK = false
+		case a.Form:
+			if p.form != nil && !p.oe.contains(a.Template) {
+				// Ignore the token
+				return true
+			}
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+			if !p.oe.contains(a.Template) {
+				p.form = p.top()
+			}
+		case a.Li:
+			p.framesetOK = false
+			for i := len(p.oe) - 1; i >= 0; i-- {
+				node := p.oe[i]
+				switch node.DataAtom {
+				case a.Li:
+					p.oe = p.oe[:i]
+				case a.Address, a.Div, a.P:
+					continue
+				default:
+					if !isSpecialElement(node) {
+						continue
+					}
+				}
+				break
+			}
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+		case a.Dd, a.Dt:
+			p.framesetOK = false
+			for i := len(p.oe) - 1; i >= 0; i-- {
+				node := p.oe[i]
+				switch node.DataAtom {
+				case a.Dd, a.Dt:
+					p.oe = p.oe[:i]
+				case a.Address, a.Div, a.P:
+					continue
+				default:
+					if !isSpecialElement(node) {
+						continue
+					}
+				}
+				break
+			}
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+		case a.Plaintext:
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+		case a.Button:
+			p.popUntil(defaultScope, a.Button)
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.framesetOK = false
+		case a.A:
+			for i := len(p.afe) - 1; i >= 0 && p.afe[i].Type != scopeMarkerNode; i-- {
+				if n := p.afe[i]; n.Type == ElementNode && n.DataAtom == a.A {
+					p.inBodyEndTagFormatting(a.A, "a")
+					p.oe.remove(n)
+					p.afe.remove(n)
+					break
+				}
+			}
+			p.reconstructActiveFormattingElements()
+			p.addFormattingElement()
+		case a.B, a.Big, a.Code, a.Em, a.Font, a.I, a.S, a.Small, a.Strike, a.Strong, a.Tt, a.U:
+			p.reconstructActiveFormattingElements()
+			p.addFormattingElement()
+		case a.Nobr:
+			p.reconstructActiveFormattingElements()
+			if p.elementInScope(defaultScope, a.Nobr) {
+				p.inBodyEndTagFormatting(a.Nobr, "nobr")
+				p.reconstructActiveFormattingElements()
+			}
+			p.addFormattingElement()
+		case a.Applet, a.Marquee, a.Object:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.afe = append(p.afe, &scopeMarker)
+			p.framesetOK = false
+		case a.Table:
+			if !p.quirks {
+				p.popUntil(buttonScope, a.P)
+			}
+			p.addElement()
+			p.framesetOK = false
+			p.im = inTableIM
+			return true
+		case a.Area, a.Br, a.Embed, a.Img, a.Input, a.Keygen, a.Wbr:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+			if p.tok.DataAtom == a.Input {
+				for _, t := range p.tok.Attr {
+					if t.Key == "type" {
+						if strings.ToLower(t.Val) == "hidden" {
+							// Skip setting framesetOK = false
+							return true
+						}
+					}
+				}
+			}
+			p.framesetOK = false
+		case a.Param, a.Source, a.Track:
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+		case a.Hr:
+			p.popUntil(buttonScope, a.P)
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+			p.framesetOK = false
+		case a.Image:
+			p.tok.DataAtom = a.Img
+			p.tok.Data = a.Img.String()
+			return false
+		case a.Isindex:
+			if p.form != nil {
+				// Ignore the token.
+				return true
+			}
+			action := ""
+			prompt := "This is a searchable index. Enter search keywords: "
+			attr := []Attribute{{Key: "name", Val: "isindex"}}
+			for _, t := range p.tok.Attr {
+				switch t.Key {
+				case "action":
+					action = t.Val
+				case "name":
+					// Ignore the attribute.
+				case "prompt":
+					prompt = t.Val
+				default:
+					attr = append(attr, t)
+				}
+			}
+			p.acknowledgeSelfClosingTag()
+			p.popUntil(buttonScope, a.P)
+			p.parseImpliedToken(StartTagToken, a.Form, a.Form.String())
+			if p.form == nil {
+				// NOTE: The 'isindex' element has been removed,
+				// and the 'template' element has not been designed to be
+				// collaborative with the index element.
+				//
+				// Ignore the token.
+				return true
+			}
+			if action != "" {
+				p.form.Attr = []Attribute{{Key: "action", Val: action}}
+			}
+			p.parseImpliedToken(StartTagToken, a.Hr, a.Hr.String())
+			p.parseImpliedToken(StartTagToken, a.Label, a.Label.String())
+			p.addText(prompt)
+			p.addChild(&Node{
+				Type:     ElementNode,
+				DataAtom: a.Input,
+				Data:     a.Input.String(),
+				Attr:     attr,
+			})
+			p.oe.pop()
+			p.parseImpliedToken(EndTagToken, a.Label, a.Label.String())
+			p.parseImpliedToken(StartTagToken, a.Hr, a.Hr.String())
+			p.parseImpliedToken(EndTagToken, a.Form, a.Form.String())
+		case a.Textarea:
+			p.addElement()
+			p.setOriginalIM()
+			p.framesetOK = false
+			p.im = textIM
+		case a.Xmp:
+			p.popUntil(buttonScope, a.P)
+			p.reconstructActiveFormattingElements()
+			p.framesetOK = false
+			p.addElement()
+			p.setOriginalIM()
+			p.im = textIM
+		case a.Iframe:
+			p.framesetOK = false
+			p.addElement()
+			p.setOriginalIM()
+			p.im = textIM
+		case a.Noembed, a.Noscript:
+			p.addElement()
+			p.setOriginalIM()
+			p.im = textIM
+		case a.Select:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.framesetOK = false
+			p.im = inSelectIM
+			return true
+		case a.Optgroup, a.Option:
+			if p.top().DataAtom == a.Option {
+				p.oe.pop()
+			}
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+		case a.Rb, a.Rtc:
+			if p.elementInScope(defaultScope, a.Ruby) {
+				p.generateImpliedEndTags()
+			}
+			p.addElement()
+		case a.Rp, a.Rt:
+			if p.elementInScope(defaultScope, a.Ruby) {
+				p.generateImpliedEndTags("rtc")
+			}
+			p.addElement()
+		case a.Math, a.Svg:
+			p.reconstructActiveFormattingElements()
+			if p.tok.DataAtom == a.Math {
+				adjustAttributeNames(p.tok.Attr, mathMLAttributeAdjustments)
+			} else {
+				adjustAttributeNames(p.tok.Attr, svgAttributeAdjustments)
+			}
+			adjustForeignAttributes(p.tok.Attr)
+			p.addElement()
+			p.top().Namespace = p.tok.Data
+			if p.hasSelfClosingToken {
+				p.oe.pop()
+				p.acknowledgeSelfClosingTag()
+			}
+			return true
+		case a.Caption, a.Col, a.Colgroup, a.Frame, a.Head, a.Tbody, a.Td, a.Tfoot, a.Th, a.Thead, a.Tr:
+			// Ignore the token.
+		default:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Body:
+			if p.elementInScope(defaultScope, a.Body) {
+				p.im = afterBodyIM
+			}
+		case a.Html:
+			if p.elementInScope(defaultScope, a.Body) {
+				p.parseImpliedToken(EndTagToken, a.Body, a.Body.String())
+				return false
+			}
+			return true
+		case a.Address, a.Article, a.Aside, a.Blockquote, a.Button, a.Center, a.Details, a.Dir, a.Div, a.Dl, a.Fieldset, a.Figcaption, a.Figure, a.Footer, a.Header, a.Hgroup, a.Listing, a.Menu, a.Nav, a.Ol, a.Pre, a.Section, a.Summary, a.Ul:
+			p.popUntil(defaultScope, p.tok.DataAtom)
+		case a.Form:
+			if p.oe.contains(a.Template) {
+				i := p.indexOfElementInScope(defaultScope, a.Form)
+				if i == -1 {
+					// Ignore the token.
+					return true
+				}
+				p.generateImpliedEndTags()
+				if p.oe[i].DataAtom != a.Form {
+					// Ignore the token.
+					return true
+				}
+				p.popUntil(defaultScope, a.Form)
+			} else {
+				node := p.form
+				p.form = nil
+				i := p.indexOfElementInScope(defaultScope, a.Form)
+				if node == nil || i == -1 || p.oe[i] != node {
+					// Ignore the token.
+					return true
+				}
+				p.generateImpliedEndTags()
+				p.oe.remove(node)
+			}
+		case a.P:
+			if !p.elementInScope(buttonScope, a.P) {
+				p.parseImpliedToken(StartTagToken, a.P, a.P.String())
+			}
+			p.popUntil(buttonScope, a.P)
+		case a.Li:
+			p.popUntil(listItemScope, a.Li)
+		case a.Dd, a.Dt:
+			p.popUntil(defaultScope, p.tok.DataAtom)
+		case a.H1, a.H2, a.H3, a.H4, a.H5, a.H6:
+			p.popUntil(defaultScope, a.H1, a.H2, a.H3, a.H4, a.H5, a.H6)
+		case a.A, a.B, a.Big, a.Code, a.Em, a.Font, a.I, a.Nobr, a.S, a.Small, a.Strike, a.Strong, a.Tt, a.U:
+			p.inBodyEndTagFormatting(p.tok.DataAtom, p.tok.Data)
+		case a.Applet, a.Marquee, a.Object:
+			if p.popUntil(defaultScope, p.tok.DataAtom) {
+				p.clearActiveFormattingElements()
+			}
+		case a.Br:
+			p.tok.Type = StartTagToken
+			return false
+		case a.Template:
+			return inHeadIM(p)
+		default:
+			p.inBodyEndTagOther(p.tok.DataAtom, p.tok.Data)
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case ErrorToken:
+		// TODO: remove this divergence from the HTML5 spec.
+		if len(p.templateStack) > 0 {
+			p.im = inTemplateIM
+			return false
+		} else {
+			for _, e := range p.oe {
+				switch e.DataAtom {
+				case a.Dd, a.Dt, a.Li, a.Optgroup, a.Option, a.P, a.Rb, a.Rp, a.Rt, a.Rtc, a.Tbody, a.Td, a.Tfoot, a.Th,
+					a.Thead, a.Tr, a.Body, a.Html:
+				default:
+					return true
+				}
+			}
+		}
+	}
+
+	return true
+}
+
+func (p *parser) inBodyEndTagFormatting(tagAtom a.Atom, tagName string) {
+	// This is the "adoption agency" algorithm, described at
+	// https://html.spec.whatwg.org/multipage/syntax.html#adoptionAgency
+
+	// TODO: this is a fairly literal line-by-line translation of that algorithm.
+	// Once the code successfully parses the comprehensive test suite, we should
+	// refactor this code to be more idiomatic.
+
+	// Steps 1-4. The outer loop.
+	for i := 0; i < 8; i++ {
+		// Step 5. Find the formatting element.
+		var formattingElement *Node
+		for j := len(p.afe) - 1; j >= 0; j-- {
+			if p.afe[j].Type == scopeMarkerNode {
+				break
+			}
+			if p.afe[j].DataAtom == tagAtom {
+				formattingElement = p.afe[j]
+				break
+			}
+		}
+		if formattingElement == nil {
+			p.inBodyEndTagOther(tagAtom, tagName)
+			return
+		}
+		feIndex := p.oe.index(formattingElement)
+		if feIndex == -1 {
+			p.afe.remove(formattingElement)
+			return
+		}
+		if !p.elementInScope(defaultScope, tagAtom) {
+			// Ignore the tag.
+			return
+		}
+
+		// Steps 9-10. Find the furthest block.
+		var furthestBlock *Node
+		for _, e := range p.oe[feIndex:] {
+			if isSpecialElement(e) {
+				furthestBlock = e
+				break
+			}
+		}
+		if furthestBlock == nil {
+			e := p.oe.pop()
+			for e != formattingElement {
+				e = p.oe.pop()
+			}
+			p.afe.remove(e)
+			return
+		}
+
+		// Steps 11-12. Find the common ancestor and bookmark node.
+		commonAncestor := p.oe[feIndex-1]
+		bookmark := p.afe.index(formattingElement)
+
+		// Step 13. The inner loop. Find the lastNode to reparent.
+		lastNode := furthestBlock
+		node := furthestBlock
+		x := p.oe.index(node)
+		// Steps 13.1-13.2
+		for j := 0; j < 3; j++ {
+			// Step 13.3.
+			x--
+			node = p.oe[x]
+			// Step 13.4 - 13.5.
+			if p.afe.index(node) == -1 {
+				p.oe.remove(node)
+				continue
+			}
+			// Step 13.6.
+			if node == formattingElement {
+				break
+			}
+			// Step 13.7.
+			clone := node.clone()
+			p.afe[p.afe.index(node)] = clone
+			p.oe[p.oe.index(node)] = clone
+			node = clone
+			// Step 13.8.
+			if lastNode == furthestBlock {
+				bookmark = p.afe.index(node) + 1
+			}
+			// Step 13.9.
+			if lastNode.Parent != nil {
+				lastNode.Parent.RemoveChild(lastNode)
+			}
+			node.AppendChild(lastNode)
+			// Step 13.10.
+			lastNode = node
+		}
+
+		// Step 14. Reparent lastNode to the common ancestor,
+		// or for misnested table nodes, to the foster parent.
+		if lastNode.Parent != nil {
+			lastNode.Parent.RemoveChild(lastNode)
+		}
+		switch commonAncestor.DataAtom {
+		case a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+			p.fosterParent(lastNode)
+		default:
+			commonAncestor.AppendChild(lastNode)
+		}
+
+		// Steps 15-17. Reparent nodes from the furthest block's children
+		// to a clone of the formatting element.
+		clone := formattingElement.clone()
+		reparentChildren(clone, furthestBlock)
+		furthestBlock.AppendChild(clone)
+
+		// Step 18. Fix up the list of active formatting elements.
+		if oldLoc := p.afe.index(formattingElement); oldLoc != -1 && oldLoc < bookmark {
+			// Move the bookmark with the rest of the list.
+			bookmark--
+		}
+		p.afe.remove(formattingElement)
+		p.afe.insert(bookmark, clone)
+
+		// Step 19. Fix up the stack of open elements.
+		p.oe.remove(formattingElement)
+		p.oe.insert(p.oe.index(furthestBlock)+1, clone)
+	}
+}
+
+// inBodyEndTagOther performs the "any other end tag" algorithm for inBodyIM.
+// "Any other end tag" handling from 12.2.6.5 The rules for parsing tokens in foreign content
+// https://html.spec.whatwg.org/multipage/syntax.html#parsing-main-inforeign
+func (p *parser) inBodyEndTagOther(tagAtom a.Atom, tagName string) {
+	for i := len(p.oe) - 1; i >= 0; i-- {
+		// Two element nodes have the same tag if they have the same Data (a
+		// string-typed field). As an optimization, for common HTML tags, each
+		// Data string is assigned a unique, non-zero DataAtom (a uint32-typed
+		// field), since integer comparison is faster than string comparison.
+		// Uncommon (custom) tags get a zero DataAtom.
+		//
+		// The if condition here is equivalent to (p.oe[i].Data == tagName).
+		if (p.oe[i].DataAtom == tagAtom) &&
+			((tagAtom != 0) || (p.oe[i].Data == tagName)) {
+			p.oe = p.oe[:i]
+			break
+		}
+		if isSpecialElement(p.oe[i]) {
+			break
+		}
+	}
+}
+
+// Section 12.2.6.4.8.
+func textIM(p *parser) bool {
+	switch p.tok.Type {
+	case ErrorToken:
+		p.oe.pop()
+	case TextToken:
+		d := p.tok.Data
+		if n := p.oe.top(); n.DataAtom == a.Textarea && n.FirstChild == nil {
+			// Ignore a newline at the start of a <textarea> block.
+			if d != "" && d[0] == '\r' {
+				d = d[1:]
+			}
+			if d != "" && d[0] == '\n' {
+				d = d[1:]
+			}
+		}
+		if d == "" {
+			return true
+		}
+		p.addText(d)
+		return true
+	case EndTagToken:
+		p.oe.pop()
+	}
+	p.im = p.originalIM
+	p.originalIM = nil
+	return p.tok.Type == EndTagToken
+}
+
+// Section 12.2.6.4.9.
+func inTableIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		p.tok.Data = strings.Replace(p.tok.Data, "\x00", "", -1)
+		switch p.oe.top().DataAtom {
+		case a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+			if strings.Trim(p.tok.Data, whitespace) == "" {
+				p.addText(p.tok.Data)
+				return true
+			}
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Caption:
+			p.clearStackToContext(tableScope)
+			p.afe = append(p.afe, &scopeMarker)
+			p.addElement()
+			p.im = inCaptionIM
+			return true
+		case a.Colgroup:
+			p.clearStackToContext(tableScope)
+			p.addElement()
+			p.im = inColumnGroupIM
+			return true
+		case a.Col:
+			p.parseImpliedToken(StartTagToken, a.Colgroup, a.Colgroup.String())
+			return false
+		case a.Tbody, a.Tfoot, a.Thead:
+			p.clearStackToContext(tableScope)
+			p.addElement()
+			p.im = inTableBodyIM
+			return true
+		case a.Td, a.Th, a.Tr:
+			p.parseImpliedToken(StartTagToken, a.Tbody, a.Tbody.String())
+			return false
+		case a.Table:
+			if p.popUntil(tableScope, a.Table) {
+				p.resetInsertionMode()
+				return false
+			}
+			// Ignore the token.
+			return true
+		case a.Style, a.Script, a.Template:
+			return inHeadIM(p)
+		case a.Input:
+			for _, t := range p.tok.Attr {
+				if t.Key == "type" && strings.ToLower(t.Val) == "hidden" {
+					p.addElement()
+					p.oe.pop()
+					return true
+				}
+			}
+			// Otherwise drop down to the default action.
+		case a.Form:
+			if p.oe.contains(a.Template) || p.form != nil {
+				// Ignore the token.
+				return true
+			}
+			p.addElement()
+			p.form = p.oe.pop()
+		case a.Select:
+			p.reconstructActiveFormattingElements()
+			switch p.top().DataAtom {
+			case a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+				p.fosterParenting = true
+			}
+			p.addElement()
+			p.fosterParenting = false
+			p.framesetOK = false
+			p.im = inSelectInTableIM
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Table:
+			if p.popUntil(tableScope, a.Table) {
+				p.resetInsertionMode()
+				return true
+			}
+			// Ignore the token.
+			return true
+		case a.Body, a.Caption, a.Col, a.Colgroup, a.Html, a.Tbody, a.Td, a.Tfoot, a.Th, a.Thead, a.Tr:
+			// Ignore the token.
+			return true
+		case a.Template:
+			return inHeadIM(p)
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	case ErrorToken:
+		return inBodyIM(p)
+	}
+
+	p.fosterParenting = true
+	defer func() { p.fosterParenting = false }()
+
+	return inBodyIM(p)
+}
+
+// Section 12.2.6.4.11.
+func inCaptionIM(p *parser) bool {
+	switch p.tok.Type {
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Caption, a.Col, a.Colgroup, a.Tbody, a.Td, a.Tfoot, a.Thead, a.Tr:
+			if p.popUntil(tableScope, a.Caption) {
+				p.clearActiveFormattingElements()
+				p.im = inTableIM
+				return false
+			} else {
+				// Ignore the token.
+				return true
+			}
+		case a.Select:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.framesetOK = false
+			p.im = inSelectInTableIM
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Caption:
+			if p.popUntil(tableScope, a.Caption) {
+				p.clearActiveFormattingElements()
+				p.im = inTableIM
+			}
+			return true
+		case a.Table:
+			if p.popUntil(tableScope, a.Caption) {
+				p.clearActiveFormattingElements()
+				p.im = inTableIM
+				return false
+			} else {
+				// Ignore the token.
+				return true
+			}
+		case a.Body, a.Col, a.Colgroup, a.Html, a.Tbody, a.Td, a.Tfoot, a.Th, a.Thead, a.Tr:
+			// Ignore the token.
+			return true
+		}
+	}
+	return inBodyIM(p)
+}
+
+// Section 12.2.6.4.12.
+func inColumnGroupIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) < len(p.tok.Data) {
+			// Add the initial whitespace to the current node.
+			p.addText(p.tok.Data[:len(p.tok.Data)-len(s)])
+			if s == "" {
+				return true
+			}
+			p.tok.Data = s
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Col:
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+			return true
+		case a.Template:
+			return inHeadIM(p)
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Colgroup:
+			if p.oe.top().DataAtom == a.Colgroup {
+				p.oe.pop()
+				p.im = inTableIM
+			}
+			return true
+		case a.Col:
+			// Ignore the token.
+			return true
+		case a.Template:
+			return inHeadIM(p)
+		}
+	case ErrorToken:
+		return inBodyIM(p)
+	}
+	if p.oe.top().DataAtom != a.Colgroup {
+		return true
+	}
+	p.oe.pop()
+	p.im = inTableIM
+	return false
+}
+
+// Section 12.2.6.4.13.
+func inTableBodyIM(p *parser) bool {
+	switch p.tok.Type {
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Tr:
+			p.clearStackToContext(tableBodyScope)
+			p.addElement()
+			p.im = inRowIM
+			return true
+		case a.Td, a.Th:
+			p.parseImpliedToken(StartTagToken, a.Tr, a.Tr.String())
+			return false
+		case a.Caption, a.Col, a.Colgroup, a.Tbody, a.Tfoot, a.Thead:
+			if p.popUntil(tableScope, a.Tbody, a.Thead, a.Tfoot) {
+				p.im = inTableIM
+				return false
+			}
+			// Ignore the token.
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Tbody, a.Tfoot, a.Thead:
+			if p.elementInScope(tableScope, p.tok.DataAtom) {
+				p.clearStackToContext(tableBodyScope)
+				p.oe.pop()
+				p.im = inTableIM
+			}
+			return true
+		case a.Table:
+			if p.popUntil(tableScope, a.Tbody, a.Thead, a.Tfoot) {
+				p.im = inTableIM
+				return false
+			}
+			// Ignore the token.
+			return true
+		case a.Body, a.Caption, a.Col, a.Colgroup, a.Html, a.Td, a.Th, a.Tr:
+			// Ignore the token.
+			return true
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	}
+
+	return inTableIM(p)
+}
+
+// Section 12.2.6.4.14.
+func inRowIM(p *parser) bool {
+	switch p.tok.Type {
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Td, a.Th:
+			p.clearStackToContext(tableRowScope)
+			p.addElement()
+			p.afe = append(p.afe, &scopeMarker)
+			p.im = inCellIM
+			return true
+		case a.Caption, a.Col, a.Colgroup, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+			if p.popUntil(tableScope, a.Tr) {
+				p.im = inTableBodyIM
+				return false
+			}
+			// Ignore the token.
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Tr:
+			if p.popUntil(tableScope, a.Tr) {
+				p.im = inTableBodyIM
+				return true
+			}
+			// Ignore the token.
+			return true
+		case a.Table:
+			if p.popUntil(tableScope, a.Tr) {
+				p.im = inTableBodyIM
+				return false
+			}
+			// Ignore the token.
+			return true
+		case a.Tbody, a.Tfoot, a.Thead:
+			if p.elementInScope(tableScope, p.tok.DataAtom) {
+				p.parseImpliedToken(EndTagToken, a.Tr, a.Tr.String())
+				return false
+			}
+			// Ignore the token.
+			return true
+		case a.Body, a.Caption, a.Col, a.Colgroup, a.Html, a.Td, a.Th:
+			// Ignore the token.
+			return true
+		}
+	}
+
+	return inTableIM(p)
+}
+
+// Section 12.2.6.4.15.
+func inCellIM(p *parser) bool {
+	switch p.tok.Type {
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Caption, a.Col, a.Colgroup, a.Tbody, a.Td, a.Tfoot, a.Th, a.Thead, a.Tr:
+			if p.popUntil(tableScope, a.Td, a.Th) {
+				// Close the cell and reprocess.
+				p.clearActiveFormattingElements()
+				p.im = inRowIM
+				return false
+			}
+			// Ignore the token.
+			return true
+		case a.Select:
+			p.reconstructActiveFormattingElements()
+			p.addElement()
+			p.framesetOK = false
+			p.im = inSelectInTableIM
+			return true
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Td, a.Th:
+			if !p.popUntil(tableScope, p.tok.DataAtom) {
+				// Ignore the token.
+				return true
+			}
+			p.clearActiveFormattingElements()
+			p.im = inRowIM
+			return true
+		case a.Body, a.Caption, a.Col, a.Colgroup, a.Html:
+			// Ignore the token.
+			return true
+		case a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr:
+			if !p.elementInScope(tableScope, p.tok.DataAtom) {
+				// Ignore the token.
+				return true
+			}
+			// Close the cell and reprocess.
+			if p.popUntil(tableScope, a.Td, a.Th) {
+				p.clearActiveFormattingElements()
+			}
+			p.im = inRowIM
+			return false
+		}
+	}
+	return inBodyIM(p)
+}
+
+// Section 12.2.6.4.16.
+func inSelectIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		p.addText(strings.Replace(p.tok.Data, "\x00", "", -1))
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Option:
+			if p.top().DataAtom == a.Option {
+				p.oe.pop()
+			}
+			p.addElement()
+		case a.Optgroup:
+			if p.top().DataAtom == a.Option {
+				p.oe.pop()
+			}
+			if p.top().DataAtom == a.Optgroup {
+				p.oe.pop()
+			}
+			p.addElement()
+		case a.Select:
+			if p.popUntil(selectScope, a.Select) {
+				p.resetInsertionMode()
+			} else {
+				// Ignore the token.
+				return true
+			}
+		case a.Input, a.Keygen, a.Textarea:
+			if p.elementInScope(selectScope, a.Select) {
+				p.parseImpliedToken(EndTagToken, a.Select, a.Select.String())
+				return false
+			}
+			// In order to properly ignore <textarea>, we need to change the tokenizer mode.
+			p.tokenizer.NextIsNotRawText()
+			// Ignore the token.
+			return true
+		case a.Script, a.Template:
+			return inHeadIM(p)
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Option:
+			if p.top().DataAtom == a.Option {
+				p.oe.pop()
+			}
+		case a.Optgroup:
+			i := len(p.oe) - 1
+			if p.oe[i].DataAtom == a.Option {
+				i--
+			}
+			if p.oe[i].DataAtom == a.Optgroup {
+				p.oe = p.oe[:i]
+			}
+		case a.Select:
+			if p.popUntil(selectScope, a.Select) {
+				p.resetInsertionMode()
+			} else {
+				// Ignore the token.
+				return true
+			}
+		case a.Template:
+			return inHeadIM(p)
+		}
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case DoctypeToken:
+		// Ignore the token.
+		return true
+	case ErrorToken:
+		return inBodyIM(p)
+	}
+
+	return true
+}
+
+// Section 12.2.6.4.17.
+func inSelectInTableIM(p *parser) bool {
+	switch p.tok.Type {
+	case StartTagToken, EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Caption, a.Table, a.Tbody, a.Tfoot, a.Thead, a.Tr, a.Td, a.Th:
+			if p.tok.Type == EndTagToken && !p.elementInScope(tableScope, p.tok.DataAtom) {
+				// Ignore the token.
+				return true
+			}
+			// This is like p.popUntil(selectScope, a.Select), but it also
+			// matches <math select>, not just <select>. Matching the MathML
+			// tag is arguably incorrect (conceptually), but it mimics what
+			// Chromium does.
+			for i := len(p.oe) - 1; i >= 0; i-- {
+				if n := p.oe[i]; n.DataAtom == a.Select {
+					p.oe = p.oe[:i]
+					break
+				}
+			}
+			p.resetInsertionMode()
+			return false
+		}
+	}
+	return inSelectIM(p)
+}
+
+// Section 12.2.6.4.18.
+func inTemplateIM(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken, CommentToken, DoctypeToken:
+		return inBodyIM(p)
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Base, a.Basefont, a.Bgsound, a.Link, a.Meta, a.Noframes, a.Script, a.Style, a.Template, a.Title:
+			return inHeadIM(p)
+		case a.Caption, a.Colgroup, a.Tbody, a.Tfoot, a.Thead:
+			p.templateStack.pop()
+			p.templateStack = append(p.templateStack, inTableIM)
+			p.im = inTableIM
+			return false
+		case a.Col:
+			p.templateStack.pop()
+			p.templateStack = append(p.templateStack, inColumnGroupIM)
+			p.im = inColumnGroupIM
+			return false
+		case a.Tr:
+			p.templateStack.pop()
+			p.templateStack = append(p.templateStack, inTableBodyIM)
+			p.im = inTableBodyIM
+			return false
+		case a.Td, a.Th:
+			p.templateStack.pop()
+			p.templateStack = append(p.templateStack, inRowIM)
+			p.im = inRowIM
+			return false
+		default:
+			p.templateStack.pop()
+			p.templateStack = append(p.templateStack, inBodyIM)
+			p.im = inBodyIM
+			return false
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Template:
+			return inHeadIM(p)
+		default:
+			// Ignore the token.
+			return true
+		}
+	case ErrorToken:
+		if !p.oe.contains(a.Template) {
+			// Ignore the token.
+			return true
+		}
+		// TODO: remove this divergence from the HTML5 spec.
+		//
+		// See https://bugs.chromium.org/p/chromium/issues/detail?id=829668
+		p.generateImpliedEndTags()
+		for i := len(p.oe) - 1; i >= 0; i-- {
+			if n := p.oe[i]; n.Namespace == "" && n.DataAtom == a.Template {
+				p.oe = p.oe[:i]
+				break
+			}
+		}
+		p.clearActiveFormattingElements()
+		p.templateStack.pop()
+		p.resetInsertionMode()
+		return false
+	}
+	return false
+}
+
+// Section 12.2.6.4.19.
+func afterBodyIM(p *parser) bool {
+	switch p.tok.Type {
+	case ErrorToken:
+		// Stop parsing.
+		return true
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) == 0 {
+			// It was all whitespace.
+			return inBodyIM(p)
+		}
+	case StartTagToken:
+		if p.tok.DataAtom == a.Html {
+			return inBodyIM(p)
+		}
+	case EndTagToken:
+		if p.tok.DataAtom == a.Html {
+			if !p.fragment {
+				p.im = afterAfterBodyIM
+			}
+			return true
+		}
+	case CommentToken:
+		// The comment is attached to the <html> element.
+		if len(p.oe) < 1 || p.oe[0].DataAtom != a.Html {
+			panic("html: bad parser state: <html> element not found, in the after-body insertion mode")
+		}
+		p.oe[0].AppendChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	}
+	p.im = inBodyIM
+	return false
+}
+
+// Section 12.2.6.4.20.
+func inFramesetIM(p *parser) bool {
+	switch p.tok.Type {
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case TextToken:
+		// Ignore all text but whitespace.
+		s := strings.Map(func(c rune) rune {
+			switch c {
+			case ' ', '\t', '\n', '\f', '\r':
+				return c
+			}
+			return -1
+		}, p.tok.Data)
+		if s != "" {
+			p.addText(s)
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Frameset:
+			p.addElement()
+		case a.Frame:
+			p.addElement()
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+		case a.Noframes:
+			return inHeadIM(p)
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Frameset:
+			if p.oe.top().DataAtom != a.Html {
+				p.oe.pop()
+				if p.oe.top().DataAtom != a.Frameset {
+					p.im = afterFramesetIM
+					return true
+				}
+			}
+		}
+	default:
+		// Ignore the token.
+	}
+	return true
+}
+
+// Section 12.2.6.4.21.
+func afterFramesetIM(p *parser) bool {
+	switch p.tok.Type {
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case TextToken:
+		// Ignore all text but whitespace.
+		s := strings.Map(func(c rune) rune {
+			switch c {
+			case ' ', '\t', '\n', '\f', '\r':
+				return c
+			}
+			return -1
+		}, p.tok.Data)
+		if s != "" {
+			p.addText(s)
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Noframes:
+			return inHeadIM(p)
+		}
+	case EndTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			p.im = afterAfterFramesetIM
+			return true
+		}
+	default:
+		// Ignore the token.
+	}
+	return true
+}
+
+// Section 12.2.6.4.22.
+func afterAfterBodyIM(p *parser) bool {
+	switch p.tok.Type {
+	case ErrorToken:
+		// Stop parsing.
+		return true
+	case TextToken:
+		s := strings.TrimLeft(p.tok.Data, whitespace)
+		if len(s) == 0 {
+			// It was all whitespace.
+			return inBodyIM(p)
+		}
+	case StartTagToken:
+		if p.tok.DataAtom == a.Html {
+			return inBodyIM(p)
+		}
+	case CommentToken:
+		p.doc.AppendChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+		return true
+	case DoctypeToken:
+		return inBodyIM(p)
+	}
+	p.im = inBodyIM
+	return false
+}
+
+// Section 12.2.6.4.23.
+func afterAfterFramesetIM(p *parser) bool {
+	switch p.tok.Type {
+	case CommentToken:
+		p.doc.AppendChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case TextToken:
+		// Ignore all text but whitespace.
+		s := strings.Map(func(c rune) rune {
+			switch c {
+			case ' ', '\t', '\n', '\f', '\r':
+				return c
+			}
+			return -1
+		}, p.tok.Data)
+		if s != "" {
+			p.tok.Data = s
+			return inBodyIM(p)
+		}
+	case StartTagToken:
+		switch p.tok.DataAtom {
+		case a.Html:
+			return inBodyIM(p)
+		case a.Noframes:
+			return inHeadIM(p)
+		}
+	case DoctypeToken:
+		return inBodyIM(p)
+	default:
+		// Ignore the token.
+	}
+	return true
+}
+
+const whitespaceOrNUL = whitespace + "\x00"
+
+// Section 12.2.6.5
+func parseForeignContent(p *parser) bool {
+	switch p.tok.Type {
+	case TextToken:
+		if p.framesetOK {
+			p.framesetOK = strings.TrimLeft(p.tok.Data, whitespaceOrNUL) == ""
+		}
+		p.tok.Data = strings.Replace(p.tok.Data, "\x00", "\ufffd", -1)
+		p.addText(p.tok.Data)
+	case CommentToken:
+		p.addChild(&Node{
+			Type: CommentNode,
+			Data: p.tok.Data,
+		})
+	case StartTagToken:
+		b := breakout[p.tok.Data]
+		if p.tok.DataAtom == a.Font {
+		loop:
+			for _, attr := range p.tok.Attr {
+				switch attr.Key {
+				case "color", "face", "size":
+					b = true
+					break loop
+				}
+			}
+		}
+		if b {
+			for i := len(p.oe) - 1; i >= 0; i-- {
+				n := p.oe[i]
+				if n.Namespace == "" || htmlIntegrationPoint(n) || mathMLTextIntegrationPoint(n) {
+					p.oe = p.oe[:i+1]
+					break
+				}
+			}
+			return false
+		}
+		switch p.top().Namespace {
+		case "math":
+			adjustAttributeNames(p.tok.Attr, mathMLAttributeAdjustments)
+		case "svg":
+			// Adjust SVG tag names. The tokenizer lower-cases tag names, but
+			// SVG wants e.g. "foreignObject" with a capital second "O".
+			if x := svgTagNameAdjustments[p.tok.Data]; x != "" {
+				p.tok.DataAtom = a.Lookup([]byte(x))
+				p.tok.Data = x
+			}
+			adjustAttributeNames(p.tok.Attr, svgAttributeAdjustments)
+		default:
+			panic("html: bad parser state: unexpected namespace")
+		}
+		adjustForeignAttributes(p.tok.Attr)
+		namespace := p.top().Namespace
+		p.addElement()
+		p.top().Namespace = namespace
+		if namespace != "" {
+			// Don't let the tokenizer go into raw text mode in foreign content
+			// (e.g. in an SVG <title> tag).
+			p.tokenizer.NextIsNotRawText()
+		}
+		if p.hasSelfClosingToken {
+			p.oe.pop()
+			p.acknowledgeSelfClosingTag()
+		}
+	case EndTagToken:
+		for i := len(p.oe) - 1; i >= 0; i-- {
+			if p.oe[i].Namespace == "" {
+				return p.im(p)
+			}
+			if strings.EqualFold(p.oe[i].Data, p.tok.Data) {
+				p.oe = p.oe[:i]
+				break
+			}
+		}
+		return true
+	default:
+		// Ignore the token.
+	}
+	return true
+}
+
+// Section 12.2.6.
+func (p *parser) inForeignContent() bool {
+	if len(p.oe) == 0 {
+		return false
+	}
+	n := p.oe[len(p.oe)-1]
+	if n.Namespace == "" {
+		return false
+	}
+	if mathMLTextIntegrationPoint(n) {
+		if p.tok.Type == StartTagToken && p.tok.DataAtom != a.Mglyph && p.tok.DataAtom != a.Malignmark {
+			return false
+		}
+		if p.tok.Type == TextToken {
+			return false
+		}
+	}
+	if n.Namespace == "math" && n.DataAtom == a.AnnotationXml && p.tok.Type == StartTagToken && p.tok.DataAtom == a.Svg {
+		return false
+	}
+	if htmlIntegrationPoint(n) && (p.tok.Type == StartTagToken || p.tok.Type == TextToken) {
+		return false
+	}
+	if p.tok.Type == ErrorToken {
+		return false
+	}
+	return true
+}
+
+// parseImpliedToken parses a token as though it had appeared in the parser's
+// input.
+func (p *parser) parseImpliedToken(t TokenType, dataAtom a.Atom, data string) {
+	realToken, selfClosing := p.tok, p.hasSelfClosingToken
+	p.tok = Token{
+		Type:     t,
+		DataAtom: dataAtom,
+		Data:     data,
+	}
+	p.hasSelfClosingToken = false
+	p.parseCurrentToken()
+	p.tok, p.hasSelfClosingToken = realToken, selfClosing
+}
+
+// parseCurrentToken runs the current token through the parsing routines
+// until it is consumed.
+func (p *parser) parseCurrentToken() {
+	if p.tok.Type == SelfClosingTagToken {
+		p.hasSelfClosingToken = true
+		p.tok.Type = StartTagToken
+	}
+
+	consumed := false
+	for !consumed {
+		if p.inForeignContent() {
+			consumed = parseForeignContent(p)
+		} else {
+			consumed = p.im(p)
+		}
+	}
+
+	if p.hasSelfClosingToken {
+		// This is a parse error, but ignore it.
+		p.hasSelfClosingToken = false
+	}
+}
+
+func (p *parser) parse() error {
+	// Iterate until EOF. Any other error will cause an early return.
+	var err error
+	for err != io.EOF {
+		// CDATA sections are allowed only in foreign content.
+		n := p.oe.top()
+		p.tokenizer.AllowCDATA(n != nil && n.Namespace != "")
+		// Read and parse the next token.
+		p.tokenizer.Next()
+		p.tok = p.tokenizer.Token()
+		if p.tok.Type == ErrorToken {
+			err = p.tokenizer.Err()
+			if err != nil && err != io.EOF {
+				return err
+			}
+		}
+		p.parseCurrentToken()
+	}
+	return nil
+}
+
+// Parse returns the parse tree for the HTML from the given Reader.
+//
+// It implements the HTML5 parsing algorithm
+// (https://html.spec.whatwg.org/multipage/syntax.html#tree-construction),
+// which is very complicated. The resultant tree can contain implicitly created
+// nodes that have no explicit <tag> listed in r's data, and nodes' parents can
+// differ from the nesting implied by a naive processing of start and end
+// <tag>s. Conversely, explicit <tag>s in r's data can be silently dropped,
+// with no corresponding node in the resulting tree.
+//
+// The input is assumed to be UTF-8 encoded.
+func Parse(r io.Reader) (*Node, error) {
+	return ParseWithOptions(r)
+}
+
+// ParseFragment parses a fragment of HTML and returns the nodes that were
+// found. If the fragment is the InnerHTML for an existing element, pass that
+// element in context.
+//
+// It has the same intricacies as Parse.
+func ParseFragment(r io.Reader, context *Node) ([]*Node, error) {
+	return ParseFragmentWithOptions(r, context)
+}
+
+// ParseOption configures a parser.
+type ParseOption func(p *parser)
+
+// ParseOptionEnableScripting configures the scripting flag.
+// https://html.spec.whatwg.org/multipage/webappapis.html#enabling-and-disabling-scripting
+//
+// By default, scripting is enabled.
+func ParseOptionEnableScripting(enable bool) ParseOption {
+	return func(p *parser) {
+		p.scripting = enable
+	}
+}
+
+// ParseWithOptions is like Parse, with options.
+func ParseWithOptions(r io.Reader, opts ...ParseOption) (*Node, error) {
+	p := &parser{
+		tokenizer: NewTokenizer(r),
+		doc: &Node{
+			Type: DocumentNode,
+		},
+		scripting:  true,
+		framesetOK: true,
+		im:         initialIM,
+	}
+
+	for _, f := range opts {
+		f(p)
+	}
+
+	err := p.parse()
+	if err != nil {
+		return nil, err
+	}
+	return p.doc, nil
+}
+
+// ParseFragmentWithOptions is like ParseFragment, with options.
+func ParseFragmentWithOptions(r io.Reader, context *Node, opts ...ParseOption) ([]*Node, error) {
+	contextTag := ""
+	if context != nil {
+		if context.Type != ElementNode {
+			return nil, errors.New("html: ParseFragment of non-element Node")
+		}
+		// The next check isn't just context.DataAtom.String() == context.Data because
+		// it is valid to pass an element whose tag isn't a known atom. For example,
+		// DataAtom == 0 and Data = "tagfromthefuture" is perfectly consistent.
+		if context.DataAtom != a.Lookup([]byte(context.Data)) {
+			return nil, fmt.Errorf("html: inconsistent Node: DataAtom=%q, Data=%q", context.DataAtom, context.Data)
+		}
+		contextTag = context.DataAtom.String()
+	}
+	p := &parser{
+		tokenizer: NewTokenizerFragment(r, contextTag),
+		doc: &Node{
+			Type: DocumentNode,
+		},
+		scripting: true,
+		fragment:  true,
+		context:   context,
+	}
+
+	for _, f := range opts {
+		f(p)
+	}
+
+	root := &Node{
+		Type:     ElementNode,
+		DataAtom: a.Html,
+		Data:     a.Html.String(),
+	}
+	p.doc.AppendChild(root)
+	p.oe = nodeStack{root}
+	if context != nil && context.DataAtom == a.Template {
+		p.templateStack = append(p.templateStack, inTemplateIM)
+	}
+	p.resetInsertionMode()
+
+	for n := context; n != nil; n = n.Parent {
+		if n.Type == ElementNode && n.DataAtom == a.Form {
+			p.form = n
+			break
+		}
+	}
+
+	err := p.parse()
+	if err != nil {
+		return nil, err
+	}
+
+	parent := p.doc
+	if context != nil {
+		parent = root
+	}
+
+	var result []*Node
+	for c := parent.FirstChild; c != nil; {
+		next := c.NextSibling
+		parent.RemoveChild(c)
+		result = append(result, c)
+		c = next
+	}
+	return result, nil
+}
diff --git a/backend/vendor/golang.org/x/net/html/render.go b/backend/vendor/golang.org/x/net/html/render.go
new file mode 100644
index 00000000..d34564f4
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/render.go
@@ -0,0 +1,271 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"bufio"
+	"errors"
+	"fmt"
+	"io"
+	"strings"
+)
+
+type writer interface {
+	io.Writer
+	io.ByteWriter
+	WriteString(string) (int, error)
+}
+
+// Render renders the parse tree n to the given writer.
+//
+// Rendering is done on a 'best effort' basis: calling Parse on the output of
+// Render will always result in something similar to the original tree, but it
+// is not necessarily an exact clone unless the original tree was 'well-formed'.
+// 'Well-formed' is not easily specified; the HTML5 specification is
+// complicated.
+//
+// Calling Parse on arbitrary input typically results in a 'well-formed' parse
+// tree. However, it is possible for Parse to yield a 'badly-formed' parse tree.
+// For example, in a 'well-formed' parse tree, no <a> element is a child of
+// another <a> element: parsing "<a><a>" results in two sibling elements.
+// Similarly, in a 'well-formed' parse tree, no <a> element is a child of a
+// <table> element: parsing "<p><table><a>" results in a <p> with two sibling
+// children; the <a> is reparented to the <table>'s parent. However, calling
+// Parse on "<a><table><a>" does not return an error, but the result has an <a>
+// element with an <a> child, and is therefore not 'well-formed'.
+//
+// Programmatically constructed trees are typically also 'well-formed', but it
+// is possible to construct a tree that looks innocuous but, when rendered and
+// re-parsed, results in a different tree. A simple example is that a solitary
+// text node would become a tree containing <html>, <head> and <body> elements.
+// Another example is that the programmatic equivalent of "a<head>b</head>c"
+// becomes "<html><head><head/><body>abc</body></html>".
+func Render(w io.Writer, n *Node) error {
+	if x, ok := w.(writer); ok {
+		return render(x, n)
+	}
+	buf := bufio.NewWriter(w)
+	if err := render(buf, n); err != nil {
+		return err
+	}
+	return buf.Flush()
+}
+
+// plaintextAbort is returned from render1 when a <plaintext> element
+// has been rendered. No more end tags should be rendered after that.
+var plaintextAbort = errors.New("html: internal error (plaintext abort)")
+
+func render(w writer, n *Node) error {
+	err := render1(w, n)
+	if err == plaintextAbort {
+		err = nil
+	}
+	return err
+}
+
+func render1(w writer, n *Node) error {
+	// Render non-element nodes; these are the easy cases.
+	switch n.Type {
+	case ErrorNode:
+		return errors.New("html: cannot render an ErrorNode node")
+	case TextNode:
+		return escape(w, n.Data)
+	case DocumentNode:
+		for c := n.FirstChild; c != nil; c = c.NextSibling {
+			if err := render1(w, c); err != nil {
+				return err
+			}
+		}
+		return nil
+	case ElementNode:
+		// No-op.
+	case CommentNode:
+		if _, err := w.WriteString("<!--"); err != nil {
+			return err
+		}
+		if _, err := w.WriteString(n.Data); err != nil {
+			return err
+		}
+		if _, err := w.WriteString("-->"); err != nil {
+			return err
+		}
+		return nil
+	case DoctypeNode:
+		if _, err := w.WriteString("<!DOCTYPE "); err != nil {
+			return err
+		}
+		if _, err := w.WriteString(n.Data); err != nil {
+			return err
+		}
+		if n.Attr != nil {
+			var p, s string
+			for _, a := range n.Attr {
+				switch a.Key {
+				case "public":
+					p = a.Val
+				case "system":
+					s = a.Val
+				}
+			}
+			if p != "" {
+				if _, err := w.WriteString(" PUBLIC "); err != nil {
+					return err
+				}
+				if err := writeQuoted(w, p); err != nil {
+					return err
+				}
+				if s != "" {
+					if err := w.WriteByte(' '); err != nil {
+						return err
+					}
+					if err := writeQuoted(w, s); err != nil {
+						return err
+					}
+				}
+			} else if s != "" {
+				if _, err := w.WriteString(" SYSTEM "); err != nil {
+					return err
+				}
+				if err := writeQuoted(w, s); err != nil {
+					return err
+				}
+			}
+		}
+		return w.WriteByte('>')
+	default:
+		return errors.New("html: unknown node type")
+	}
+
+	// Render the <xxx> opening tag.
+	if err := w.WriteByte('<'); err != nil {
+		return err
+	}
+	if _, err := w.WriteString(n.Data); err != nil {
+		return err
+	}
+	for _, a := range n.Attr {
+		if err := w.WriteByte(' '); err != nil {
+			return err
+		}
+		if a.Namespace != "" {
+			if _, err := w.WriteString(a.Namespace); err != nil {
+				return err
+			}
+			if err := w.WriteByte(':'); err != nil {
+				return err
+			}
+		}
+		if _, err := w.WriteString(a.Key); err != nil {
+			return err
+		}
+		if _, err := w.WriteString(`="`); err != nil {
+			return err
+		}
+		if err := escape(w, a.Val); err != nil {
+			return err
+		}
+		if err := w.WriteByte('"'); err != nil {
+			return err
+		}
+	}
+	if voidElements[n.Data] {
+		if n.FirstChild != nil {
+			return fmt.Errorf("html: void element <%s> has child nodes", n.Data)
+		}
+		_, err := w.WriteString("/>")
+		return err
+	}
+	if err := w.WriteByte('>'); err != nil {
+		return err
+	}
+
+	// Add initial newline where there is danger of a newline beging ignored.
+	if c := n.FirstChild; c != nil && c.Type == TextNode && strings.HasPrefix(c.Data, "\n") {
+		switch n.Data {
+		case "pre", "listing", "textarea":
+			if err := w.WriteByte('\n'); err != nil {
+				return err
+			}
+		}
+	}
+
+	// Render any child nodes.
+	switch n.Data {
+	case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "xmp":
+		for c := n.FirstChild; c != nil; c = c.NextSibling {
+			if c.Type == TextNode {
+				if _, err := w.WriteString(c.Data); err != nil {
+					return err
+				}
+			} else {
+				if err := render1(w, c); err != nil {
+					return err
+				}
+			}
+		}
+		if n.Data == "plaintext" {
+			// Don't render anything else. <plaintext> must be the
+			// last element in the file, with no closing tag.
+			return plaintextAbort
+		}
+	default:
+		for c := n.FirstChild; c != nil; c = c.NextSibling {
+			if err := render1(w, c); err != nil {
+				return err
+			}
+		}
+	}
+
+	// Render the </xxx> closing tag.
+	if _, err := w.WriteString("</"); err != nil {
+		return err
+	}
+	if _, err := w.WriteString(n.Data); err != nil {
+		return err
+	}
+	return w.WriteByte('>')
+}
+
+// writeQuoted writes s to w surrounded by quotes. Normally it will use double
+// quotes, but if s contains a double quote, it will use single quotes.
+// It is used for writing the identifiers in a doctype declaration.
+// In valid HTML, they can't contain both types of quotes.
+func writeQuoted(w writer, s string) error {
+	var q byte = '"'
+	if strings.Contains(s, `"`) {
+		q = '\''
+	}
+	if err := w.WriteByte(q); err != nil {
+		return err
+	}
+	if _, err := w.WriteString(s); err != nil {
+		return err
+	}
+	if err := w.WriteByte(q); err != nil {
+		return err
+	}
+	return nil
+}
+
+// Section 12.1.2, "Elements", gives this list of void elements. Void elements
+// are those that can't have any contents.
+var voidElements = map[string]bool{
+	"area":    true,
+	"base":    true,
+	"br":      true,
+	"col":     true,
+	"command": true,
+	"embed":   true,
+	"hr":      true,
+	"img":     true,
+	"input":   true,
+	"keygen":  true,
+	"link":    true,
+	"meta":    true,
+	"param":   true,
+	"source":  true,
+	"track":   true,
+	"wbr":     true,
+}
diff --git a/backend/vendor/golang.org/x/net/html/token.go b/backend/vendor/golang.org/x/net/html/token.go
new file mode 100644
index 00000000..e3c01d7c
--- /dev/null
+++ b/backend/vendor/golang.org/x/net/html/token.go
@@ -0,0 +1,1219 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package html
+
+import (
+	"bytes"
+	"errors"
+	"io"
+	"strconv"
+	"strings"
+
+	"golang.org/x/net/html/atom"
+)
+
+// A TokenType is the type of a Token.
+type TokenType uint32
+
+const (
+	// ErrorToken means that an error occurred during tokenization.
+	ErrorToken TokenType = iota
+	// TextToken means a text node.
+	TextToken
+	// A StartTagToken looks like <a>.
+	StartTagToken
+	// An EndTagToken looks like </a>.
+	EndTagToken
+	// A SelfClosingTagToken tag looks like <br/>.
+	SelfClosingTagToken
+	// A CommentToken looks like <!--x-->.
+	CommentToken
+	// A DoctypeToken looks like <!DOCTYPE x>
+	DoctypeToken
+)
+
+// ErrBufferExceeded means that the buffering limit was exceeded.
+var ErrBufferExceeded = errors.New("max buffer exceeded")
+
+// String returns a string representation of the TokenType.
+func (t TokenType) String() string {
+	switch t {
+	case ErrorToken:
+		return "Error"
+	case TextToken:
+		return "Text"
+	case StartTagToken:
+		return "StartTag"
+	case EndTagToken:
+		return "EndTag"
+	case SelfClosingTagToken:
+		return "SelfClosingTag"
+	case CommentToken:
+		return "Comment"
+	case DoctypeToken:
+		return "Doctype"
+	}
+	return "Invalid(" + strconv.Itoa(int(t)) + ")"
+}
+
+// An Attribute is an attribute namespace-key-value triple. Namespace is
+// non-empty for foreign attributes like xlink, Key is alphabetic (and hence
+// does not contain escapable characters like '&', '<' or '>'), and Val is
+// unescaped (it looks like "a<b" rather than "a&lt;b").
+//
+// Namespace is only used by the parser, not the tokenizer.
+type Attribute struct {
+	Namespace, Key, Val string
+}
+
+// A Token consists of a TokenType and some Data (tag name for start and end
+// tags, content for text, comments and doctypes). A tag Token may also contain
+// a slice of Attributes. Data is unescaped for all Tokens (it looks like "a<b"
+// rather than "a&lt;b"). For tag Tokens, DataAtom is the atom for Data, or
+// zero if Data is not a known tag name.
+type Token struct {
+	Type     TokenType
+	DataAtom atom.Atom
+	Data     string
+	Attr     []Attribute
+}
+
+// tagString returns a string representation of a tag Token's Data and Attr.
+func (t Token) tagString() string {
+	if len(t.Attr) == 0 {
+		return t.Data
+	}
+	buf := bytes.NewBufferString(t.Data)
+	for _, a := range t.Attr {
+		buf.WriteByte(' ')
+		buf.WriteString(a.Key)
+		buf.WriteString(`="`)
+		escape(buf, a.Val)
+		buf.WriteByte('"')
+	}
+	return buf.String()
+}
+
+// String returns a string representation of the Token.
+func (t Token) String() string {
+	switch t.Type {
+	case ErrorToken:
+		return ""
+	case TextToken:
+		return EscapeString(t.Data)
+	case StartTagToken:
+		return "<" + t.tagString() + ">"
+	case EndTagToken:
+		return "</" + t.tagString() + ">"
+	case SelfClosingTagToken:
+		return "<" + t.tagString() + "/>"
+	case CommentToken:
+		return "<!--" + t.Data + "-->"
+	case DoctypeToken:
+		return "<!DOCTYPE " + t.Data + ">"
+	}
+	return "Invalid(" + strconv.Itoa(int(t.Type)) + ")"
+}
+
+// span is a range of bytes in a Tokenizer's buffer. The start is inclusive,
+// the end is exclusive.
+type span struct {
+	start, end int
+}
+
+// A Tokenizer returns a stream of HTML Tokens.
+type Tokenizer struct {
+	// r is the source of the HTML text.
+	r io.Reader
+	// tt is the TokenType of the current token.
+	tt TokenType
+	// err is the first error encountered during tokenization. It is possible
+	// for tt != Error && err != nil to hold: this means that Next returned a
+	// valid token but the subsequent Next call will return an error token.
+	// For example, if the HTML text input was just "plain", then the first
+	// Next call would set z.err to io.EOF but return a TextToken, and all
+	// subsequent Next calls would return an ErrorToken.
+	// err is never reset. Once it becomes non-nil, it stays non-nil.
+	err error
+	// readErr is the error returned by the io.Reader r. It is separate from
+	// err because it is valid for an io.Reader to return (n int, err1 error)
+	// such that n > 0 && err1 != nil, and callers should always process the
+	// n > 0 bytes before considering the error err1.
+	readErr error
+	// buf[raw.start:raw.end] holds the raw bytes of the current token.
+	// buf[raw.end:] is buffered input that will yield future tokens.
+	raw span
+	buf []byte
+	// maxBuf limits the data buffered in buf. A value of 0 means unlimited.
+	maxBuf int
+	// buf[data.start:data.end] holds the raw bytes of the current token's data:
+	// a text token's text, a tag token's tag name, etc.
+	data span
+	// pendingAttr is the attribute key and value currently being tokenized.
+	// When complete, pendingAttr is pushed onto attr. nAttrReturned is
+	// incremented on each call to TagAttr.
+	pendingAttr   [2]span
+	attr          [][2]span
+	nAttrReturned int
+	// rawTag is the "script" in "</script>" that closes the next token. If
+	// non-empty, the subsequent call to Next will return a raw or RCDATA text
+	// token: one that treats "<p>" as text instead of an element.
+	// rawTag's contents are lower-cased.
+	rawTag string
+	// textIsRaw is whether the current text token's data is not escaped.
+	textIsRaw bool
+	// convertNUL is whether NUL bytes in the current token's data should
+	// be converted into \ufffd replacement characters.
+	convertNUL bool
+	// allowCDATA is whether CDATA sections are allowed in the current context.
+	allowCDATA bool
+}
+
+// AllowCDATA sets whether or not the tokenizer recognizes <![CDATA[foo]]> as
+// the text "foo". The default value is false, which means to recognize it as
+// a bogus comment "<!-- [CDATA[foo]] -->" instead.
+//
+// Strictly speaking, an HTML5 compliant tokenizer should allow CDATA if and
+// only if tokenizing foreign content, such as MathML and SVG. However,
+// tracking foreign-contentness is difficult to do purely in the tokenizer,
+// as opposed to the parser, due to HTML integration points: an <svg> element
+// can contain a <foreignObject> that is foreign-to-SVG but not foreign-to-
+// HTML. For strict compliance with the HTML5 tokenization algorithm, it is the
+// responsibility of the user of a tokenizer to call AllowCDATA as appropriate.
+// In practice, if using the tokenizer without caring whether MathML or SVG
+// CDATA is text or comments, such as tokenizing HTML to find all the anchor
+// text, it is acceptable to ignore this responsibility.
+func (z *Tokenizer) AllowCDATA(allowCDATA bool) {
+	z.allowCDATA = allowCDATA
+}
+
+// NextIsNotRawText instructs the tokenizer that the next token should not be
+// considered as 'raw text'. Some elements, such as script and title elements,
+// normally require the next token after the opening tag to be 'raw text' that
+// has no child elements. For example, tokenizing "<title>a<b>c</b>d</title>"
+// yields a start tag token for "<title>", a text token for "a<b>c</b>d", and
+// an end tag token for "</title>". There are no distinct start tag or end tag
+// tokens for the "<b>" and "</b>".
+//
+// This tokenizer implementation will generally look for raw text at the right
+// times. Strictly speaking, an HTML5 compliant tokenizer should not look for
+// raw text if in foreign content: <title> generally needs raw text, but a
+// <title> inside an <svg> does not. Another example is that a <textarea>
+// generally needs raw text, but a <textarea> is not allowed as an immediate
+// child of a <select>; in normal parsing, a <textarea> implies </select>, but
+// one cannot close the implicit element when parsing a <select>'s InnerHTML.
+// Similarly to AllowCDATA, tracking the correct moment to override raw-text-
+// ness is difficult to do purely in the tokenizer, as opposed to the parser.
+// For strict compliance with the HTML5 tokenization algorithm, it is the
+// responsibility of the user of a tokenizer to call NextIsNotRawText as
+// appropriate. In practice, like AllowCDATA, it is acceptable to ignore this
+// responsibility for basic usage.
+//
+// Note that this 'raw text' concept is different from the one offered by the
+// Tokenizer.Raw method.
+func (z *Tokenizer) NextIsNotRawText() {
+	z.rawTag = ""
+}
+
+// Err returns the error associated with the most recent ErrorToken token.
+// This is typically io.EOF, meaning the end of tokenization.
+func (z *Tokenizer) Err() error {
+	if z.tt != ErrorToken {
+		return nil
+	}
+	return z.err
+}
+
+// readByte returns the next byte from the input stream, doing a buffered read
+// from z.r into z.buf if necessary. z.buf[z.raw.start:z.raw.end] remains a contiguous byte
+// slice that holds all the bytes read so far for the current token.
+// It sets z.err if the underlying reader returns an error.
+// Pre-condition: z.err == nil.
+func (z *Tokenizer) readByte() byte {
+	if z.raw.end >= len(z.buf) {
+		// Our buffer is exhausted and we have to read from z.r. Check if the
+		// previous read resulted in an error.
+		if z.readErr != nil {
+			z.err = z.readErr
+			return 0
+		}
+		// We copy z.buf[z.raw.start:z.raw.end] to the beginning of z.buf. If the length
+		// z.raw.end - z.raw.start is more than half the capacity of z.buf, then we
+		// allocate a new buffer before the copy.
+		c := cap(z.buf)
+		d := z.raw.end - z.raw.start
+		var buf1 []byte
+		if 2*d > c {
+			buf1 = make([]byte, d, 2*c)
+		} else {
+			buf1 = z.buf[:d]
+		}
+		copy(buf1, z.buf[z.raw.start:z.raw.end])
+		if x := z.raw.start; x != 0 {
+			// Adjust the data/attr spans to refer to the same contents after the copy.
+			z.data.start -= x
+			z.data.end -= x
+			z.pendingAttr[0].start -= x
+			z.pendingAttr[0].end -= x
+			z.pendingAttr[1].start -= x
+			z.pendingAttr[1].end -= x
+			for i := range z.attr {
+				z.attr[i][0].start -= x
+				z.attr[i][0].end -= x
+				z.attr[i][1].start -= x
+				z.attr[i][1].end -= x
+			}
+		}
+		z.raw.start, z.raw.end, z.buf = 0, d, buf1[:d]
+		// Now that we have copied the live bytes to the start of the buffer,
+		// we read from z.r into the remainder.
+		var n int
+		n, z.readErr = readAtLeastOneByte(z.r, buf1[d:cap(buf1)])
+		if n == 0 {
+			z.err = z.readErr
+			return 0
+		}
+		z.buf = buf1[:d+n]
+	}
+	x := z.buf[z.raw.end]
+	z.raw.end++
+	if z.maxBuf > 0 && z.raw.end-z.raw.start >= z.maxBuf {
+		z.err = ErrBufferExceeded
+		return 0
+	}
+	return x
+}
+
+// Buffered returns a slice containing data buffered but not yet tokenized.
+func (z *Tokenizer) Buffered() []byte {
+	return z.buf[z.raw.end:]
+}
+
+// readAtLeastOneByte wraps an io.Reader so that reading cannot return (0, nil).
+// It returns io.ErrNoProgress if the underlying r.Read method returns (0, nil)
+// too many times in succession.
+func readAtLeastOneByte(r io.Reader, b []byte) (int, error) {
+	for i := 0; i < 100; i++ {
+		n, err := r.Read(b)
+		if n != 0 || err != nil {
+			return n, err
+		}
+	}
+	return 0, io.ErrNoProgress
+}
+
+// skipWhiteSpace skips past any white space.
+func (z *Tokenizer) skipWhiteSpace() {
+	if z.err != nil {
+		return
+	}
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			return
+		}
+		switch c {
+		case ' ', '\n', '\r', '\t', '\f':
+			// No-op.
+		default:
+			z.raw.end--
+			return
+		}
+	}
+}
+
+// readRawOrRCDATA reads until the next "</foo>", where "foo" is z.rawTag and
+// is typically something like "script" or "textarea".
+func (z *Tokenizer) readRawOrRCDATA() {
+	if z.rawTag == "script" {
+		z.readScript()
+		z.textIsRaw = true
+		z.rawTag = ""
+		return
+	}
+loop:
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			break loop
+		}
+		if c != '<' {
+			continue loop
+		}
+		c = z.readByte()
+		if z.err != nil {
+			break loop
+		}
+		if c != '/' {
+			continue loop
+		}
+		if z.readRawEndTag() || z.err != nil {
+			break loop
+		}
+	}
+	z.data.end = z.raw.end
+	// A textarea's or title's RCDATA can contain escaped entities.
+	z.textIsRaw = z.rawTag != "textarea" && z.rawTag != "title"
+	z.rawTag = ""
+}
+
+// readRawEndTag attempts to read a tag like "</foo>", where "foo" is z.rawTag.
+// If it succeeds, it backs up the input position to reconsume the tag and
+// returns true. Otherwise it returns false. The opening "</" has already been
+// consumed.
+func (z *Tokenizer) readRawEndTag() bool {
+	for i := 0; i < len(z.rawTag); i++ {
+		c := z.readByte()
+		if z.err != nil {
+			return false
+		}
+		if c != z.rawTag[i] && c != z.rawTag[i]-('a'-'A') {
+			z.raw.end--
+			return false
+		}
+	}
+	c := z.readByte()
+	if z.err != nil {
+		return false
+	}
+	switch c {
+	case ' ', '\n', '\r', '\t', '\f', '/', '>':
+		// The 3 is 2 for the leading "</" plus 1 for the trailing character c.
+		z.raw.end -= 3 + len(z.rawTag)
+		return true
+	}
+	z.raw.end--
+	return false
+}
+
+// readScript reads until the next </script> tag, following the byzantine
+// rules for escaping/hiding the closing tag.
+func (z *Tokenizer) readScript() {
+	defer func() {
+		z.data.end = z.raw.end
+	}()
+	var c byte
+
+scriptData:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c == '<' {
+		goto scriptDataLessThanSign
+	}
+	goto scriptData
+
+scriptDataLessThanSign:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '/':
+		goto scriptDataEndTagOpen
+	case '!':
+		goto scriptDataEscapeStart
+	}
+	z.raw.end--
+	goto scriptData
+
+scriptDataEndTagOpen:
+	if z.readRawEndTag() || z.err != nil {
+		return
+	}
+	goto scriptData
+
+scriptDataEscapeStart:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c == '-' {
+		goto scriptDataEscapeStartDash
+	}
+	z.raw.end--
+	goto scriptData
+
+scriptDataEscapeStartDash:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c == '-' {
+		goto scriptDataEscapedDashDash
+	}
+	z.raw.end--
+	goto scriptData
+
+scriptDataEscaped:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataEscapedDash
+	case '<':
+		goto scriptDataEscapedLessThanSign
+	}
+	goto scriptDataEscaped
+
+scriptDataEscapedDash:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataEscapedDashDash
+	case '<':
+		goto scriptDataEscapedLessThanSign
+	}
+	goto scriptDataEscaped
+
+scriptDataEscapedDashDash:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataEscapedDashDash
+	case '<':
+		goto scriptDataEscapedLessThanSign
+	case '>':
+		goto scriptData
+	}
+	goto scriptDataEscaped
+
+scriptDataEscapedLessThanSign:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c == '/' {
+		goto scriptDataEscapedEndTagOpen
+	}
+	if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' {
+		goto scriptDataDoubleEscapeStart
+	}
+	z.raw.end--
+	goto scriptData
+
+scriptDataEscapedEndTagOpen:
+	if z.readRawEndTag() || z.err != nil {
+		return
+	}
+	goto scriptDataEscaped
+
+scriptDataDoubleEscapeStart:
+	z.raw.end--
+	for i := 0; i < len("script"); i++ {
+		c = z.readByte()
+		if z.err != nil {
+			return
+		}
+		if c != "script"[i] && c != "SCRIPT"[i] {
+			z.raw.end--
+			goto scriptDataEscaped
+		}
+	}
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case ' ', '\n', '\r', '\t', '\f', '/', '>':
+		goto scriptDataDoubleEscaped
+	}
+	z.raw.end--
+	goto scriptDataEscaped
+
+scriptDataDoubleEscaped:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataDoubleEscapedDash
+	case '<':
+		goto scriptDataDoubleEscapedLessThanSign
+	}
+	goto scriptDataDoubleEscaped
+
+scriptDataDoubleEscapedDash:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataDoubleEscapedDashDash
+	case '<':
+		goto scriptDataDoubleEscapedLessThanSign
+	}
+	goto scriptDataDoubleEscaped
+
+scriptDataDoubleEscapedDashDash:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch c {
+	case '-':
+		goto scriptDataDoubleEscapedDashDash
+	case '<':
+		goto scriptDataDoubleEscapedLessThanSign
+	case '>':
+		goto scriptData
+	}
+	goto scriptDataDoubleEscaped
+
+scriptDataDoubleEscapedLessThanSign:
+	c = z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c == '/' {
+		goto scriptDataDoubleEscapeEnd
+	}
+	z.raw.end--
+	goto scriptDataDoubleEscaped
+
+scriptDataDoubleEscapeEnd:
+	if z.readRawEndTag() {
+		z.raw.end += len("</script>")
+		goto scriptDataEscaped
+	}
+	if z.err != nil {
+		return
+	}
+	goto scriptDataDoubleEscaped
+}
+
+// readComment reads the next comment token starting with "<!--". The opening
+// "<!--" has already been consumed.
+func (z *Tokenizer) readComment() {
+	z.data.start = z.raw.end
+	defer func() {
+		if z.data.end < z.data.start {
+			// It's a comment with no data, like <!-->.
+			z.data.end = z.data.start
+		}
+	}()
+	for dashCount := 2; ; {
+		c := z.readByte()
+		if z.err != nil {
+			// Ignore up to two dashes at EOF.
+			if dashCount > 2 {
+				dashCount = 2
+			}
+			z.data.end = z.raw.end - dashCount
+			return
+		}
+		switch c {
+		case '-':
+			dashCount++
+			continue
+		case '>':
+			if dashCount >= 2 {
+				z.data.end = z.raw.end - len("-->")
+				return
+			}
+		case '!':
+			if dashCount >= 2 {
+				c = z.readByte()
+				if z.err != nil {
+					z.data.end = z.raw.end
+					return
+				}
+				if c == '>' {
+					z.data.end = z.raw.end - len("--!>")
+					return
+				}
+			}
+		}
+		dashCount = 0
+	}
+}
+
+// readUntilCloseAngle reads until the next ">".
+func (z *Tokenizer) readUntilCloseAngle() {
+	z.data.start = z.raw.end
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return
+		}
+		if c == '>' {
+			z.data.end = z.raw.end - len(">")
+			return
+		}
+	}
+}
+
+// readMarkupDeclaration reads the next token starting with "<!". It might be
+// a "<!--comment-->", a "<!DOCTYPE foo>", a "<![CDATA[section]]>" or
+// "<!a bogus comment". The opening "<!" has already been consumed.
+func (z *Tokenizer) readMarkupDeclaration() TokenType {
+	z.data.start = z.raw.end
+	var c [2]byte
+	for i := 0; i < 2; i++ {
+		c[i] = z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return CommentToken
+		}
+	}
+	if c[0] == '-' && c[1] == '-' {
+		z.readComment()
+		return CommentToken
+	}
+	z.raw.end -= 2
+	if z.readDoctype() {
+		return DoctypeToken
+	}
+	if z.allowCDATA && z.readCDATA() {
+		z.convertNUL = true
+		return TextToken
+	}
+	// It's a bogus comment.
+	z.readUntilCloseAngle()
+	return CommentToken
+}
+
+// readDoctype attempts to read a doctype declaration and returns true if
+// successful. The opening "<!" has already been consumed.
+func (z *Tokenizer) readDoctype() bool {
+	const s = "DOCTYPE"
+	for i := 0; i < len(s); i++ {
+		c := z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return false
+		}
+		if c != s[i] && c != s[i]+('a'-'A') {
+			// Back up to read the fragment of "DOCTYPE" again.
+			z.raw.end = z.data.start
+			return false
+		}
+	}
+	if z.skipWhiteSpace(); z.err != nil {
+		z.data.start = z.raw.end
+		z.data.end = z.raw.end
+		return true
+	}
+	z.readUntilCloseAngle()
+	return true
+}
+
+// readCDATA attempts to read a CDATA section and returns true if
+// successful. The opening "<!" has already been consumed.
+func (z *Tokenizer) readCDATA() bool {
+	const s = "[CDATA["
+	for i := 0; i < len(s); i++ {
+		c := z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return false
+		}
+		if c != s[i] {
+			// Back up to read the fragment of "[CDATA[" again.
+			z.raw.end = z.data.start
+			return false
+		}
+	}
+	z.data.start = z.raw.end
+	brackets := 0
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return true
+		}
+		switch c {
+		case ']':
+			brackets++
+		case '>':
+			if brackets >= 2 {
+				z.data.end = z.raw.end - len("]]>")
+				return true
+			}
+			brackets = 0
+		default:
+			brackets = 0
+		}
+	}
+}
+
+// startTagIn returns whether the start tag in z.buf[z.data.start:z.data.end]
+// case-insensitively matches any element of ss.
+func (z *Tokenizer) startTagIn(ss ...string) bool {
+loop:
+	for _, s := range ss {
+		if z.data.end-z.data.start != len(s) {
+			continue loop
+		}
+		for i := 0; i < len(s); i++ {
+			c := z.buf[z.data.start+i]
+			if 'A' <= c && c <= 'Z' {
+				c += 'a' - 'A'
+			}
+			if c != s[i] {
+				continue loop
+			}
+		}
+		return true
+	}
+	return false
+}
+
+// readStartTag reads the next start tag token. The opening "<a" has already
+// been consumed, where 'a' means anything in [A-Za-z].
+func (z *Tokenizer) readStartTag() TokenType {
+	z.readTag(true)
+	if z.err != nil {
+		return ErrorToken
+	}
+	// Several tags flag the tokenizer's next token as raw.
+	c, raw := z.buf[z.data.start], false
+	if 'A' <= c && c <= 'Z' {
+		c += 'a' - 'A'
+	}
+	switch c {
+	case 'i':
+		raw = z.startTagIn("iframe")
+	case 'n':
+		raw = z.startTagIn("noembed", "noframes", "noscript")
+	case 'p':
+		raw = z.startTagIn("plaintext")
+	case 's':
+		raw = z.startTagIn("script", "style")
+	case 't':
+		raw = z.startTagIn("textarea", "title")
+	case 'x':
+		raw = z.startTagIn("xmp")
+	}
+	if raw {
+		z.rawTag = strings.ToLower(string(z.buf[z.data.start:z.data.end]))
+	}
+	// Look for a self-closing token like "<br/>".
+	if z.err == nil && z.buf[z.raw.end-2] == '/' {
+		return SelfClosingTagToken
+	}
+	return StartTagToken
+}
+
+// readTag reads the next tag token and its attributes. If saveAttr, those
+// attributes are saved in z.attr, otherwise z.attr is set to an empty slice.
+// The opening "<a" or "</a" has already been consumed, where 'a' means anything
+// in [A-Za-z].
+func (z *Tokenizer) readTag(saveAttr bool) {
+	z.attr = z.attr[:0]
+	z.nAttrReturned = 0
+	// Read the tag name and attribute key/value pairs.
+	z.readTagName()
+	if z.skipWhiteSpace(); z.err != nil {
+		return
+	}
+	for {
+		c := z.readByte()
+		if z.err != nil || c == '>' {
+			break
+		}
+		z.raw.end--
+		z.readTagAttrKey()
+		z.readTagAttrVal()
+		// Save pendingAttr if saveAttr and that attribute has a non-empty key.
+		if saveAttr && z.pendingAttr[0].start != z.pendingAttr[0].end {
+			z.attr = append(z.attr, z.pendingAttr)
+		}
+		if z.skipWhiteSpace(); z.err != nil {
+			break
+		}
+	}
+}
+
+// readTagName sets z.data to the "div" in "<div k=v>". The reader (z.raw.end)
+// is positioned such that the first byte of the tag name (the "d" in "<div")
+// has already been consumed.
+func (z *Tokenizer) readTagName() {
+	z.data.start = z.raw.end - 1
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			z.data.end = z.raw.end
+			return
+		}
+		switch c {
+		case ' ', '\n', '\r', '\t', '\f':
+			z.data.end = z.raw.end - 1
+			return
+		case '/', '>':
+			z.raw.end--
+			z.data.end = z.raw.end
+			return
+		}
+	}
+}
+
+// readTagAttrKey sets z.pendingAttr[0] to the "k" in "<div k=v>".
+// Precondition: z.err == nil.
+func (z *Tokenizer) readTagAttrKey() {
+	z.pendingAttr[0].start = z.raw.end
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			z.pendingAttr[0].end = z.raw.end
+			return
+		}
+		switch c {
+		case ' ', '\n', '\r', '\t', '\f', '/':
+			z.pendingAttr[0].end = z.raw.end - 1
+			return
+		case '=', '>':
+			z.raw.end--
+			z.pendingAttr[0].end = z.raw.end
+			return
+		}
+	}
+}
+
+// readTagAttrVal sets z.pendingAttr[1] to the "v" in "<div k=v>".
+func (z *Tokenizer) readTagAttrVal() {
+	z.pendingAttr[1].start = z.raw.end
+	z.pendingAttr[1].end = z.raw.end
+	if z.skipWhiteSpace(); z.err != nil {
+		return
+	}
+	c := z.readByte()
+	if z.err != nil {
+		return
+	}
+	if c != '=' {
+		z.raw.end--
+		return
+	}
+	if z.skipWhiteSpace(); z.err != nil {
+		return
+	}
+	quote := z.readByte()
+	if z.err != nil {
+		return
+	}
+	switch quote {
+	case '>':
+		z.raw.end--
+		return
+
+	case '\'', '"':
+		z.pendingAttr[1].start = z.raw.end
+		for {
+			c := z.readByte()
+			if z.err != nil {
+				z.pendingAttr[1].end = z.raw.end
+				return
+			}
+			if c == quote {
+				z.pendingAttr[1].end = z.raw.end - 1
+				return
+			}
+		}
+
+	default:
+		z.pendingAttr[1].start = z.raw.end - 1
+		for {
+			c := z.readByte()
+			if z.err != nil {
+				z.pendingAttr[1].end = z.raw.end
+				return
+			}
+			switch c {
+			case ' ', '\n', '\r', '\t', '\f':
+				z.pendingAttr[1].end = z.raw.end - 1
+				return
+			case '>':
+				z.raw.end--
+				z.pendingAttr[1].end = z.raw.end
+				return
+			}
+		}
+	}
+}
+
+// Next scans the next token and returns its type.
+func (z *Tokenizer) Next() TokenType {
+	z.raw.start = z.raw.end
+	z.data.start = z.raw.end
+	z.data.end = z.raw.end
+	if z.err != nil {
+		z.tt = ErrorToken
+		return z.tt
+	}
+	if z.rawTag != "" {
+		if z.rawTag == "plaintext" {
+			// Read everything up to EOF.
+			for z.err == nil {
+				z.readByte()
+			}
+			z.data.end = z.raw.end
+			z.textIsRaw = true
+		} else {
+			z.readRawOrRCDATA()
+		}
+		if z.data.end > z.data.start {
+			z.tt = TextToken
+			z.convertNUL = true
+			return z.tt
+		}
+	}
+	z.textIsRaw = false
+	z.convertNUL = false
+
+loop:
+	for {
+		c := z.readByte()
+		if z.err != nil {
+			break loop
+		}
+		if c != '<' {
+			continue loop
+		}
+
+		// Check if the '<' we have just read is part of a tag, comment
+		// or doctype. If not, it's part of the accumulated text token.
+		c = z.readByte()
+		if z.err != nil {
+			break loop
+		}
+		var tokenType TokenType
+		switch {
+		case 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z':
+			tokenType = StartTagToken
+		case c == '/':
+			tokenType = EndTagToken
+		case c == '!' || c == '?':
+			// We use CommentToken to mean any of "<!--actual comments-->",
+			// "<!DOCTYPE declarations>" and "<?xml processing instructions?>".
+			tokenType = CommentToken
+		default:
+			// Reconsume the current character.
+			z.raw.end--
+			continue
+		}
+
+		// We have a non-text token, but we might have accumulated some text
+		// before that. If so, we return the text first, and return the non-
+		// text token on the subsequent call to Next.
+		if x := z.raw.end - len("<a"); z.raw.start < x {
+			z.raw.end = x
+			z.data.end = x
+			z.tt = TextToken
+			return z.tt
+		}
+		switch tokenType {
+		case StartTagToken:
+			z.tt = z.readStartTag()
+			return z.tt
+		case EndTagToken:
+			c = z.readByte()
+			if z.err != nil {
+				break loop
+			}
+			if c == '>' {
+				// "</>" does not generate a token at all. Generate an empty comment
+				// to allow passthrough clients to pick up the data using Raw.
+				// Reset the tokenizer state and start again.
+				z.tt = CommentToken
+				return z.tt
+			}
+			if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' {
+				z.readTag(false)
+				if z.err != nil {
+					z.tt = ErrorToken
+				} else {
+					z.tt = EndTagToken
+				}
+				return z.tt
+			}
+			z.raw.end--
+			z.readUntilCloseAngle()
+			z.tt = CommentToken
+			return z.tt
+		case CommentToken:
+			if c == '!' {
+				z.tt = z.readMarkupDeclaration()
+				return z.tt
+			}
+			z.raw.end--
+			z.readUntilCloseAngle()
+			z.tt = CommentToken
+			return z.tt
+		}
+	}
+	if z.raw.start < z.raw.end {
+		z.data.end = z.raw.end
+		z.tt = TextToken
+		return z.tt
+	}
+	z.tt = ErrorToken
+	return z.tt
+}
+
+// Raw returns the unmodified text of the current token. Calling Next, Token,
+// Text, TagName or TagAttr may change the contents of the returned slice.
+func (z *Tokenizer) Raw() []byte {
+	return z.buf[z.raw.start:z.raw.end]
+}
+
+// convertNewlines converts "\r" and "\r\n" in s to "\n".
+// The conversion happens in place, but the resulting slice may be shorter.
+func convertNewlines(s []byte) []byte {
+	for i, c := range s {
+		if c != '\r' {
+			continue
+		}
+
+		src := i + 1
+		if src >= len(s) || s[src] != '\n' {
+			s[i] = '\n'
+			continue
+		}
+
+		dst := i
+		for src < len(s) {
+			if s[src] == '\r' {
+				if src+1 < len(s) && s[src+1] == '\n' {
+					src++
+				}
+				s[dst] = '\n'
+			} else {
+				s[dst] = s[src]
+			}
+			src++
+			dst++
+		}
+		return s[:dst]
+	}
+	return s
+}
+
+var (
+	nul         = []byte("\x00")
+	replacement = []byte("\ufffd")
+)
+
+// Text returns the unescaped text of a text, comment or doctype token. The
+// contents of the returned slice may change on the next call to Next.
+func (z *Tokenizer) Text() []byte {
+	switch z.tt {
+	case TextToken, CommentToken, DoctypeToken:
+		s := z.buf[z.data.start:z.data.end]
+		z.data.start = z.raw.end
+		z.data.end = z.raw.end
+		s = convertNewlines(s)
+		if (z.convertNUL || z.tt == CommentToken) && bytes.Contains(s, nul) {
+			s = bytes.Replace(s, nul, replacement, -1)
+		}
+		if !z.textIsRaw {
+			s = unescape(s, false)
+		}
+		return s
+	}
+	return nil
+}
+
+// TagName returns the lower-cased name of a tag token (the `img` out of
+// `<IMG SRC="foo">`) and whether the tag has attributes.
+// The contents of the returned slice may change on the next call to Next.
+func (z *Tokenizer) TagName() (name []byte, hasAttr bool) {
+	if z.data.start < z.data.end {
+		switch z.tt {
+		case StartTagToken, EndTagToken, SelfClosingTagToken:
+			s := z.buf[z.data.start:z.data.end]
+			z.data.start = z.raw.end
+			z.data.end = z.raw.end
+			return lower(s), z.nAttrReturned < len(z.attr)
+		}
+	}
+	return nil, false
+}
+
+// TagAttr returns the lower-cased key and unescaped value of the next unparsed
+// attribute for the current tag token and whether there are more attributes.
+// The contents of the returned slices may change on the next call to Next.
+func (z *Tokenizer) TagAttr() (key, val []byte, moreAttr bool) {
+	if z.nAttrReturned < len(z.attr) {
+		switch z.tt {
+		case StartTagToken, SelfClosingTagToken:
+			x := z.attr[z.nAttrReturned]
+			z.nAttrReturned++
+			key = z.buf[x[0].start:x[0].end]
+			val = z.buf[x[1].start:x[1].end]
+			return lower(key), unescape(convertNewlines(val), true), z.nAttrReturned < len(z.attr)
+		}
+	}
+	return nil, nil, false
+}
+
+// Token returns the current Token. The result's Data and Attr values remain
+// valid after subsequent Next calls.
+func (z *Tokenizer) Token() Token {
+	t := Token{Type: z.tt}
+	switch z.tt {
+	case TextToken, CommentToken, DoctypeToken:
+		t.Data = string(z.Text())
+	case StartTagToken, SelfClosingTagToken, EndTagToken:
+		name, moreAttr := z.TagName()
+		for moreAttr {
+			var key, val []byte
+			key, val, moreAttr = z.TagAttr()
+			t.Attr = append(t.Attr, Attribute{"", atom.String(key), string(val)})
+		}
+		if a := atom.Lookup(name); a != 0 {
+			t.DataAtom, t.Data = a, a.String()
+		} else {
+			t.DataAtom, t.Data = 0, string(name)
+		}
+	}
+	return t
+}
+
+// SetMaxBuf sets a limit on the amount of data buffered during tokenization.
+// A value of 0 means unlimited.
+func (z *Tokenizer) SetMaxBuf(n int) {
+	z.maxBuf = n
+}
+
+// NewTokenizer returns a new HTML Tokenizer for the given Reader.
+// The input is assumed to be UTF-8 encoded.
+func NewTokenizer(r io.Reader) *Tokenizer {
+	return NewTokenizerFragment(r, "")
+}
+
+// NewTokenizerFragment returns a new HTML Tokenizer for the given Reader, for
+// tokenizing an existing element's InnerHTML fragment. contextTag is that
+// element's tag, such as "div" or "iframe".
+//
+// For example, how the InnerHTML "a<b" is tokenized depends on whether it is
+// for a <p> tag or a <script> tag.
+//
+// The input is assumed to be UTF-8 encoded.
+func NewTokenizerFragment(r io.Reader, contextTag string) *Tokenizer {
+	z := &Tokenizer{
+		r:   r,
+		buf: make([]byte, 0, 4096),
+	}
+	if contextTag != "" {
+		switch s := strings.ToLower(contextTag); s {
+		case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "title", "textarea", "xmp":
+			z.rawTag = s
+		}
+	}
+	return z
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/.gitignore b/backend/vendor/gopkg.in/russross/blackfriday.v2/.gitignore
new file mode 100644
index 00000000..75623dcc
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/.gitignore
@@ -0,0 +1,8 @@
+*.out
+*.swp
+*.8
+*.6
+_obj
+_test*
+markdown
+tags
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/.travis.yml b/backend/vendor/gopkg.in/russross/blackfriday.v2/.travis.yml
new file mode 100644
index 00000000..a4eb2577
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/.travis.yml
@@ -0,0 +1,18 @@
+# Travis CI (http://travis-ci.org/) is a continuous integration service for
+# open source projects. This file configures it to run unit tests for
+# blackfriday.
+
+language: go
+
+go:
+    - 1.5
+    - 1.6
+    - 1.7
+
+install:
+    - go get -d -t -v ./...
+    - go build -v ./...
+
+script:
+    - go test -v ./...
+    - go test -run=^$ -bench=BenchmarkReference -benchmem
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/LICENSE.txt b/backend/vendor/gopkg.in/russross/blackfriday.v2/LICENSE.txt
new file mode 100644
index 00000000..2885af36
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/LICENSE.txt
@@ -0,0 +1,29 @@
+Blackfriday is distributed under the Simplified BSD License:
+
+> Copyright © 2011 Russ Ross
+> All rights reserved.
+>
+> Redistribution and use in source and binary forms, with or without
+> modification, are permitted provided that the following conditions
+> are met:
+>
+> 1.  Redistributions of source code must retain the above copyright
+>     notice, this list of conditions and the following disclaimer.
+>
+> 2.  Redistributions in binary form must reproduce the above
+>     copyright notice, this list of conditions and the following
+>     disclaimer in the documentation and/or other materials provided with
+>     the distribution.
+>
+> THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+> "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+> LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+> FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+> COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+> INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+> BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+> LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+> CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+> LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+> ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+> POSSIBILITY OF SUCH DAMAGE.
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/README.md b/backend/vendor/gopkg.in/russross/blackfriday.v2/README.md
new file mode 100644
index 00000000..2e0db355
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/README.md
@@ -0,0 +1,283 @@
+Blackfriday [![Build Status](https://travis-ci.org/russross/blackfriday.svg?branch=master)](https://travis-ci.org/russross/blackfriday)
+===========
+
+Blackfriday is a [Markdown][1] processor implemented in [Go][2]. It
+is paranoid about its input (so you can safely feed it user-supplied
+data), it is fast, it supports common extensions (tables, smart
+punctuation substitutions, etc.), and it is safe for all utf-8
+(unicode) input.
+
+HTML output is currently supported, along with Smartypants
+extensions.
+
+It started as a translation from C of [Sundown][3].
+
+
+Installation
+------------
+
+Blackfriday is compatible with any modern Go release. With Go 1.7 and git
+installed:
+
+    go get gopkg.in/russross/blackfriday.v2
+
+will download, compile, and install the package into your `$GOPATH`
+directory hierarchy. Alternatively, you can achieve the same if you
+import it into a project:
+
+    import "gopkg.in/russross/blackfriday.v2"
+
+and `go get` without parameters.
+
+
+Versions
+--------
+
+Currently maintained and recommended version of Blackfriday is `v2`. It's being
+developed on its own branch: https://github.com/russross/blackfriday/v2. You
+should install and import it via [gopkg.in][6] at
+`gopkg.in/russross/blackfriday.v2`.
+
+Version 2 offers a number of improvements over v1:
+
+* Cleaned up API
+* A separate call to [`Parse`][4], which produces an abstract syntax tree for
+  the document
+* Latest bug fixes
+* Flexibility to easily add your own rendering extensions
+
+Potential drawbacks:
+
+* Our benchmarks show v2 to be slightly slower than v1. Currently in the
+  ballpark of around 15%.
+* API breakage. If you can't afford modifying your code to adhere to the new API
+  and don't care too much about the new features, v2 is probably not for you.
+* Several bug fixes are trailing behind and still need to be forward-ported to
+  v2. See issue [#348](https://github.com/russross/blackfriday/issues/348) for
+  tracking.
+
+Usage
+-----
+
+For the most sensible markdown processing, it is as simple as getting your input
+into a byte slice and calling:
+
+```go
+output := blackfriday.Run(input)
+```
+
+Your input will be parsed and the output rendered with a set of most popular
+extensions enabled. If you want the most basic feature set, corresponding with
+the bare Markdown specification, use:
+
+```go
+output := blackfriday.Run(input, blackfriday.WithNoExtensions())
+```
+
+### Sanitize untrusted content
+
+Blackfriday itself does nothing to protect against malicious content. If you are
+dealing with user-supplied markdown, we recommend running Blackfriday's output
+through HTML sanitizer such as [Bluemonday][5].
+
+Here's an example of simple usage of Blackfriday together with Bluemonday:
+
+```go
+import (
+    "github.com/microcosm-cc/bluemonday"
+    "github.com/russross/blackfriday"
+)
+
+// ...
+unsafe := blackfriday.Run(input)
+html := bluemonday.UGCPolicy().SanitizeBytes(unsafe)
+```
+
+### Custom options
+
+If you want to customize the set of options, use `blackfriday.WithExtensions`,
+`blackfriday.WithRenderer` and `blackfriday.WithRefOverride`.
+
+You can also check out `blackfriday-tool` for a more complete example
+of how to use it. Download and install it using:
+
+    go get github.com/russross/blackfriday-tool
+
+This is a simple command-line tool that allows you to process a
+markdown file using a standalone program.  You can also browse the
+source directly on github if you are just looking for some example
+code:
+
+* <http://github.com/russross/blackfriday-tool>
+
+Note that if you have not already done so, installing
+`blackfriday-tool` will be sufficient to download and install
+blackfriday in addition to the tool itself. The tool binary will be
+installed in `$GOPATH/bin`.  This is a statically-linked binary that
+can be copied to wherever you need it without worrying about
+dependencies and library versions.
+
+
+Features
+--------
+
+All features of Sundown are supported, including:
+
+*   **Compatibility**. The Markdown v1.0.3 test suite passes with
+    the `--tidy` option.  Without `--tidy`, the differences are
+    mostly in whitespace and entity escaping, where blackfriday is
+    more consistent and cleaner.
+
+*   **Common extensions**, including table support, fenced code
+    blocks, autolinks, strikethroughs, non-strict emphasis, etc.
+
+*   **Safety**. Blackfriday is paranoid when parsing, making it safe
+    to feed untrusted user input without fear of bad things
+    happening. The test suite stress tests this and there are no
+    known inputs that make it crash.  If you find one, please let me
+    know and send me the input that does it.
+
+    NOTE: "safety" in this context means *runtime safety only*. In order to
+    protect yourself against JavaScript injection in untrusted content, see
+    [this example](https://github.com/russross/blackfriday#sanitize-untrusted-content).
+
+*   **Fast processing**. It is fast enough to render on-demand in
+    most web applications without having to cache the output.
+
+*   **Thread safety**. You can run multiple parsers in different
+    goroutines without ill effect. There is no dependence on global
+    shared state.
+
+*   **Minimal dependencies**. Blackfriday only depends on standard
+    library packages in Go. The source code is pretty
+    self-contained, so it is easy to add to any project, including
+    Google App Engine projects.
+
+*   **Standards compliant**. Output successfully validates using the
+    W3C validation tool for HTML 4.01 and XHTML 1.0 Transitional.
+
+
+Extensions
+----------
+
+In addition to the standard markdown syntax, this package
+implements the following extensions:
+
+*   **Intra-word emphasis supression**. The `_` character is
+    commonly used inside words when discussing code, so having
+    markdown interpret it as an emphasis command is usually the
+    wrong thing. Blackfriday lets you treat all emphasis markers as
+    normal characters when they occur inside a word.
+
+*   **Tables**. Tables can be created by drawing them in the input
+    using a simple syntax:
+
+    ```
+    Name    | Age
+    --------|------
+    Bob     | 27
+    Alice   | 23
+    ```
+
+*   **Fenced code blocks**. In addition to the normal 4-space
+    indentation to mark code blocks, you can explicitly mark them
+    and supply a language (to make syntax highlighting simple). Just
+    mark it like this:
+
+        ```go
+        func getTrue() bool {
+            return true
+        }
+        ```
+
+    You can use 3 or more backticks to mark the beginning of the
+    block, and the same number to mark the end of the block.
+
+*   **Definition lists**. A simple definition list is made of a single-line
+    term followed by a colon and the definition for that term.
+
+        Cat
+        : Fluffy animal everyone likes
+        
+        Internet
+        : Vector of transmission for pictures of cats
+
+    Terms must be separated from the previous definition by a blank line.
+
+*   **Footnotes**. A marker in the text that will become a superscript number;
+    a footnote definition that will be placed in a list of footnotes at the
+    end of the document. A footnote looks like this:
+
+        This is a footnote.[^1]
+        
+        [^1]: the footnote text.
+
+*   **Autolinking**. Blackfriday can find URLs that have not been
+    explicitly marked as links and turn them into links.
+
+*   **Strikethrough**. Use two tildes (`~~`) to mark text that
+    should be crossed out.
+
+*   **Hard line breaks**. With this extension enabled newlines in the input
+    translate into line breaks in the output. This extension is off by default.
+
+*   **Smart quotes**. Smartypants-style punctuation substitution is
+    supported, turning normal double- and single-quote marks into
+    curly quotes, etc.
+
+*   **LaTeX-style dash parsing** is an additional option, where `--`
+    is translated into `&ndash;`, and `---` is translated into
+    `&mdash;`. This differs from most smartypants processors, which
+    turn a single hyphen into an ndash and a double hyphen into an
+    mdash.
+
+*   **Smart fractions**, where anything that looks like a fraction
+    is translated into suitable HTML (instead of just a few special
+    cases like most smartypant processors). For example, `4/5`
+    becomes `<sup>4</sup>&frasl;<sub>5</sub>`, which renders as
+    <sup>4</sup>&frasl;<sub>5</sub>.
+
+
+Other renderers
+---------------
+
+Blackfriday is structured to allow alternative rendering engines. Here
+are a few of note:
+
+*   [github_flavored_markdown](https://godoc.org/github.com/shurcooL/github_flavored_markdown):
+    provides a GitHub Flavored Markdown renderer with fenced code block
+    highlighting, clickable heading anchor links.
+
+    It's not customizable, and its goal is to produce HTML output
+    equivalent to the [GitHub Markdown API endpoint](https://developer.github.com/v3/markdown/#render-a-markdown-document-in-raw-mode),
+    except the rendering is performed locally.
+
+*   [markdownfmt](https://github.com/shurcooL/markdownfmt): like gofmt,
+    but for markdown.
+
+*   [LaTeX output](https://bitbucket.org/ambrevar/blackfriday-latex):
+    renders output as LaTeX.
+
+
+Todo
+----
+
+*   More unit testing
+*   Improve unicode support. It does not understand all unicode
+    rules (about what constitutes a letter, a punctuation symbol,
+    etc.), so it may fail to detect word boundaries correctly in
+    some instances. It is safe on all utf-8 input.
+
+
+License
+-------
+
+[Blackfriday is distributed under the Simplified BSD License](LICENSE.txt)
+
+
+   [1]: https://daringfireball.net/projects/markdown/ "Markdown"
+   [2]: https://golang.org/ "Go Language"
+   [3]: https://github.com/vmg/sundown "Sundown"
+   [4]: https://godoc.org/gopkg.in/russross/blackfriday.v2#Parse "Parse func"
+   [5]: https://github.com/microcosm-cc/bluemonday "Bluemonday"
+   [6]: https://labix.org/gopkg.in "gopkg.in"
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/block.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/block.go
new file mode 100644
index 00000000..d7da33f2
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/block.go
@@ -0,0 +1,1549 @@
+//
+// Blackfriday Markdown Processor
+// Available at http://github.com/russross/blackfriday
+//
+// Copyright © 2011 Russ Ross <russ@russross.com>.
+// Distributed under the Simplified BSD License.
+// See README.md for details.
+//
+
+//
+// Functions to parse block-level elements.
+//
+
+package blackfriday
+
+import (
+	"bytes"
+	"html"
+	"regexp"
+
+	"github.com/shurcooL/sanitized_anchor_name"
+)
+
+const (
+	charEntity = "&(?:#x[a-f0-9]{1,8}|#[0-9]{1,8}|[a-z][a-z0-9]{1,31});"
+	escapable  = "[!\"#$%&'()*+,./:;<=>?@[\\\\\\]^_`{|}~-]"
+)
+
+var (
+	reBackslashOrAmp      = regexp.MustCompile("[\\&]")
+	reEntityOrEscapedChar = regexp.MustCompile("(?i)\\\\" + escapable + "|" + charEntity)
+)
+
+// Parse block-level data.
+// Note: this function and many that it calls assume that
+// the input buffer ends with a newline.
+func (p *Markdown) block(data []byte) {
+	// this is called recursively: enforce a maximum depth
+	if p.nesting >= p.maxNesting {
+		return
+	}
+	p.nesting++
+
+	// parse out one block-level construct at a time
+	for len(data) > 0 {
+		// prefixed heading:
+		//
+		// # Heading 1
+		// ## Heading 2
+		// ...
+		// ###### Heading 6
+		if p.isPrefixHeading(data) {
+			data = data[p.prefixHeading(data):]
+			continue
+		}
+
+		// block of preformatted HTML:
+		//
+		// <div>
+		//     ...
+		// </div>
+		if data[0] == '<' {
+			if i := p.html(data, true); i > 0 {
+				data = data[i:]
+				continue
+			}
+		}
+
+		// title block
+		//
+		// % stuff
+		// % more stuff
+		// % even more stuff
+		if p.extensions&Titleblock != 0 {
+			if data[0] == '%' {
+				if i := p.titleBlock(data, true); i > 0 {
+					data = data[i:]
+					continue
+				}
+			}
+		}
+
+		// blank lines.  note: returns the # of bytes to skip
+		if i := p.isEmpty(data); i > 0 {
+			data = data[i:]
+			continue
+		}
+
+		// indented code block:
+		//
+		//     func max(a, b int) int {
+		//         if a > b {
+		//             return a
+		//         }
+		//         return b
+		//      }
+		if p.codePrefix(data) > 0 {
+			data = data[p.code(data):]
+			continue
+		}
+
+		// fenced code block:
+		//
+		// ``` go
+		// func fact(n int) int {
+		//     if n <= 1 {
+		//         return n
+		//     }
+		//     return n * fact(n-1)
+		// }
+		// ```
+		if p.extensions&FencedCode != 0 {
+			if i := p.fencedCodeBlock(data, true); i > 0 {
+				data = data[i:]
+				continue
+			}
+		}
+
+		// horizontal rule:
+		//
+		// ------
+		// or
+		// ******
+		// or
+		// ______
+		if p.isHRule(data) {
+			p.addBlock(HorizontalRule, nil)
+			var i int
+			for i = 0; i < len(data) && data[i] != '\n'; i++ {
+			}
+			data = data[i:]
+			continue
+		}
+
+		// block quote:
+		//
+		// > A big quote I found somewhere
+		// > on the web
+		if p.quotePrefix(data) > 0 {
+			data = data[p.quote(data):]
+			continue
+		}
+
+		// table:
+		//
+		// Name  | Age | Phone
+		// ------|-----|---------
+		// Bob   | 31  | 555-1234
+		// Alice | 27  | 555-4321
+		if p.extensions&Tables != 0 {
+			if i := p.table(data); i > 0 {
+				data = data[i:]
+				continue
+			}
+		}
+
+		// an itemized/unordered list:
+		//
+		// * Item 1
+		// * Item 2
+		//
+		// also works with + or -
+		if p.uliPrefix(data) > 0 {
+			data = data[p.list(data, 0):]
+			continue
+		}
+
+		// a numbered/ordered list:
+		//
+		// 1. Item 1
+		// 2. Item 2
+		if p.oliPrefix(data) > 0 {
+			data = data[p.list(data, ListTypeOrdered):]
+			continue
+		}
+
+		// definition lists:
+		//
+		// Term 1
+		// :   Definition a
+		// :   Definition b
+		//
+		// Term 2
+		// :   Definition c
+		if p.extensions&DefinitionLists != 0 {
+			if p.dliPrefix(data) > 0 {
+				data = data[p.list(data, ListTypeDefinition):]
+				continue
+			}
+		}
+
+		// anything else must look like a normal paragraph
+		// note: this finds underlined headings, too
+		data = data[p.paragraph(data):]
+	}
+
+	p.nesting--
+}
+
+func (p *Markdown) addBlock(typ NodeType, content []byte) *Node {
+	p.closeUnmatchedBlocks()
+	container := p.addChild(typ, 0)
+	container.content = content
+	return container
+}
+
+func (p *Markdown) isPrefixHeading(data []byte) bool {
+	if data[0] != '#' {
+		return false
+	}
+
+	if p.extensions&SpaceHeadings != 0 {
+		level := 0
+		for level < 6 && level < len(data) && data[level] == '#' {
+			level++
+		}
+		if level == len(data) || data[level] != ' ' {
+			return false
+		}
+	}
+	return true
+}
+
+func (p *Markdown) prefixHeading(data []byte) int {
+	level := 0
+	for level < 6 && level < len(data) && data[level] == '#' {
+		level++
+	}
+	i := skipChar(data, level, ' ')
+	end := skipUntilChar(data, i, '\n')
+	skip := end
+	id := ""
+	if p.extensions&HeadingIDs != 0 {
+		j, k := 0, 0
+		// find start/end of heading id
+		for j = i; j < end-1 && (data[j] != '{' || data[j+1] != '#'); j++ {
+		}
+		for k = j + 1; k < end && data[k] != '}'; k++ {
+		}
+		// extract heading id iff found
+		if j < end && k < end {
+			id = string(data[j+2 : k])
+			end = j
+			skip = k + 1
+			for end > 0 && data[end-1] == ' ' {
+				end--
+			}
+		}
+	}
+	for end > 0 && data[end-1] == '#' {
+		if isBackslashEscaped(data, end-1) {
+			break
+		}
+		end--
+	}
+	for end > 0 && data[end-1] == ' ' {
+		end--
+	}
+	if end > i {
+		if id == "" && p.extensions&AutoHeadingIDs != 0 {
+			id = sanitized_anchor_name.Create(string(data[i:end]))
+		}
+		block := p.addBlock(Heading, data[i:end])
+		block.HeadingID = id
+		block.Level = level
+	}
+	return skip
+}
+
+func (p *Markdown) isUnderlinedHeading(data []byte) int {
+	// test of level 1 heading
+	if data[0] == '=' {
+		i := skipChar(data, 1, '=')
+		i = skipChar(data, i, ' ')
+		if i < len(data) && data[i] == '\n' {
+			return 1
+		}
+		return 0
+	}
+
+	// test of level 2 heading
+	if data[0] == '-' {
+		i := skipChar(data, 1, '-')
+		i = skipChar(data, i, ' ')
+		if i < len(data) && data[i] == '\n' {
+			return 2
+		}
+		return 0
+	}
+
+	return 0
+}
+
+func (p *Markdown) titleBlock(data []byte, doRender bool) int {
+	if data[0] != '%' {
+		return 0
+	}
+	splitData := bytes.Split(data, []byte("\n"))
+	var i int
+	for idx, b := range splitData {
+		if !bytes.HasPrefix(b, []byte("%")) {
+			i = idx // - 1
+			break
+		}
+	}
+
+	data = bytes.Join(splitData[0:i], []byte("\n"))
+	consumed := len(data)
+	data = bytes.TrimPrefix(data, []byte("% "))
+	data = bytes.Replace(data, []byte("\n% "), []byte("\n"), -1)
+	block := p.addBlock(Heading, data)
+	block.Level = 1
+	block.IsTitleblock = true
+
+	return consumed
+}
+
+func (p *Markdown) html(data []byte, doRender bool) int {
+	var i, j int
+
+	// identify the opening tag
+	if data[0] != '<' {
+		return 0
+	}
+	curtag, tagfound := p.htmlFindTag(data[1:])
+
+	// handle special cases
+	if !tagfound {
+		// check for an HTML comment
+		if size := p.htmlComment(data, doRender); size > 0 {
+			return size
+		}
+
+		// check for an <hr> tag
+		if size := p.htmlHr(data, doRender); size > 0 {
+			return size
+		}
+
+		// no special case recognized
+		return 0
+	}
+
+	// look for an unindented matching closing tag
+	// followed by a blank line
+	found := false
+	/*
+		closetag := []byte("\n</" + curtag + ">")
+		j = len(curtag) + 1
+		for !found {
+			// scan for a closing tag at the beginning of a line
+			if skip := bytes.Index(data[j:], closetag); skip >= 0 {
+				j += skip + len(closetag)
+			} else {
+				break
+			}
+
+			// see if it is the only thing on the line
+			if skip := p.isEmpty(data[j:]); skip > 0 {
+				// see if it is followed by a blank line/eof
+				j += skip
+				if j >= len(data) {
+					found = true
+					i = j
+				} else {
+					if skip := p.isEmpty(data[j:]); skip > 0 {
+						j += skip
+						found = true
+						i = j
+					}
+				}
+			}
+		}
+	*/
+
+	// if not found, try a second pass looking for indented match
+	// but not if tag is "ins" or "del" (following original Markdown.pl)
+	if !found && curtag != "ins" && curtag != "del" {
+		i = 1
+		for i < len(data) {
+			i++
+			for i < len(data) && !(data[i-1] == '<' && data[i] == '/') {
+				i++
+			}
+
+			if i+2+len(curtag) >= len(data) {
+				break
+			}
+
+			j = p.htmlFindEnd(curtag, data[i-1:])
+
+			if j > 0 {
+				i += j - 1
+				found = true
+				break
+			}
+		}
+	}
+
+	if !found {
+		return 0
+	}
+
+	// the end of the block has been found
+	if doRender {
+		// trim newlines
+		end := i
+		for end > 0 && data[end-1] == '\n' {
+			end--
+		}
+		finalizeHTMLBlock(p.addBlock(HTMLBlock, data[:end]))
+	}
+
+	return i
+}
+
+func finalizeHTMLBlock(block *Node) {
+	block.Literal = block.content
+	block.content = nil
+}
+
+// HTML comment, lax form
+func (p *Markdown) htmlComment(data []byte, doRender bool) int {
+	i := p.inlineHTMLComment(data)
+	// needs to end with a blank line
+	if j := p.isEmpty(data[i:]); j > 0 {
+		size := i + j
+		if doRender {
+			// trim trailing newlines
+			end := size
+			for end > 0 && data[end-1] == '\n' {
+				end--
+			}
+			block := p.addBlock(HTMLBlock, data[:end])
+			finalizeHTMLBlock(block)
+		}
+		return size
+	}
+	return 0
+}
+
+// HR, which is the only self-closing block tag considered
+func (p *Markdown) htmlHr(data []byte, doRender bool) int {
+	if len(data) < 4 {
+		return 0
+	}
+	if data[0] != '<' || (data[1] != 'h' && data[1] != 'H') || (data[2] != 'r' && data[2] != 'R') {
+		return 0
+	}
+	if data[3] != ' ' && data[3] != '/' && data[3] != '>' {
+		// not an <hr> tag after all; at least not a valid one
+		return 0
+	}
+	i := 3
+	for i < len(data) && data[i] != '>' && data[i] != '\n' {
+		i++
+	}
+	if i < len(data) && data[i] == '>' {
+		i++
+		if j := p.isEmpty(data[i:]); j > 0 {
+			size := i + j
+			if doRender {
+				// trim newlines
+				end := size
+				for end > 0 && data[end-1] == '\n' {
+					end--
+				}
+				finalizeHTMLBlock(p.addBlock(HTMLBlock, data[:end]))
+			}
+			return size
+		}
+	}
+	return 0
+}
+
+func (p *Markdown) htmlFindTag(data []byte) (string, bool) {
+	i := 0
+	for i < len(data) && isalnum(data[i]) {
+		i++
+	}
+	key := string(data[:i])
+	if _, ok := blockTags[key]; ok {
+		return key, true
+	}
+	return "", false
+}
+
+func (p *Markdown) htmlFindEnd(tag string, data []byte) int {
+	// assume data[0] == '<' && data[1] == '/' already tested
+	if tag == "hr" {
+		return 2
+	}
+	// check if tag is a match
+	closetag := []byte("</" + tag + ">")
+	if !bytes.HasPrefix(data, closetag) {
+		return 0
+	}
+	i := len(closetag)
+
+	// check that the rest of the line is blank
+	skip := 0
+	if skip = p.isEmpty(data[i:]); skip == 0 {
+		return 0
+	}
+	i += skip
+	skip = 0
+
+	if i >= len(data) {
+		return i
+	}
+
+	if p.extensions&LaxHTMLBlocks != 0 {
+		return i
+	}
+	if skip = p.isEmpty(data[i:]); skip == 0 {
+		// following line must be blank
+		return 0
+	}
+
+	return i + skip
+}
+
+func (*Markdown) isEmpty(data []byte) int {
+	// it is okay to call isEmpty on an empty buffer
+	if len(data) == 0 {
+		return 0
+	}
+
+	var i int
+	for i = 0; i < len(data) && data[i] != '\n'; i++ {
+		if data[i] != ' ' && data[i] != '\t' {
+			return 0
+		}
+	}
+	if i < len(data) && data[i] == '\n' {
+		i++
+	}
+	return i
+}
+
+func (*Markdown) isHRule(data []byte) bool {
+	i := 0
+
+	// skip up to three spaces
+	for i < 3 && data[i] == ' ' {
+		i++
+	}
+
+	// look at the hrule char
+	if data[i] != '*' && data[i] != '-' && data[i] != '_' {
+		return false
+	}
+	c := data[i]
+
+	// the whole line must be the char or whitespace
+	n := 0
+	for i < len(data) && data[i] != '\n' {
+		switch {
+		case data[i] == c:
+			n++
+		case data[i] != ' ':
+			return false
+		}
+		i++
+	}
+
+	return n >= 3
+}
+
+// isFenceLine checks if there's a fence line (e.g., ``` or ``` go) at the beginning of data,
+// and returns the end index if so, or 0 otherwise. It also returns the marker found.
+// If syntax is not nil, it gets set to the syntax specified in the fence line.
+func isFenceLine(data []byte, syntax *string, oldmarker string) (end int, marker string) {
+	i, size := 0, 0
+
+	// skip up to three spaces
+	for i < len(data) && i < 3 && data[i] == ' ' {
+		i++
+	}
+
+	// check for the marker characters: ~ or `
+	if i >= len(data) {
+		return 0, ""
+	}
+	if data[i] != '~' && data[i] != '`' {
+		return 0, ""
+	}
+
+	c := data[i]
+
+	// the whole line must be the same char or whitespace
+	for i < len(data) && data[i] == c {
+		size++
+		i++
+	}
+
+	// the marker char must occur at least 3 times
+	if size < 3 {
+		return 0, ""
+	}
+	marker = string(data[i-size : i])
+
+	// if this is the end marker, it must match the beginning marker
+	if oldmarker != "" && marker != oldmarker {
+		return 0, ""
+	}
+
+	// TODO(shurcooL): It's probably a good idea to simplify the 2 code paths here
+	// into one, always get the syntax, and discard it if the caller doesn't care.
+	if syntax != nil {
+		syn := 0
+		i = skipChar(data, i, ' ')
+
+		if i >= len(data) {
+			if i == len(data) {
+				return i, marker
+			}
+			return 0, ""
+		}
+
+		syntaxStart := i
+
+		if data[i] == '{' {
+			i++
+			syntaxStart++
+
+			for i < len(data) && data[i] != '}' && data[i] != '\n' {
+				syn++
+				i++
+			}
+
+			if i >= len(data) || data[i] != '}' {
+				return 0, ""
+			}
+
+			// strip all whitespace at the beginning and the end
+			// of the {} block
+			for syn > 0 && isspace(data[syntaxStart]) {
+				syntaxStart++
+				syn--
+			}
+
+			for syn > 0 && isspace(data[syntaxStart+syn-1]) {
+				syn--
+			}
+
+			i++
+		} else {
+			for i < len(data) && !isspace(data[i]) {
+				syn++
+				i++
+			}
+		}
+
+		*syntax = string(data[syntaxStart : syntaxStart+syn])
+	}
+
+	i = skipChar(data, i, ' ')
+	if i >= len(data) || data[i] != '\n' {
+		if i == len(data) {
+			return i, marker
+		}
+		return 0, ""
+	}
+	return i + 1, marker // Take newline into account.
+}
+
+// fencedCodeBlock returns the end index if data contains a fenced code block at the beginning,
+// or 0 otherwise. It writes to out if doRender is true, otherwise it has no side effects.
+// If doRender is true, a final newline is mandatory to recognize the fenced code block.
+func (p *Markdown) fencedCodeBlock(data []byte, doRender bool) int {
+	var syntax string
+	beg, marker := isFenceLine(data, &syntax, "")
+	if beg == 0 || beg >= len(data) {
+		return 0
+	}
+
+	var work bytes.Buffer
+	work.Write([]byte(syntax))
+	work.WriteByte('\n')
+
+	for {
+		// safe to assume beg < len(data)
+
+		// check for the end of the code block
+		fenceEnd, _ := isFenceLine(data[beg:], nil, marker)
+		if fenceEnd != 0 {
+			beg += fenceEnd
+			break
+		}
+
+		// copy the current line
+		end := skipUntilChar(data, beg, '\n') + 1
+
+		// did we reach the end of the buffer without a closing marker?
+		if end >= len(data) {
+			return 0
+		}
+
+		// verbatim copy to the working buffer
+		if doRender {
+			work.Write(data[beg:end])
+		}
+		beg = end
+	}
+
+	if doRender {
+		block := p.addBlock(CodeBlock, work.Bytes()) // TODO: get rid of temp buffer
+		block.IsFenced = true
+		finalizeCodeBlock(block)
+	}
+
+	return beg
+}
+
+func unescapeChar(str []byte) []byte {
+	if str[0] == '\\' {
+		return []byte{str[1]}
+	}
+	return []byte(html.UnescapeString(string(str)))
+}
+
+func unescapeString(str []byte) []byte {
+	if reBackslashOrAmp.Match(str) {
+		return reEntityOrEscapedChar.ReplaceAllFunc(str, unescapeChar)
+	}
+	return str
+}
+
+func finalizeCodeBlock(block *Node) {
+	if block.IsFenced {
+		newlinePos := bytes.IndexByte(block.content, '\n')
+		firstLine := block.content[:newlinePos]
+		rest := block.content[newlinePos+1:]
+		block.Info = unescapeString(bytes.Trim(firstLine, "\n"))
+		block.Literal = rest
+	} else {
+		block.Literal = block.content
+	}
+	block.content = nil
+}
+
+func (p *Markdown) table(data []byte) int {
+	table := p.addBlock(Table, nil)
+	i, columns := p.tableHeader(data)
+	if i == 0 {
+		p.tip = table.Parent
+		table.Unlink()
+		return 0
+	}
+
+	p.addBlock(TableBody, nil)
+
+	for i < len(data) {
+		pipes, rowStart := 0, i
+		for ; i < len(data) && data[i] != '\n'; i++ {
+			if data[i] == '|' {
+				pipes++
+			}
+		}
+
+		if pipes == 0 {
+			i = rowStart
+			break
+		}
+
+		// include the newline in data sent to tableRow
+		if i < len(data) && data[i] == '\n' {
+			i++
+		}
+		p.tableRow(data[rowStart:i], columns, false)
+	}
+
+	return i
+}
+
+// check if the specified position is preceded by an odd number of backslashes
+func isBackslashEscaped(data []byte, i int) bool {
+	backslashes := 0
+	for i-backslashes-1 >= 0 && data[i-backslashes-1] == '\\' {
+		backslashes++
+	}
+	return backslashes&1 == 1
+}
+
+func (p *Markdown) tableHeader(data []byte) (size int, columns []CellAlignFlags) {
+	i := 0
+	colCount := 1
+	for i = 0; i < len(data) && data[i] != '\n'; i++ {
+		if data[i] == '|' && !isBackslashEscaped(data, i) {
+			colCount++
+		}
+	}
+
+	// doesn't look like a table header
+	if colCount == 1 {
+		return
+	}
+
+	// include the newline in the data sent to tableRow
+	j := i
+	if j < len(data) && data[j] == '\n' {
+		j++
+	}
+	header := data[:j]
+
+	// column count ignores pipes at beginning or end of line
+	if data[0] == '|' {
+		colCount--
+	}
+	if i > 2 && data[i-1] == '|' && !isBackslashEscaped(data, i-1) {
+		colCount--
+	}
+
+	columns = make([]CellAlignFlags, colCount)
+
+	// move on to the header underline
+	i++
+	if i >= len(data) {
+		return
+	}
+
+	if data[i] == '|' && !isBackslashEscaped(data, i) {
+		i++
+	}
+	i = skipChar(data, i, ' ')
+
+	// each column header is of form: / *:?-+:? *|/ with # dashes + # colons >= 3
+	// and trailing | optional on last column
+	col := 0
+	for i < len(data) && data[i] != '\n' {
+		dashes := 0
+
+		if data[i] == ':' {
+			i++
+			columns[col] |= TableAlignmentLeft
+			dashes++
+		}
+		for i < len(data) && data[i] == '-' {
+			i++
+			dashes++
+		}
+		if i < len(data) && data[i] == ':' {
+			i++
+			columns[col] |= TableAlignmentRight
+			dashes++
+		}
+		for i < len(data) && data[i] == ' ' {
+			i++
+		}
+		if i == len(data) {
+			return
+		}
+		// end of column test is messy
+		switch {
+		case dashes < 3:
+			// not a valid column
+			return
+
+		case data[i] == '|' && !isBackslashEscaped(data, i):
+			// marker found, now skip past trailing whitespace
+			col++
+			i++
+			for i < len(data) && data[i] == ' ' {
+				i++
+			}
+
+			// trailing junk found after last column
+			if col >= colCount && i < len(data) && data[i] != '\n' {
+				return
+			}
+
+		case (data[i] != '|' || isBackslashEscaped(data, i)) && col+1 < colCount:
+			// something else found where marker was required
+			return
+
+		case data[i] == '\n':
+			// marker is optional for the last column
+			col++
+
+		default:
+			// trailing junk found after last column
+			return
+		}
+	}
+	if col != colCount {
+		return
+	}
+
+	p.addBlock(TableHead, nil)
+	p.tableRow(header, columns, true)
+	size = i
+	if size < len(data) && data[size] == '\n' {
+		size++
+	}
+	return
+}
+
+func (p *Markdown) tableRow(data []byte, columns []CellAlignFlags, header bool) {
+	p.addBlock(TableRow, nil)
+	i, col := 0, 0
+
+	if data[i] == '|' && !isBackslashEscaped(data, i) {
+		i++
+	}
+
+	for col = 0; col < len(columns) && i < len(data); col++ {
+		for i < len(data) && data[i] == ' ' {
+			i++
+		}
+
+		cellStart := i
+
+		for i < len(data) && (data[i] != '|' || isBackslashEscaped(data, i)) && data[i] != '\n' {
+			i++
+		}
+
+		cellEnd := i
+
+		// skip the end-of-cell marker, possibly taking us past end of buffer
+		i++
+
+		for cellEnd > cellStart && cellEnd-1 < len(data) && data[cellEnd-1] == ' ' {
+			cellEnd--
+		}
+
+		cell := p.addBlock(TableCell, data[cellStart:cellEnd])
+		cell.IsHeader = header
+		cell.Align = columns[col]
+	}
+
+	// pad it out with empty columns to get the right number
+	for ; col < len(columns); col++ {
+		cell := p.addBlock(TableCell, nil)
+		cell.IsHeader = header
+		cell.Align = columns[col]
+	}
+
+	// silently ignore rows with too many cells
+}
+
+// returns blockquote prefix length
+func (p *Markdown) quotePrefix(data []byte) int {
+	i := 0
+	for i < 3 && i < len(data) && data[i] == ' ' {
+		i++
+	}
+	if i < len(data) && data[i] == '>' {
+		if i+1 < len(data) && data[i+1] == ' ' {
+			return i + 2
+		}
+		return i + 1
+	}
+	return 0
+}
+
+// blockquote ends with at least one blank line
+// followed by something without a blockquote prefix
+func (p *Markdown) terminateBlockquote(data []byte, beg, end int) bool {
+	if p.isEmpty(data[beg:]) <= 0 {
+		return false
+	}
+	if end >= len(data) {
+		return true
+	}
+	return p.quotePrefix(data[end:]) == 0 && p.isEmpty(data[end:]) == 0
+}
+
+// parse a blockquote fragment
+func (p *Markdown) quote(data []byte) int {
+	block := p.addBlock(BlockQuote, nil)
+	var raw bytes.Buffer
+	beg, end := 0, 0
+	for beg < len(data) {
+		end = beg
+		// Step over whole lines, collecting them. While doing that, check for
+		// fenced code and if one's found, incorporate it altogether,
+		// irregardless of any contents inside it
+		for end < len(data) && data[end] != '\n' {
+			if p.extensions&FencedCode != 0 {
+				if i := p.fencedCodeBlock(data[end:], false); i > 0 {
+					// -1 to compensate for the extra end++ after the loop:
+					end += i - 1
+					break
+				}
+			}
+			end++
+		}
+		if end < len(data) && data[end] == '\n' {
+			end++
+		}
+		if pre := p.quotePrefix(data[beg:]); pre > 0 {
+			// skip the prefix
+			beg += pre
+		} else if p.terminateBlockquote(data, beg, end) {
+			break
+		}
+		// this line is part of the blockquote
+		raw.Write(data[beg:end])
+		beg = end
+	}
+	p.block(raw.Bytes())
+	p.finalize(block)
+	return end
+}
+
+// returns prefix length for block code
+func (p *Markdown) codePrefix(data []byte) int {
+	if len(data) >= 1 && data[0] == '\t' {
+		return 1
+	}
+	if len(data) >= 4 && data[0] == ' ' && data[1] == ' ' && data[2] == ' ' && data[3] == ' ' {
+		return 4
+	}
+	return 0
+}
+
+func (p *Markdown) code(data []byte) int {
+	var work bytes.Buffer
+
+	i := 0
+	for i < len(data) {
+		beg := i
+		for i < len(data) && data[i] != '\n' {
+			i++
+		}
+		if i < len(data) && data[i] == '\n' {
+			i++
+		}
+
+		blankline := p.isEmpty(data[beg:i]) > 0
+		if pre := p.codePrefix(data[beg:i]); pre > 0 {
+			beg += pre
+		} else if !blankline {
+			// non-empty, non-prefixed line breaks the pre
+			i = beg
+			break
+		}
+
+		// verbatim copy to the working buffer
+		if blankline {
+			work.WriteByte('\n')
+		} else {
+			work.Write(data[beg:i])
+		}
+	}
+
+	// trim all the \n off the end of work
+	workbytes := work.Bytes()
+	eol := len(workbytes)
+	for eol > 0 && workbytes[eol-1] == '\n' {
+		eol--
+	}
+	if eol != len(workbytes) {
+		work.Truncate(eol)
+	}
+
+	work.WriteByte('\n')
+
+	block := p.addBlock(CodeBlock, work.Bytes()) // TODO: get rid of temp buffer
+	block.IsFenced = false
+	finalizeCodeBlock(block)
+
+	return i
+}
+
+// returns unordered list item prefix
+func (p *Markdown) uliPrefix(data []byte) int {
+	i := 0
+	// start with up to 3 spaces
+	for i < len(data) && i < 3 && data[i] == ' ' {
+		i++
+	}
+	if i >= len(data)-1 {
+		return 0
+	}
+	// need one of {'*', '+', '-'} followed by a space or a tab
+	if (data[i] != '*' && data[i] != '+' && data[i] != '-') ||
+		(data[i+1] != ' ' && data[i+1] != '\t') {
+		return 0
+	}
+	return i + 2
+}
+
+// returns ordered list item prefix
+func (p *Markdown) oliPrefix(data []byte) int {
+	i := 0
+
+	// start with up to 3 spaces
+	for i < 3 && i < len(data) && data[i] == ' ' {
+		i++
+	}
+
+	// count the digits
+	start := i
+	for i < len(data) && data[i] >= '0' && data[i] <= '9' {
+		i++
+	}
+	if start == i || i >= len(data)-1 {
+		return 0
+	}
+
+	// we need >= 1 digits followed by a dot and a space or a tab
+	if data[i] != '.' || !(data[i+1] == ' ' || data[i+1] == '\t') {
+		return 0
+	}
+	return i + 2
+}
+
+// returns definition list item prefix
+func (p *Markdown) dliPrefix(data []byte) int {
+	if len(data) < 2 {
+		return 0
+	}
+	i := 0
+	// need a ':' followed by a space or a tab
+	if data[i] != ':' || !(data[i+1] == ' ' || data[i+1] == '\t') {
+		return 0
+	}
+	for i < len(data) && data[i] == ' ' {
+		i++
+	}
+	return i + 2
+}
+
+// parse ordered or unordered list block
+func (p *Markdown) list(data []byte, flags ListType) int {
+	i := 0
+	flags |= ListItemBeginningOfList
+	block := p.addBlock(List, nil)
+	block.ListFlags = flags
+	block.Tight = true
+
+	for i < len(data) {
+		skip := p.listItem(data[i:], &flags)
+		if flags&ListItemContainsBlock != 0 {
+			block.ListData.Tight = false
+		}
+		i += skip
+		if skip == 0 || flags&ListItemEndOfList != 0 {
+			break
+		}
+		flags &= ^ListItemBeginningOfList
+	}
+
+	above := block.Parent
+	finalizeList(block)
+	p.tip = above
+	return i
+}
+
+// Returns true if block ends with a blank line, descending if needed
+// into lists and sublists.
+func endsWithBlankLine(block *Node) bool {
+	// TODO: figure this out. Always false now.
+	for block != nil {
+		//if block.lastLineBlank {
+		//return true
+		//}
+		t := block.Type
+		if t == List || t == Item {
+			block = block.LastChild
+		} else {
+			break
+		}
+	}
+	return false
+}
+
+func finalizeList(block *Node) {
+	block.open = false
+	item := block.FirstChild
+	for item != nil {
+		// check for non-final list item ending with blank line:
+		if endsWithBlankLine(item) && item.Next != nil {
+			block.ListData.Tight = false
+			break
+		}
+		// recurse into children of list item, to see if there are spaces
+		// between any of them:
+		subItem := item.FirstChild
+		for subItem != nil {
+			if endsWithBlankLine(subItem) && (item.Next != nil || subItem.Next != nil) {
+				block.ListData.Tight = false
+				break
+			}
+			subItem = subItem.Next
+		}
+		item = item.Next
+	}
+}
+
+// Parse a single list item.
+// Assumes initial prefix is already removed if this is a sublist.
+func (p *Markdown) listItem(data []byte, flags *ListType) int {
+	// keep track of the indentation of the first line
+	itemIndent := 0
+	if data[0] == '\t' {
+		itemIndent += 4
+	} else {
+		for itemIndent < 3 && data[itemIndent] == ' ' {
+			itemIndent++
+		}
+	}
+
+	var bulletChar byte = '*'
+	i := p.uliPrefix(data)
+	if i == 0 {
+		i = p.oliPrefix(data)
+	} else {
+		bulletChar = data[i-2]
+	}
+	if i == 0 {
+		i = p.dliPrefix(data)
+		// reset definition term flag
+		if i > 0 {
+			*flags &= ^ListTypeTerm
+		}
+	}
+	if i == 0 {
+		// if in definition list, set term flag and continue
+		if *flags&ListTypeDefinition != 0 {
+			*flags |= ListTypeTerm
+		} else {
+			return 0
+		}
+	}
+
+	// skip leading whitespace on first line
+	for i < len(data) && data[i] == ' ' {
+		i++
+	}
+
+	// find the end of the line
+	line := i
+	for i > 0 && i < len(data) && data[i-1] != '\n' {
+		i++
+	}
+
+	// get working buffer
+	var raw bytes.Buffer
+
+	// put the first line into the working buffer
+	raw.Write(data[line:i])
+	line = i
+
+	// process the following lines
+	containsBlankLine := false
+	sublist := 0
+
+gatherlines:
+	for line < len(data) {
+		i++
+
+		// find the end of this line
+		for i < len(data) && data[i-1] != '\n' {
+			i++
+		}
+
+		// if it is an empty line, guess that it is part of this item
+		// and move on to the next line
+		if p.isEmpty(data[line:i]) > 0 {
+			containsBlankLine = true
+			line = i
+			continue
+		}
+
+		// calculate the indentation
+		indent := 0
+		indentIndex := 0
+		if data[line] == '\t' {
+			indentIndex++
+			indent += 4
+		} else {
+			for indent < 4 && line+indent < i && data[line+indent] == ' ' {
+				indent++
+				indentIndex++
+			}
+		}
+
+		chunk := data[line+indentIndex : i]
+
+		// evaluate how this line fits in
+		switch {
+		// is this a nested list item?
+		case (p.uliPrefix(chunk) > 0 && !p.isHRule(chunk)) ||
+			p.oliPrefix(chunk) > 0 ||
+			p.dliPrefix(chunk) > 0:
+
+			if containsBlankLine {
+				*flags |= ListItemContainsBlock
+			}
+
+			// to be a nested list, it must be indented more
+			// if not, it is the next item in the same list
+			if indent <= itemIndent {
+				break gatherlines
+			}
+
+			// is this the first item in the nested list?
+			if sublist == 0 {
+				sublist = raw.Len()
+			}
+
+		// is this a nested prefix heading?
+		case p.isPrefixHeading(chunk):
+			// if the heading is not indented, it is not nested in the list
+			// and thus ends the list
+			if containsBlankLine && indent < 4 {
+				*flags |= ListItemEndOfList
+				break gatherlines
+			}
+			*flags |= ListItemContainsBlock
+
+		// anything following an empty line is only part
+		// of this item if it is indented 4 spaces
+		// (regardless of the indentation of the beginning of the item)
+		case containsBlankLine && indent < 4:
+			if *flags&ListTypeDefinition != 0 && i < len(data)-1 {
+				// is the next item still a part of this list?
+				next := i
+				for next < len(data) && data[next] != '\n' {
+					next++
+				}
+				for next < len(data)-1 && data[next] == '\n' {
+					next++
+				}
+				if i < len(data)-1 && data[i] != ':' && data[next] != ':' {
+					*flags |= ListItemEndOfList
+				}
+			} else {
+				*flags |= ListItemEndOfList
+			}
+			break gatherlines
+
+		// a blank line means this should be parsed as a block
+		case containsBlankLine:
+			raw.WriteByte('\n')
+			*flags |= ListItemContainsBlock
+		}
+
+		// if this line was preceded by one or more blanks,
+		// re-introduce the blank into the buffer
+		if containsBlankLine {
+			containsBlankLine = false
+			raw.WriteByte('\n')
+		}
+
+		// add the line into the working buffer without prefix
+		raw.Write(data[line+indentIndex : i])
+
+		line = i
+	}
+
+	rawBytes := raw.Bytes()
+
+	block := p.addBlock(Item, nil)
+	block.ListFlags = *flags
+	block.Tight = false
+	block.BulletChar = bulletChar
+	block.Delimiter = '.' // Only '.' is possible in Markdown, but ')' will also be possible in CommonMark
+
+	// render the contents of the list item
+	if *flags&ListItemContainsBlock != 0 && *flags&ListTypeTerm == 0 {
+		// intermediate render of block item, except for definition term
+		if sublist > 0 {
+			p.block(rawBytes[:sublist])
+			p.block(rawBytes[sublist:])
+		} else {
+			p.block(rawBytes)
+		}
+	} else {
+		// intermediate render of inline item
+		if sublist > 0 {
+			child := p.addChild(Paragraph, 0)
+			child.content = rawBytes[:sublist]
+			p.block(rawBytes[sublist:])
+		} else {
+			child := p.addChild(Paragraph, 0)
+			child.content = rawBytes
+		}
+	}
+	return line
+}
+
+// render a single paragraph that has already been parsed out
+func (p *Markdown) renderParagraph(data []byte) {
+	if len(data) == 0 {
+		return
+	}
+
+	// trim leading spaces
+	beg := 0
+	for data[beg] == ' ' {
+		beg++
+	}
+
+	end := len(data)
+	// trim trailing newline
+	if data[len(data)-1] == '\n' {
+		end--
+	}
+
+	// trim trailing spaces
+	for end > beg && data[end-1] == ' ' {
+		end--
+	}
+
+	p.addBlock(Paragraph, data[beg:end])
+}
+
+func (p *Markdown) paragraph(data []byte) int {
+	// prev: index of 1st char of previous line
+	// line: index of 1st char of current line
+	// i: index of cursor/end of current line
+	var prev, line, i int
+	tabSize := TabSizeDefault
+	if p.extensions&TabSizeEight != 0 {
+		tabSize = TabSizeDouble
+	}
+	// keep going until we find something to mark the end of the paragraph
+	for i < len(data) {
+		// mark the beginning of the current line
+		prev = line
+		current := data[i:]
+		line = i
+
+		// did we find a reference or a footnote? If so, end a paragraph
+		// preceding it and report that we have consumed up to the end of that
+		// reference:
+		if refEnd := isReference(p, current, tabSize); refEnd > 0 {
+			p.renderParagraph(data[:i])
+			return i + refEnd
+		}
+
+		// did we find a blank line marking the end of the paragraph?
+		if n := p.isEmpty(current); n > 0 {
+			// did this blank line followed by a definition list item?
+			if p.extensions&DefinitionLists != 0 {
+				if i < len(data)-1 && data[i+1] == ':' {
+					return p.list(data[prev:], ListTypeDefinition)
+				}
+			}
+
+			p.renderParagraph(data[:i])
+			return i + n
+		}
+
+		// an underline under some text marks a heading, so our paragraph ended on prev line
+		if i > 0 {
+			if level := p.isUnderlinedHeading(current); level > 0 {
+				// render the paragraph
+				p.renderParagraph(data[:prev])
+
+				// ignore leading and trailing whitespace
+				eol := i - 1
+				for prev < eol && data[prev] == ' ' {
+					prev++
+				}
+				for eol > prev && data[eol-1] == ' ' {
+					eol--
+				}
+
+				id := ""
+				if p.extensions&AutoHeadingIDs != 0 {
+					id = sanitized_anchor_name.Create(string(data[prev:eol]))
+				}
+
+				block := p.addBlock(Heading, data[prev:eol])
+				block.Level = level
+				block.HeadingID = id
+
+				// find the end of the underline
+				for i < len(data) && data[i] != '\n' {
+					i++
+				}
+				return i
+			}
+		}
+
+		// if the next line starts a block of HTML, then the paragraph ends here
+		if p.extensions&LaxHTMLBlocks != 0 {
+			if data[i] == '<' && p.html(current, false) > 0 {
+				// rewind to before the HTML block
+				p.renderParagraph(data[:i])
+				return i
+			}
+		}
+
+		// if there's a prefixed heading or a horizontal rule after this, paragraph is over
+		if p.isPrefixHeading(current) || p.isHRule(current) {
+			p.renderParagraph(data[:i])
+			return i
+		}
+
+		// if there's a fenced code block, paragraph is over
+		if p.extensions&FencedCode != 0 {
+			if p.fencedCodeBlock(current, false) > 0 {
+				p.renderParagraph(data[:i])
+				return i
+			}
+		}
+
+		// if there's a definition list item, prev line is a definition term
+		if p.extensions&DefinitionLists != 0 {
+			if p.dliPrefix(current) != 0 {
+				ret := p.list(data[prev:], ListTypeDefinition)
+				return ret
+			}
+		}
+
+		// if there's a list after this, paragraph is over
+		if p.extensions&NoEmptyLineBeforeBlock != 0 {
+			if p.uliPrefix(current) != 0 ||
+				p.oliPrefix(current) != 0 ||
+				p.quotePrefix(current) != 0 ||
+				p.codePrefix(current) != 0 {
+				p.renderParagraph(data[:i])
+				return i
+			}
+		}
+
+		// otherwise, scan to the beginning of the next line
+		nl := bytes.IndexByte(data[i:], '\n')
+		if nl >= 0 {
+			i += nl + 1
+		} else {
+			i += len(data[i:])
+		}
+	}
+
+	p.renderParagraph(data[:i])
+	return i
+}
+
+func skipChar(data []byte, start int, char byte) int {
+	i := start
+	for i < len(data) && data[i] == char {
+		i++
+	}
+	return i
+}
+
+func skipUntilChar(text []byte, start int, char byte) int {
+	i := start
+	for i < len(text) && text[i] != char {
+		i++
+	}
+	return i
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/doc.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/doc.go
new file mode 100644
index 00000000..5b3fa987
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/doc.go
@@ -0,0 +1,18 @@
+// Package blackfriday is a markdown processor.
+//
+// It translates plain text with simple formatting rules into an AST, which can
+// then be further processed to HTML (provided by Blackfriday itself) or other
+// formats (provided by the community).
+//
+// The simplest way to invoke Blackfriday is to call the Run function. It will
+// take a text input and produce a text output in HTML (or other format).
+//
+// A slightly more sophisticated way to use Blackfriday is to create a Markdown
+// processor and to call Parse, which returns a syntax tree for the input
+// document. You can leverage Blackfriday's parsing for content extraction from
+// markdown documents. You can assign a custom renderer and set various options
+// to the Markdown processor.
+//
+// If you're interested in calling Blackfriday from command line, see
+// https://github.com/russross/blackfriday-tool.
+package blackfriday
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/esc.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/esc.go
new file mode 100644
index 00000000..6385f27c
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/esc.go
@@ -0,0 +1,34 @@
+package blackfriday
+
+import (
+	"html"
+	"io"
+)
+
+var htmlEscaper = [256][]byte{
+	'&': []byte("&amp;"),
+	'<': []byte("&lt;"),
+	'>': []byte("&gt;"),
+	'"': []byte("&quot;"),
+}
+
+func escapeHTML(w io.Writer, s []byte) {
+	var start, end int
+	for end < len(s) {
+		escSeq := htmlEscaper[s[end]]
+		if escSeq != nil {
+			w.Write(s[start:end])
+			w.Write(escSeq)
+			start = end + 1
+		}
+		end++
+	}
+	if start < len(s) && end <= len(s) {
+		w.Write(s[start:end])
+	}
+}
+
+func escLink(w io.Writer, text []byte) {
+	unesc := html.UnescapeString(string(text))
+	escapeHTML(w, []byte(unesc))
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/html.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/html.go
new file mode 100644
index 00000000..25fb185e
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/html.go
@@ -0,0 +1,940 @@
+//
+// Blackfriday Markdown Processor
+// Available at http://github.com/russross/blackfriday
+//
+// Copyright © 2011 Russ Ross <russ@russross.com>.
+// Distributed under the Simplified BSD License.
+// See README.md for details.
+//
+
+//
+//
+// HTML rendering backend
+//
+//
+
+package blackfriday
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"regexp"
+	"strings"
+)
+
+// HTMLFlags control optional behavior of HTML renderer.
+type HTMLFlags int
+
+// HTML renderer configuration options.
+const (
+	HTMLFlagsNone           HTMLFlags = 0
+	SkipHTML                HTMLFlags = 1 << iota // Skip preformatted HTML blocks
+	SkipImages                                    // Skip embedded images
+	SkipLinks                                     // Skip all links
+	Safelink                                      // Only link to trusted protocols
+	NofollowLinks                                 // Only link with rel="nofollow"
+	NoreferrerLinks                               // Only link with rel="noreferrer"
+	HrefTargetBlank                               // Add a blank target
+	CompletePage                                  // Generate a complete HTML page
+	UseXHTML                                      // Generate XHTML output instead of HTML
+	FootnoteReturnLinks                           // Generate a link at the end of a footnote to return to the source
+	Smartypants                                   // Enable smart punctuation substitutions
+	SmartypantsFractions                          // Enable smart fractions (with Smartypants)
+	SmartypantsDashes                             // Enable smart dashes (with Smartypants)
+	SmartypantsLatexDashes                        // Enable LaTeX-style dashes (with Smartypants)
+	SmartypantsAngledQuotes                       // Enable angled double quotes (with Smartypants) for double quotes rendering
+	SmartypantsQuotesNBSP                         // Enable « French guillemets » (with Smartypants)
+	TOC                                           // Generate a table of contents
+)
+
+var (
+	htmlTagRe = regexp.MustCompile("(?i)^" + htmlTag)
+)
+
+const (
+	htmlTag = "(?:" + openTag + "|" + closeTag + "|" + htmlComment + "|" +
+		processingInstruction + "|" + declaration + "|" + cdata + ")"
+	closeTag              = "</" + tagName + "\\s*[>]"
+	openTag               = "<" + tagName + attribute + "*" + "\\s*/?>"
+	attribute             = "(?:" + "\\s+" + attributeName + attributeValueSpec + "?)"
+	attributeValue        = "(?:" + unquotedValue + "|" + singleQuotedValue + "|" + doubleQuotedValue + ")"
+	attributeValueSpec    = "(?:" + "\\s*=" + "\\s*" + attributeValue + ")"
+	attributeName         = "[a-zA-Z_:][a-zA-Z0-9:._-]*"
+	cdata                 = "<!\\[CDATA\\[[\\s\\S]*?\\]\\]>"
+	declaration           = "<![A-Z]+" + "\\s+[^>]*>"
+	doubleQuotedValue     = "\"[^\"]*\""
+	htmlComment           = "<!---->|<!--(?:-?[^>-])(?:-?[^-])*-->"
+	processingInstruction = "[<][?].*?[?][>]"
+	singleQuotedValue     = "'[^']*'"
+	tagName               = "[A-Za-z][A-Za-z0-9-]*"
+	unquotedValue         = "[^\"'=<>`\\x00-\\x20]+"
+)
+
+// HTMLRendererParameters is a collection of supplementary parameters tweaking
+// the behavior of various parts of HTML renderer.
+type HTMLRendererParameters struct {
+	// Prepend this text to each relative URL.
+	AbsolutePrefix string
+	// Add this text to each footnote anchor, to ensure uniqueness.
+	FootnoteAnchorPrefix string
+	// Show this text inside the <a> tag for a footnote return link, if the
+	// HTML_FOOTNOTE_RETURN_LINKS flag is enabled. If blank, the string
+	// <sup>[return]</sup> is used.
+	FootnoteReturnLinkContents string
+	// If set, add this text to the front of each Heading ID, to ensure
+	// uniqueness.
+	HeadingIDPrefix string
+	// If set, add this text to the back of each Heading ID, to ensure uniqueness.
+	HeadingIDSuffix string
+
+	Title string // Document title (used if CompletePage is set)
+	CSS   string // Optional CSS file URL (used if CompletePage is set)
+	Icon  string // Optional icon file URL (used if CompletePage is set)
+
+	Flags HTMLFlags // Flags allow customizing this renderer's behavior
+}
+
+// HTMLRenderer is a type that implements the Renderer interface for HTML output.
+//
+// Do not create this directly, instead use the NewHTMLRenderer function.
+type HTMLRenderer struct {
+	HTMLRendererParameters
+
+	closeTag string // how to end singleton tags: either " />" or ">"
+
+	// Track heading IDs to prevent ID collision in a single generation.
+	headingIDs map[string]int
+
+	lastOutputLen int
+	disableTags   int
+
+	sr *SPRenderer
+}
+
+const (
+	xhtmlClose = " />"
+	htmlClose  = ">"
+)
+
+// NewHTMLRenderer creates and configures an HTMLRenderer object, which
+// satisfies the Renderer interface.
+func NewHTMLRenderer(params HTMLRendererParameters) *HTMLRenderer {
+	// configure the rendering engine
+	closeTag := htmlClose
+	if params.Flags&UseXHTML != 0 {
+		closeTag = xhtmlClose
+	}
+
+	if params.FootnoteReturnLinkContents == "" {
+		params.FootnoteReturnLinkContents = `<sup>[return]</sup>`
+	}
+
+	return &HTMLRenderer{
+		HTMLRendererParameters: params,
+
+		closeTag:   closeTag,
+		headingIDs: make(map[string]int),
+
+		sr: NewSmartypantsRenderer(params.Flags),
+	}
+}
+
+func isHTMLTag(tag []byte, tagname string) bool {
+	found, _ := findHTMLTagPos(tag, tagname)
+	return found
+}
+
+// Look for a character, but ignore it when it's in any kind of quotes, it
+// might be JavaScript
+func skipUntilCharIgnoreQuotes(html []byte, start int, char byte) int {
+	inSingleQuote := false
+	inDoubleQuote := false
+	inGraveQuote := false
+	i := start
+	for i < len(html) {
+		switch {
+		case html[i] == char && !inSingleQuote && !inDoubleQuote && !inGraveQuote:
+			return i
+		case html[i] == '\'':
+			inSingleQuote = !inSingleQuote
+		case html[i] == '"':
+			inDoubleQuote = !inDoubleQuote
+		case html[i] == '`':
+			inGraveQuote = !inGraveQuote
+		}
+		i++
+	}
+	return start
+}
+
+func findHTMLTagPos(tag []byte, tagname string) (bool, int) {
+	i := 0
+	if i < len(tag) && tag[0] != '<' {
+		return false, -1
+	}
+	i++
+	i = skipSpace(tag, i)
+
+	if i < len(tag) && tag[i] == '/' {
+		i++
+	}
+
+	i = skipSpace(tag, i)
+	j := 0
+	for ; i < len(tag); i, j = i+1, j+1 {
+		if j >= len(tagname) {
+			break
+		}
+
+		if strings.ToLower(string(tag[i]))[0] != tagname[j] {
+			return false, -1
+		}
+	}
+
+	if i == len(tag) {
+		return false, -1
+	}
+
+	rightAngle := skipUntilCharIgnoreQuotes(tag, i, '>')
+	if rightAngle >= i {
+		return true, rightAngle
+	}
+
+	return false, -1
+}
+
+func skipSpace(tag []byte, i int) int {
+	for i < len(tag) && isspace(tag[i]) {
+		i++
+	}
+	return i
+}
+
+func isRelativeLink(link []byte) (yes bool) {
+	// a tag begin with '#'
+	if link[0] == '#' {
+		return true
+	}
+
+	// link begin with '/' but not '//', the second maybe a protocol relative link
+	if len(link) >= 2 && link[0] == '/' && link[1] != '/' {
+		return true
+	}
+
+	// only the root '/'
+	if len(link) == 1 && link[0] == '/' {
+		return true
+	}
+
+	// current directory : begin with "./"
+	if bytes.HasPrefix(link, []byte("./")) {
+		return true
+	}
+
+	// parent directory : begin with "../"
+	if bytes.HasPrefix(link, []byte("../")) {
+		return true
+	}
+
+	return false
+}
+
+func (r *HTMLRenderer) ensureUniqueHeadingID(id string) string {
+	for count, found := r.headingIDs[id]; found; count, found = r.headingIDs[id] {
+		tmp := fmt.Sprintf("%s-%d", id, count+1)
+
+		if _, tmpFound := r.headingIDs[tmp]; !tmpFound {
+			r.headingIDs[id] = count + 1
+			id = tmp
+		} else {
+			id = id + "-1"
+		}
+	}
+
+	if _, found := r.headingIDs[id]; !found {
+		r.headingIDs[id] = 0
+	}
+
+	return id
+}
+
+func (r *HTMLRenderer) addAbsPrefix(link []byte) []byte {
+	if r.AbsolutePrefix != "" && isRelativeLink(link) && link[0] != '.' {
+		newDest := r.AbsolutePrefix
+		if link[0] != '/' {
+			newDest += "/"
+		}
+		newDest += string(link)
+		return []byte(newDest)
+	}
+	return link
+}
+
+func appendLinkAttrs(attrs []string, flags HTMLFlags, link []byte) []string {
+	if isRelativeLink(link) {
+		return attrs
+	}
+	val := []string{}
+	if flags&NofollowLinks != 0 {
+		val = append(val, "nofollow")
+	}
+	if flags&NoreferrerLinks != 0 {
+		val = append(val, "noreferrer")
+	}
+	if flags&HrefTargetBlank != 0 {
+		attrs = append(attrs, "target=\"_blank\"")
+	}
+	if len(val) == 0 {
+		return attrs
+	}
+	attr := fmt.Sprintf("rel=%q", strings.Join(val, " "))
+	return append(attrs, attr)
+}
+
+func isMailto(link []byte) bool {
+	return bytes.HasPrefix(link, []byte("mailto:"))
+}
+
+func needSkipLink(flags HTMLFlags, dest []byte) bool {
+	if flags&SkipLinks != 0 {
+		return true
+	}
+	return flags&Safelink != 0 && !isSafeLink(dest) && !isMailto(dest)
+}
+
+func isSmartypantable(node *Node) bool {
+	pt := node.Parent.Type
+	return pt != Link && pt != CodeBlock && pt != Code
+}
+
+func appendLanguageAttr(attrs []string, info []byte) []string {
+	if len(info) == 0 {
+		return attrs
+	}
+	endOfLang := bytes.IndexAny(info, "\t ")
+	if endOfLang < 0 {
+		endOfLang = len(info)
+	}
+	return append(attrs, fmt.Sprintf("class=\"language-%s\"", info[:endOfLang]))
+}
+
+func (r *HTMLRenderer) tag(w io.Writer, name []byte, attrs []string) {
+	w.Write(name)
+	if len(attrs) > 0 {
+		w.Write(spaceBytes)
+		w.Write([]byte(strings.Join(attrs, " ")))
+	}
+	w.Write(gtBytes)
+	r.lastOutputLen = 1
+}
+
+func footnoteRef(prefix string, node *Node) []byte {
+	urlFrag := prefix + string(slugify(node.Destination))
+	anchor := fmt.Sprintf(`<a rel="footnote" href="#fn:%s">%d</a>`, urlFrag, node.NoteID)
+	return []byte(fmt.Sprintf(`<sup class="footnote-ref" id="fnref:%s">%s</sup>`, urlFrag, anchor))
+}
+
+func footnoteItem(prefix string, slug []byte) []byte {
+	return []byte(fmt.Sprintf(`<li id="fn:%s%s">`, prefix, slug))
+}
+
+func footnoteReturnLink(prefix, returnLink string, slug []byte) []byte {
+	const format = ` <a class="footnote-return" href="#fnref:%s%s">%s</a>`
+	return []byte(fmt.Sprintf(format, prefix, slug, returnLink))
+}
+
+func itemOpenCR(node *Node) bool {
+	if node.Prev == nil {
+		return false
+	}
+	ld := node.Parent.ListData
+	return !ld.Tight && ld.ListFlags&ListTypeDefinition == 0
+}
+
+func skipParagraphTags(node *Node) bool {
+	grandparent := node.Parent.Parent
+	if grandparent == nil || grandparent.Type != List {
+		return false
+	}
+	tightOrTerm := grandparent.Tight || node.Parent.ListFlags&ListTypeTerm != 0
+	return grandparent.Type == List && tightOrTerm
+}
+
+func cellAlignment(align CellAlignFlags) string {
+	switch align {
+	case TableAlignmentLeft:
+		return "left"
+	case TableAlignmentRight:
+		return "right"
+	case TableAlignmentCenter:
+		return "center"
+	default:
+		return ""
+	}
+}
+
+func (r *HTMLRenderer) out(w io.Writer, text []byte) {
+	if r.disableTags > 0 {
+		w.Write(htmlTagRe.ReplaceAll(text, []byte{}))
+	} else {
+		w.Write(text)
+	}
+	r.lastOutputLen = len(text)
+}
+
+func (r *HTMLRenderer) cr(w io.Writer) {
+	if r.lastOutputLen > 0 {
+		r.out(w, nlBytes)
+	}
+}
+
+var (
+	nlBytes    = []byte{'\n'}
+	gtBytes    = []byte{'>'}
+	spaceBytes = []byte{' '}
+)
+
+var (
+	brTag              = []byte("<br>")
+	brXHTMLTag         = []byte("<br />")
+	emTag              = []byte("<em>")
+	emCloseTag         = []byte("</em>")
+	strongTag          = []byte("<strong>")
+	strongCloseTag     = []byte("</strong>")
+	delTag             = []byte("<del>")
+	delCloseTag        = []byte("</del>")
+	ttTag              = []byte("<tt>")
+	ttCloseTag         = []byte("</tt>")
+	aTag               = []byte("<a")
+	aCloseTag          = []byte("</a>")
+	preTag             = []byte("<pre>")
+	preCloseTag        = []byte("</pre>")
+	codeTag            = []byte("<code>")
+	codeCloseTag       = []byte("</code>")
+	pTag               = []byte("<p>")
+	pCloseTag          = []byte("</p>")
+	blockquoteTag      = []byte("<blockquote>")
+	blockquoteCloseTag = []byte("</blockquote>")
+	hrTag              = []byte("<hr>")
+	hrXHTMLTag         = []byte("<hr />")
+	ulTag              = []byte("<ul>")
+	ulCloseTag         = []byte("</ul>")
+	olTag              = []byte("<ol>")
+	olCloseTag         = []byte("</ol>")
+	dlTag              = []byte("<dl>")
+	dlCloseTag         = []byte("</dl>")
+	liTag              = []byte("<li>")
+	liCloseTag         = []byte("</li>")
+	ddTag              = []byte("<dd>")
+	ddCloseTag         = []byte("</dd>")
+	dtTag              = []byte("<dt>")
+	dtCloseTag         = []byte("</dt>")
+	tableTag           = []byte("<table>")
+	tableCloseTag      = []byte("</table>")
+	tdTag              = []byte("<td")
+	tdCloseTag         = []byte("</td>")
+	thTag              = []byte("<th")
+	thCloseTag         = []byte("</th>")
+	theadTag           = []byte("<thead>")
+	theadCloseTag      = []byte("</thead>")
+	tbodyTag           = []byte("<tbody>")
+	tbodyCloseTag      = []byte("</tbody>")
+	trTag              = []byte("<tr>")
+	trCloseTag         = []byte("</tr>")
+	h1Tag              = []byte("<h1")
+	h1CloseTag         = []byte("</h1>")
+	h2Tag              = []byte("<h2")
+	h2CloseTag         = []byte("</h2>")
+	h3Tag              = []byte("<h3")
+	h3CloseTag         = []byte("</h3>")
+	h4Tag              = []byte("<h4")
+	h4CloseTag         = []byte("</h4>")
+	h5Tag              = []byte("<h5")
+	h5CloseTag         = []byte("</h5>")
+	h6Tag              = []byte("<h6")
+	h6CloseTag         = []byte("</h6>")
+
+	footnotesDivBytes      = []byte("\n<div class=\"footnotes\">\n\n")
+	footnotesCloseDivBytes = []byte("\n</div>\n")
+)
+
+func headingTagsFromLevel(level int) ([]byte, []byte) {
+	switch level {
+	case 1:
+		return h1Tag, h1CloseTag
+	case 2:
+		return h2Tag, h2CloseTag
+	case 3:
+		return h3Tag, h3CloseTag
+	case 4:
+		return h4Tag, h4CloseTag
+	case 5:
+		return h5Tag, h5CloseTag
+	default:
+		return h6Tag, h6CloseTag
+	}
+}
+
+func (r *HTMLRenderer) outHRTag(w io.Writer) {
+	if r.Flags&UseXHTML == 0 {
+		r.out(w, hrTag)
+	} else {
+		r.out(w, hrXHTMLTag)
+	}
+}
+
+// RenderNode is a default renderer of a single node of a syntax tree. For
+// block nodes it will be called twice: first time with entering=true, second
+// time with entering=false, so that it could know when it's working on an open
+// tag and when on close. It writes the result to w.
+//
+// The return value is a way to tell the calling walker to adjust its walk
+// pattern: e.g. it can terminate the traversal by returning Terminate. Or it
+// can ask the walker to skip a subtree of this node by returning SkipChildren.
+// The typical behavior is to return GoToNext, which asks for the usual
+// traversal to the next node.
+func (r *HTMLRenderer) RenderNode(w io.Writer, node *Node, entering bool) WalkStatus {
+	attrs := []string{}
+	switch node.Type {
+	case Text:
+		if r.Flags&Smartypants != 0 {
+			var tmp bytes.Buffer
+			escapeHTML(&tmp, node.Literal)
+			r.sr.Process(w, tmp.Bytes())
+		} else {
+			if node.Parent.Type == Link {
+				escLink(w, node.Literal)
+			} else {
+				escapeHTML(w, node.Literal)
+			}
+		}
+	case Softbreak:
+		r.cr(w)
+		// TODO: make it configurable via out(renderer.softbreak)
+	case Hardbreak:
+		if r.Flags&UseXHTML == 0 {
+			r.out(w, brTag)
+		} else {
+			r.out(w, brXHTMLTag)
+		}
+		r.cr(w)
+	case Emph:
+		if entering {
+			r.out(w, emTag)
+		} else {
+			r.out(w, emCloseTag)
+		}
+	case Strong:
+		if entering {
+			r.out(w, strongTag)
+		} else {
+			r.out(w, strongCloseTag)
+		}
+	case Del:
+		if entering {
+			r.out(w, delTag)
+		} else {
+			r.out(w, delCloseTag)
+		}
+	case HTMLSpan:
+		if r.Flags&SkipHTML != 0 {
+			break
+		}
+		r.out(w, node.Literal)
+	case Link:
+		// mark it but don't link it if it is not a safe link: no smartypants
+		dest := node.LinkData.Destination
+		if needSkipLink(r.Flags, dest) {
+			if entering {
+				r.out(w, ttTag)
+			} else {
+				r.out(w, ttCloseTag)
+			}
+		} else {
+			if entering {
+				dest = r.addAbsPrefix(dest)
+				var hrefBuf bytes.Buffer
+				hrefBuf.WriteString("href=\"")
+				escLink(&hrefBuf, dest)
+				hrefBuf.WriteByte('"')
+				attrs = append(attrs, hrefBuf.String())
+				if node.NoteID != 0 {
+					r.out(w, footnoteRef(r.FootnoteAnchorPrefix, node))
+					break
+				}
+				attrs = appendLinkAttrs(attrs, r.Flags, dest)
+				if len(node.LinkData.Title) > 0 {
+					var titleBuff bytes.Buffer
+					titleBuff.WriteString("title=\"")
+					escapeHTML(&titleBuff, node.LinkData.Title)
+					titleBuff.WriteByte('"')
+					attrs = append(attrs, titleBuff.String())
+				}
+				r.tag(w, aTag, attrs)
+			} else {
+				if node.NoteID != 0 {
+					break
+				}
+				r.out(w, aCloseTag)
+			}
+		}
+	case Image:
+		if r.Flags&SkipImages != 0 {
+			return SkipChildren
+		}
+		if entering {
+			dest := node.LinkData.Destination
+			dest = r.addAbsPrefix(dest)
+			if r.disableTags == 0 {
+				//if options.safe && potentiallyUnsafe(dest) {
+				//out(w, `<img src="" alt="`)
+				//} else {
+				r.out(w, []byte(`<img src="`))
+				escLink(w, dest)
+				r.out(w, []byte(`" alt="`))
+				//}
+			}
+			r.disableTags++
+		} else {
+			r.disableTags--
+			if r.disableTags == 0 {
+				if node.LinkData.Title != nil {
+					r.out(w, []byte(`" title="`))
+					escapeHTML(w, node.LinkData.Title)
+				}
+				r.out(w, []byte(`" />`))
+			}
+		}
+	case Code:
+		r.out(w, codeTag)
+		escapeHTML(w, node.Literal)
+		r.out(w, codeCloseTag)
+	case Document:
+		break
+	case Paragraph:
+		if skipParagraphTags(node) {
+			break
+		}
+		if entering {
+			// TODO: untangle this clusterfuck about when the newlines need
+			// to be added and when not.
+			if node.Prev != nil {
+				switch node.Prev.Type {
+				case HTMLBlock, List, Paragraph, Heading, CodeBlock, BlockQuote, HorizontalRule:
+					r.cr(w)
+				}
+			}
+			if node.Parent.Type == BlockQuote && node.Prev == nil {
+				r.cr(w)
+			}
+			r.out(w, pTag)
+		} else {
+			r.out(w, pCloseTag)
+			if !(node.Parent.Type == Item && node.Next == nil) {
+				r.cr(w)
+			}
+		}
+	case BlockQuote:
+		if entering {
+			r.cr(w)
+			r.out(w, blockquoteTag)
+		} else {
+			r.out(w, blockquoteCloseTag)
+			r.cr(w)
+		}
+	case HTMLBlock:
+		if r.Flags&SkipHTML != 0 {
+			break
+		}
+		r.cr(w)
+		r.out(w, node.Literal)
+		r.cr(w)
+	case Heading:
+		openTag, closeTag := headingTagsFromLevel(node.Level)
+		if entering {
+			if node.IsTitleblock {
+				attrs = append(attrs, `class="title"`)
+			}
+			if node.HeadingID != "" {
+				id := r.ensureUniqueHeadingID(node.HeadingID)
+				if r.HeadingIDPrefix != "" {
+					id = r.HeadingIDPrefix + id
+				}
+				if r.HeadingIDSuffix != "" {
+					id = id + r.HeadingIDSuffix
+				}
+				attrs = append(attrs, fmt.Sprintf(`id="%s"`, id))
+			}
+			r.cr(w)
+			r.tag(w, openTag, attrs)
+		} else {
+			r.out(w, closeTag)
+			if !(node.Parent.Type == Item && node.Next == nil) {
+				r.cr(w)
+			}
+		}
+	case HorizontalRule:
+		r.cr(w)
+		r.outHRTag(w)
+		r.cr(w)
+	case List:
+		openTag := ulTag
+		closeTag := ulCloseTag
+		if node.ListFlags&ListTypeOrdered != 0 {
+			openTag = olTag
+			closeTag = olCloseTag
+		}
+		if node.ListFlags&ListTypeDefinition != 0 {
+			openTag = dlTag
+			closeTag = dlCloseTag
+		}
+		if entering {
+			if node.IsFootnotesList {
+				r.out(w, footnotesDivBytes)
+				r.outHRTag(w)
+				r.cr(w)
+			}
+			r.cr(w)
+			if node.Parent.Type == Item && node.Parent.Parent.Tight {
+				r.cr(w)
+			}
+			r.tag(w, openTag[:len(openTag)-1], attrs)
+			r.cr(w)
+		} else {
+			r.out(w, closeTag)
+			//cr(w)
+			//if node.parent.Type != Item {
+			//	cr(w)
+			//}
+			if node.Parent.Type == Item && node.Next != nil {
+				r.cr(w)
+			}
+			if node.Parent.Type == Document || node.Parent.Type == BlockQuote {
+				r.cr(w)
+			}
+			if node.IsFootnotesList {
+				r.out(w, footnotesCloseDivBytes)
+			}
+		}
+	case Item:
+		openTag := liTag
+		closeTag := liCloseTag
+		if node.ListFlags&ListTypeDefinition != 0 {
+			openTag = ddTag
+			closeTag = ddCloseTag
+		}
+		if node.ListFlags&ListTypeTerm != 0 {
+			openTag = dtTag
+			closeTag = dtCloseTag
+		}
+		if entering {
+			if itemOpenCR(node) {
+				r.cr(w)
+			}
+			if node.ListData.RefLink != nil {
+				slug := slugify(node.ListData.RefLink)
+				r.out(w, footnoteItem(r.FootnoteAnchorPrefix, slug))
+				break
+			}
+			r.out(w, openTag)
+		} else {
+			if node.ListData.RefLink != nil {
+				slug := slugify(node.ListData.RefLink)
+				if r.Flags&FootnoteReturnLinks != 0 {
+					r.out(w, footnoteReturnLink(r.FootnoteAnchorPrefix, r.FootnoteReturnLinkContents, slug))
+				}
+			}
+			r.out(w, closeTag)
+			r.cr(w)
+		}
+	case CodeBlock:
+		attrs = appendLanguageAttr(attrs, node.Info)
+		r.cr(w)
+		r.out(w, preTag)
+		r.tag(w, codeTag[:len(codeTag)-1], attrs)
+		escapeHTML(w, node.Literal)
+		r.out(w, codeCloseTag)
+		r.out(w, preCloseTag)
+		if node.Parent.Type != Item {
+			r.cr(w)
+		}
+	case Table:
+		if entering {
+			r.cr(w)
+			r.out(w, tableTag)
+		} else {
+			r.out(w, tableCloseTag)
+			r.cr(w)
+		}
+	case TableCell:
+		openTag := tdTag
+		closeTag := tdCloseTag
+		if node.IsHeader {
+			openTag = thTag
+			closeTag = thCloseTag
+		}
+		if entering {
+			align := cellAlignment(node.Align)
+			if align != "" {
+				attrs = append(attrs, fmt.Sprintf(`align="%s"`, align))
+			}
+			if node.Prev == nil {
+				r.cr(w)
+			}
+			r.tag(w, openTag, attrs)
+		} else {
+			r.out(w, closeTag)
+			r.cr(w)
+		}
+	case TableHead:
+		if entering {
+			r.cr(w)
+			r.out(w, theadTag)
+		} else {
+			r.out(w, theadCloseTag)
+			r.cr(w)
+		}
+	case TableBody:
+		if entering {
+			r.cr(w)
+			r.out(w, tbodyTag)
+			// XXX: this is to adhere to a rather silly test. Should fix test.
+			if node.FirstChild == nil {
+				r.cr(w)
+			}
+		} else {
+			r.out(w, tbodyCloseTag)
+			r.cr(w)
+		}
+	case TableRow:
+		if entering {
+			r.cr(w)
+			r.out(w, trTag)
+		} else {
+			r.out(w, trCloseTag)
+			r.cr(w)
+		}
+	default:
+		panic("Unknown node type " + node.Type.String())
+	}
+	return GoToNext
+}
+
+// RenderHeader writes HTML document preamble and TOC if requested.
+func (r *HTMLRenderer) RenderHeader(w io.Writer, ast *Node) {
+	r.writeDocumentHeader(w)
+	if r.Flags&TOC != 0 {
+		r.writeTOC(w, ast)
+	}
+}
+
+// RenderFooter writes HTML document footer.
+func (r *HTMLRenderer) RenderFooter(w io.Writer, ast *Node) {
+	if r.Flags&CompletePage == 0 {
+		return
+	}
+	io.WriteString(w, "\n</body>\n</html>\n")
+}
+
+func (r *HTMLRenderer) writeDocumentHeader(w io.Writer) {
+	if r.Flags&CompletePage == 0 {
+		return
+	}
+	ending := ""
+	if r.Flags&UseXHTML != 0 {
+		io.WriteString(w, "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\" ")
+		io.WriteString(w, "\"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n")
+		io.WriteString(w, "<html xmlns=\"http://www.w3.org/1999/xhtml\">\n")
+		ending = " /"
+	} else {
+		io.WriteString(w, "<!DOCTYPE html>\n")
+		io.WriteString(w, "<html>\n")
+	}
+	io.WriteString(w, "<head>\n")
+	io.WriteString(w, "  <title>")
+	if r.Flags&Smartypants != 0 {
+		r.sr.Process(w, []byte(r.Title))
+	} else {
+		escapeHTML(w, []byte(r.Title))
+	}
+	io.WriteString(w, "</title>\n")
+	io.WriteString(w, "  <meta name=\"GENERATOR\" content=\"Blackfriday Markdown Processor v")
+	io.WriteString(w, Version)
+	io.WriteString(w, "\"")
+	io.WriteString(w, ending)
+	io.WriteString(w, ">\n")
+	io.WriteString(w, "  <meta charset=\"utf-8\"")
+	io.WriteString(w, ending)
+	io.WriteString(w, ">\n")
+	if r.CSS != "" {
+		io.WriteString(w, "  <link rel=\"stylesheet\" type=\"text/css\" href=\"")
+		escapeHTML(w, []byte(r.CSS))
+		io.WriteString(w, "\"")
+		io.WriteString(w, ending)
+		io.WriteString(w, ">\n")
+	}
+	if r.Icon != "" {
+		io.WriteString(w, "  <link rel=\"icon\" type=\"image/x-icon\" href=\"")
+		escapeHTML(w, []byte(r.Icon))
+		io.WriteString(w, "\"")
+		io.WriteString(w, ending)
+		io.WriteString(w, ">\n")
+	}
+	io.WriteString(w, "</head>\n")
+	io.WriteString(w, "<body>\n\n")
+}
+
+func (r *HTMLRenderer) writeTOC(w io.Writer, ast *Node) {
+	buf := bytes.Buffer{}
+
+	inHeading := false
+	tocLevel := 0
+	headingCount := 0
+
+	ast.Walk(func(node *Node, entering bool) WalkStatus {
+		if node.Type == Heading && !node.HeadingData.IsTitleblock {
+			inHeading = entering
+			if entering {
+				node.HeadingID = fmt.Sprintf("toc_%d", headingCount)
+				if node.Level == tocLevel {
+					buf.WriteString("</li>\n\n<li>")
+				} else if node.Level < tocLevel {
+					for node.Level < tocLevel {
+						tocLevel--
+						buf.WriteString("</li>\n</ul>")
+					}
+					buf.WriteString("</li>\n\n<li>")
+				} else {
+					for node.Level > tocLevel {
+						tocLevel++
+						buf.WriteString("\n<ul>\n<li>")
+					}
+				}
+
+				fmt.Fprintf(&buf, `<a href="#toc_%d">`, headingCount)
+				headingCount++
+			} else {
+				buf.WriteString("</a>")
+			}
+			return GoToNext
+		}
+
+		if inHeading {
+			return r.RenderNode(&buf, node, entering)
+		}
+
+		return GoToNext
+	})
+
+	for ; tocLevel > 0; tocLevel-- {
+		buf.WriteString("</li>\n</ul>")
+	}
+
+	if buf.Len() > 0 {
+		io.WriteString(w, "<nav>\n")
+		w.Write(buf.Bytes())
+		io.WriteString(w, "\n\n</nav>\n")
+	}
+	r.lastOutputLen = buf.Len()
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/inline.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/inline.go
new file mode 100644
index 00000000..3d633106
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/inline.go
@@ -0,0 +1,1214 @@
+//
+// Blackfriday Markdown Processor
+// Available at http://github.com/russross/blackfriday
+//
+// Copyright © 2011 Russ Ross <russ@russross.com>.
+// Distributed under the Simplified BSD License.
+// See README.md for details.
+//
+
+//
+// Functions to parse inline elements.
+//
+
+package blackfriday
+
+import (
+	"bytes"
+	"regexp"
+	"strconv"
+)
+
+var (
+	urlRe    = `((https?|ftp):\/\/|\/)[-A-Za-z0-9+&@#\/%?=~_|!:,.;\(\)]+`
+	anchorRe = regexp.MustCompile(`^(<a\shref="` + urlRe + `"(\stitle="[^"<>]+")?\s?>` + urlRe + `<\/a>)`)
+
+	// TODO: improve this regexp to catch all possible entities:
+	htmlEntityRe = regexp.MustCompile(`&[a-z]{2,5};`)
+)
+
+// Functions to parse text within a block
+// Each function returns the number of chars taken care of
+// data is the complete block being rendered
+// offset is the number of valid chars before the current cursor
+
+func (p *Markdown) inline(currBlock *Node, data []byte) {
+	// handlers might call us recursively: enforce a maximum depth
+	if p.nesting >= p.maxNesting || len(data) == 0 {
+		return
+	}
+	p.nesting++
+	beg, end := 0, 0
+	for end < len(data) {
+		handler := p.inlineCallback[data[end]]
+		if handler != nil {
+			if consumed, node := handler(p, data, end); consumed == 0 {
+				// No action from the callback.
+				end++
+			} else {
+				// Copy inactive chars into the output.
+				currBlock.AppendChild(text(data[beg:end]))
+				if node != nil {
+					currBlock.AppendChild(node)
+				}
+				// Skip past whatever the callback used.
+				beg = end + consumed
+				end = beg
+			}
+		} else {
+			end++
+		}
+	}
+	if beg < len(data) {
+		if data[end-1] == '\n' {
+			end--
+		}
+		currBlock.AppendChild(text(data[beg:end]))
+	}
+	p.nesting--
+}
+
+// single and double emphasis parsing
+func emphasis(p *Markdown, data []byte, offset int) (int, *Node) {
+	data = data[offset:]
+	c := data[0]
+
+	if len(data) > 2 && data[1] != c {
+		// whitespace cannot follow an opening emphasis;
+		// strikethrough only takes two characters '~~'
+		if c == '~' || isspace(data[1]) {
+			return 0, nil
+		}
+		ret, node := helperEmphasis(p, data[1:], c)
+		if ret == 0 {
+			return 0, nil
+		}
+
+		return ret + 1, node
+	}
+
+	if len(data) > 3 && data[1] == c && data[2] != c {
+		if isspace(data[2]) {
+			return 0, nil
+		}
+		ret, node := helperDoubleEmphasis(p, data[2:], c)
+		if ret == 0 {
+			return 0, nil
+		}
+
+		return ret + 2, node
+	}
+
+	if len(data) > 4 && data[1] == c && data[2] == c && data[3] != c {
+		if c == '~' || isspace(data[3]) {
+			return 0, nil
+		}
+		ret, node := helperTripleEmphasis(p, data, 3, c)
+		if ret == 0 {
+			return 0, nil
+		}
+
+		return ret + 3, node
+	}
+
+	return 0, nil
+}
+
+func codeSpan(p *Markdown, data []byte, offset int) (int, *Node) {
+	data = data[offset:]
+
+	nb := 0
+
+	// count the number of backticks in the delimiter
+	for nb < len(data) && data[nb] == '`' {
+		nb++
+	}
+
+	// find the next delimiter
+	i, end := 0, 0
+	for end = nb; end < len(data) && i < nb; end++ {
+		if data[end] == '`' {
+			i++
+		} else {
+			i = 0
+		}
+	}
+
+	// no matching delimiter?
+	if i < nb && end >= len(data) {
+		return 0, nil
+	}
+
+	// trim outside whitespace
+	fBegin := nb
+	for fBegin < end && data[fBegin] == ' ' {
+		fBegin++
+	}
+
+	fEnd := end - nb
+	for fEnd > fBegin && data[fEnd-1] == ' ' {
+		fEnd--
+	}
+
+	// render the code span
+	if fBegin != fEnd {
+		code := NewNode(Code)
+		code.Literal = data[fBegin:fEnd]
+		return end, code
+	}
+
+	return end, nil
+}
+
+// newline preceded by two spaces becomes <br>
+func maybeLineBreak(p *Markdown, data []byte, offset int) (int, *Node) {
+	origOffset := offset
+	for offset < len(data) && data[offset] == ' ' {
+		offset++
+	}
+
+	if offset < len(data) && data[offset] == '\n' {
+		if offset-origOffset >= 2 {
+			return offset - origOffset + 1, NewNode(Hardbreak)
+		}
+		return offset - origOffset, nil
+	}
+	return 0, nil
+}
+
+// newline without two spaces works when HardLineBreak is enabled
+func lineBreak(p *Markdown, data []byte, offset int) (int, *Node) {
+	if p.extensions&HardLineBreak != 0 {
+		return 1, NewNode(Hardbreak)
+	}
+	return 0, nil
+}
+
+type linkType int
+
+const (
+	linkNormal linkType = iota
+	linkImg
+	linkDeferredFootnote
+	linkInlineFootnote
+)
+
+func isReferenceStyleLink(data []byte, pos int, t linkType) bool {
+	if t == linkDeferredFootnote {
+		return false
+	}
+	return pos < len(data)-1 && data[pos] == '[' && data[pos+1] != '^'
+}
+
+func maybeImage(p *Markdown, data []byte, offset int) (int, *Node) {
+	if offset < len(data)-1 && data[offset+1] == '[' {
+		return link(p, data, offset)
+	}
+	return 0, nil
+}
+
+func maybeInlineFootnote(p *Markdown, data []byte, offset int) (int, *Node) {
+	if offset < len(data)-1 && data[offset+1] == '[' {
+		return link(p, data, offset)
+	}
+	return 0, nil
+}
+
+// '[': parse a link or an image or a footnote
+func link(p *Markdown, data []byte, offset int) (int, *Node) {
+	// no links allowed inside regular links, footnote, and deferred footnotes
+	if p.insideLink && (offset > 0 && data[offset-1] == '[' || len(data)-1 > offset && data[offset+1] == '^') {
+		return 0, nil
+	}
+
+	var t linkType
+	switch {
+	// special case: ![^text] == deferred footnote (that follows something with
+	// an exclamation point)
+	case p.extensions&Footnotes != 0 && len(data)-1 > offset && data[offset+1] == '^':
+		t = linkDeferredFootnote
+	// ![alt] == image
+	case offset >= 0 && data[offset] == '!':
+		t = linkImg
+		offset++
+	// ^[text] == inline footnote
+	// [^refId] == deferred footnote
+	case p.extensions&Footnotes != 0:
+		if offset >= 0 && data[offset] == '^' {
+			t = linkInlineFootnote
+			offset++
+		} else if len(data)-1 > offset && data[offset+1] == '^' {
+			t = linkDeferredFootnote
+		}
+	// [text] == regular link
+	default:
+		t = linkNormal
+	}
+
+	data = data[offset:]
+
+	var (
+		i                       = 1
+		noteID                  int
+		title, link, altContent []byte
+		textHasNl               = false
+	)
+
+	if t == linkDeferredFootnote {
+		i++
+	}
+
+	// look for the matching closing bracket
+	for level := 1; level > 0 && i < len(data); i++ {
+		switch {
+		case data[i] == '\n':
+			textHasNl = true
+
+		case data[i-1] == '\\':
+			continue
+
+		case data[i] == '[':
+			level++
+
+		case data[i] == ']':
+			level--
+			if level <= 0 {
+				i-- // compensate for extra i++ in for loop
+			}
+		}
+	}
+
+	if i >= len(data) {
+		return 0, nil
+	}
+
+	txtE := i
+	i++
+	var footnoteNode *Node
+
+	// skip any amount of whitespace or newline
+	// (this is much more lax than original markdown syntax)
+	for i < len(data) && isspace(data[i]) {
+		i++
+	}
+
+	// inline style link
+	switch {
+	case i < len(data) && data[i] == '(':
+		// skip initial whitespace
+		i++
+
+		for i < len(data) && isspace(data[i]) {
+			i++
+		}
+
+		linkB := i
+
+		// look for link end: ' " )
+	findlinkend:
+		for i < len(data) {
+			switch {
+			case data[i] == '\\':
+				i += 2
+
+			case data[i] == ')' || data[i] == '\'' || data[i] == '"':
+				break findlinkend
+
+			default:
+				i++
+			}
+		}
+
+		if i >= len(data) {
+			return 0, nil
+		}
+		linkE := i
+
+		// look for title end if present
+		titleB, titleE := 0, 0
+		if data[i] == '\'' || data[i] == '"' {
+			i++
+			titleB = i
+
+		findtitleend:
+			for i < len(data) {
+				switch {
+				case data[i] == '\\':
+					i += 2
+
+				case data[i] == ')':
+					break findtitleend
+
+				default:
+					i++
+				}
+			}
+
+			if i >= len(data) {
+				return 0, nil
+			}
+
+			// skip whitespace after title
+			titleE = i - 1
+			for titleE > titleB && isspace(data[titleE]) {
+				titleE--
+			}
+
+			// check for closing quote presence
+			if data[titleE] != '\'' && data[titleE] != '"' {
+				titleB, titleE = 0, 0
+				linkE = i
+			}
+		}
+
+		// remove whitespace at the end of the link
+		for linkE > linkB && isspace(data[linkE-1]) {
+			linkE--
+		}
+
+		// remove optional angle brackets around the link
+		if data[linkB] == '<' {
+			linkB++
+		}
+		if data[linkE-1] == '>' {
+			linkE--
+		}
+
+		// build escaped link and title
+		if linkE > linkB {
+			link = data[linkB:linkE]
+		}
+
+		if titleE > titleB {
+			title = data[titleB:titleE]
+		}
+
+		i++
+
+	// reference style link
+	case isReferenceStyleLink(data, i, t):
+		var id []byte
+		altContentConsidered := false
+
+		// look for the id
+		i++
+		linkB := i
+		for i < len(data) && data[i] != ']' {
+			i++
+		}
+		if i >= len(data) {
+			return 0, nil
+		}
+		linkE := i
+
+		// find the reference
+		if linkB == linkE {
+			if textHasNl {
+				var b bytes.Buffer
+
+				for j := 1; j < txtE; j++ {
+					switch {
+					case data[j] != '\n':
+						b.WriteByte(data[j])
+					case data[j-1] != ' ':
+						b.WriteByte(' ')
+					}
+				}
+
+				id = b.Bytes()
+			} else {
+				id = data[1:txtE]
+				altContentConsidered = true
+			}
+		} else {
+			id = data[linkB:linkE]
+		}
+
+		// find the reference with matching id
+		lr, ok := p.getRef(string(id))
+		if !ok {
+			return 0, nil
+		}
+
+		// keep link and title from reference
+		link = lr.link
+		title = lr.title
+		if altContentConsidered {
+			altContent = lr.text
+		}
+		i++
+
+	// shortcut reference style link or reference or inline footnote
+	default:
+		var id []byte
+
+		// craft the id
+		if textHasNl {
+			var b bytes.Buffer
+
+			for j := 1; j < txtE; j++ {
+				switch {
+				case data[j] != '\n':
+					b.WriteByte(data[j])
+				case data[j-1] != ' ':
+					b.WriteByte(' ')
+				}
+			}
+
+			id = b.Bytes()
+		} else {
+			if t == linkDeferredFootnote {
+				id = data[2:txtE] // get rid of the ^
+			} else {
+				id = data[1:txtE]
+			}
+		}
+
+		footnoteNode = NewNode(Item)
+		if t == linkInlineFootnote {
+			// create a new reference
+			noteID = len(p.notes) + 1
+
+			var fragment []byte
+			if len(id) > 0 {
+				if len(id) < 16 {
+					fragment = make([]byte, len(id))
+				} else {
+					fragment = make([]byte, 16)
+				}
+				copy(fragment, slugify(id))
+			} else {
+				fragment = append([]byte("footnote-"), []byte(strconv.Itoa(noteID))...)
+			}
+
+			ref := &reference{
+				noteID:   noteID,
+				hasBlock: false,
+				link:     fragment,
+				title:    id,
+				footnote: footnoteNode,
+			}
+
+			p.notes = append(p.notes, ref)
+
+			link = ref.link
+			title = ref.title
+		} else {
+			// find the reference with matching id
+			lr, ok := p.getRef(string(id))
+			if !ok {
+				return 0, nil
+			}
+
+			if t == linkDeferredFootnote {
+				lr.noteID = len(p.notes) + 1
+				lr.footnote = footnoteNode
+				p.notes = append(p.notes, lr)
+			}
+
+			// keep link and title from reference
+			link = lr.link
+			// if inline footnote, title == footnote contents
+			title = lr.title
+			noteID = lr.noteID
+		}
+
+		// rewind the whitespace
+		i = txtE + 1
+	}
+
+	var uLink []byte
+	if t == linkNormal || t == linkImg {
+		if len(link) > 0 {
+			var uLinkBuf bytes.Buffer
+			unescapeText(&uLinkBuf, link)
+			uLink = uLinkBuf.Bytes()
+		}
+
+		// links need something to click on and somewhere to go
+		if len(uLink) == 0 || (t == linkNormal && txtE <= 1) {
+			return 0, nil
+		}
+	}
+
+	// call the relevant rendering function
+	var linkNode *Node
+	switch t {
+	case linkNormal:
+		linkNode = NewNode(Link)
+		linkNode.Destination = normalizeURI(uLink)
+		linkNode.Title = title
+		if len(altContent) > 0 {
+			linkNode.AppendChild(text(altContent))
+		} else {
+			// links cannot contain other links, so turn off link parsing
+			// temporarily and recurse
+			insideLink := p.insideLink
+			p.insideLink = true
+			p.inline(linkNode, data[1:txtE])
+			p.insideLink = insideLink
+		}
+
+	case linkImg:
+		linkNode = NewNode(Image)
+		linkNode.Destination = uLink
+		linkNode.Title = title
+		linkNode.AppendChild(text(data[1:txtE]))
+		i++
+
+	case linkInlineFootnote, linkDeferredFootnote:
+		linkNode = NewNode(Link)
+		linkNode.Destination = link
+		linkNode.Title = title
+		linkNode.NoteID = noteID
+		linkNode.Footnote = footnoteNode
+		if t == linkInlineFootnote {
+			i++
+		}
+
+	default:
+		return 0, nil
+	}
+
+	return i, linkNode
+}
+
+func (p *Markdown) inlineHTMLComment(data []byte) int {
+	if len(data) < 5 {
+		return 0
+	}
+	if data[0] != '<' || data[1] != '!' || data[2] != '-' || data[3] != '-' {
+		return 0
+	}
+	i := 5
+	// scan for an end-of-comment marker, across lines if necessary
+	for i < len(data) && !(data[i-2] == '-' && data[i-1] == '-' && data[i] == '>') {
+		i++
+	}
+	// no end-of-comment marker
+	if i >= len(data) {
+		return 0
+	}
+	return i + 1
+}
+
+func stripMailto(link []byte) []byte {
+	if bytes.HasPrefix(link, []byte("mailto://")) {
+		return link[9:]
+	} else if bytes.HasPrefix(link, []byte("mailto:")) {
+		return link[7:]
+	} else {
+		return link
+	}
+}
+
+// autolinkType specifies a kind of autolink that gets detected.
+type autolinkType int
+
+// These are the possible flag values for the autolink renderer.
+const (
+	notAutolink autolinkType = iota
+	normalAutolink
+	emailAutolink
+)
+
+// '<' when tags or autolinks are allowed
+func leftAngle(p *Markdown, data []byte, offset int) (int, *Node) {
+	data = data[offset:]
+	altype, end := tagLength(data)
+	if size := p.inlineHTMLComment(data); size > 0 {
+		end = size
+	}
+	if end > 2 {
+		if altype != notAutolink {
+			var uLink bytes.Buffer
+			unescapeText(&uLink, data[1:end+1-2])
+			if uLink.Len() > 0 {
+				link := uLink.Bytes()
+				node := NewNode(Link)
+				node.Destination = link
+				if altype == emailAutolink {
+					node.Destination = append([]byte("mailto:"), link...)
+				}
+				node.AppendChild(text(stripMailto(link)))
+				return end, node
+			}
+		} else {
+			htmlTag := NewNode(HTMLSpan)
+			htmlTag.Literal = data[:end]
+			return end, htmlTag
+		}
+	}
+
+	return end, nil
+}
+
+// '\\' backslash escape
+var escapeChars = []byte("\\`*_{}[]()#+-.!:|&<>~")
+
+func escape(p *Markdown, data []byte, offset int) (int, *Node) {
+	data = data[offset:]
+
+	if len(data) > 1 {
+		if p.extensions&BackslashLineBreak != 0 && data[1] == '\n' {
+			return 2, NewNode(Hardbreak)
+		}
+		if bytes.IndexByte(escapeChars, data[1]) < 0 {
+			return 0, nil
+		}
+
+		return 2, text(data[1:2])
+	}
+
+	return 2, nil
+}
+
+func unescapeText(ob *bytes.Buffer, src []byte) {
+	i := 0
+	for i < len(src) {
+		org := i
+		for i < len(src) && src[i] != '\\' {
+			i++
+		}
+
+		if i > org {
+			ob.Write(src[org:i])
+		}
+
+		if i+1 >= len(src) {
+			break
+		}
+
+		ob.WriteByte(src[i+1])
+		i += 2
+	}
+}
+
+// '&' escaped when it doesn't belong to an entity
+// valid entities are assumed to be anything matching &#?[A-Za-z0-9]+;
+func entity(p *Markdown, data []byte, offset int) (int, *Node) {
+	data = data[offset:]
+
+	end := 1
+
+	if end < len(data) && data[end] == '#' {
+		end++
+	}
+
+	for end < len(data) && isalnum(data[end]) {
+		end++
+	}
+
+	if end < len(data) && data[end] == ';' {
+		end++ // real entity
+	} else {
+		return 0, nil // lone '&'
+	}
+
+	ent := data[:end]
+	// undo &amp; escaping or it will be converted to &amp;amp; by another
+	// escaper in the renderer
+	if bytes.Equal(ent, []byte("&amp;")) {
+		ent = []byte{'&'}
+	}
+
+	return end, text(ent)
+}
+
+func linkEndsWithEntity(data []byte, linkEnd int) bool {
+	entityRanges := htmlEntityRe.FindAllIndex(data[:linkEnd], -1)
+	return entityRanges != nil && entityRanges[len(entityRanges)-1][1] == linkEnd
+}
+
+// hasPrefixCaseInsensitive is a custom implementation of
+//     strings.HasPrefix(strings.ToLower(s), prefix)
+// we rolled our own because ToLower pulls in a huge machinery of lowercasing
+// anything from Unicode and that's very slow. Since this func will only be
+// used on ASCII protocol prefixes, we can take shortcuts.
+func hasPrefixCaseInsensitive(s, prefix []byte) bool {
+	if len(s) < len(prefix) {
+		return false
+	}
+	delta := byte('a' - 'A')
+	for i, b := range prefix {
+		if b != s[i] && b != s[i]+delta {
+			return false
+		}
+	}
+	return true
+}
+
+var protocolPrefixes = [][]byte{
+	[]byte("http://"),
+	[]byte("https://"),
+	[]byte("ftp://"),
+	[]byte("file://"),
+	[]byte("mailto:"),
+}
+
+const shortestPrefix = 6 // len("ftp://"), the shortest of the above
+
+func maybeAutoLink(p *Markdown, data []byte, offset int) (int, *Node) {
+	// quick check to rule out most false hits
+	if p.insideLink || len(data) < offset+shortestPrefix {
+		return 0, nil
+	}
+	for _, prefix := range protocolPrefixes {
+		endOfHead := offset + 8 // 8 is the len() of the longest prefix
+		if endOfHead > len(data) {
+			endOfHead = len(data)
+		}
+		if hasPrefixCaseInsensitive(data[offset:endOfHead], prefix) {
+			return autoLink(p, data, offset)
+		}
+	}
+	return 0, nil
+}
+
+func autoLink(p *Markdown, data []byte, offset int) (int, *Node) {
+	// Now a more expensive check to see if we're not inside an anchor element
+	anchorStart := offset
+	offsetFromAnchor := 0
+	for anchorStart > 0 && data[anchorStart] != '<' {
+		anchorStart--
+		offsetFromAnchor++
+	}
+
+	anchorStr := anchorRe.Find(data[anchorStart:])
+	if anchorStr != nil {
+		anchorClose := NewNode(HTMLSpan)
+		anchorClose.Literal = anchorStr[offsetFromAnchor:]
+		return len(anchorStr) - offsetFromAnchor, anchorClose
+	}
+
+	// scan backward for a word boundary
+	rewind := 0
+	for offset-rewind > 0 && rewind <= 7 && isletter(data[offset-rewind-1]) {
+		rewind++
+	}
+	if rewind > 6 { // longest supported protocol is "mailto" which has 6 letters
+		return 0, nil
+	}
+
+	origData := data
+	data = data[offset-rewind:]
+
+	if !isSafeLink(data) {
+		return 0, nil
+	}
+
+	linkEnd := 0
+	for linkEnd < len(data) && !isEndOfLink(data[linkEnd]) {
+		linkEnd++
+	}
+
+	// Skip punctuation at the end of the link
+	if (data[linkEnd-1] == '.' || data[linkEnd-1] == ',') && data[linkEnd-2] != '\\' {
+		linkEnd--
+	}
+
+	// But don't skip semicolon if it's a part of escaped entity:
+	if data[linkEnd-1] == ';' && data[linkEnd-2] != '\\' && !linkEndsWithEntity(data, linkEnd) {
+		linkEnd--
+	}
+
+	// See if the link finishes with a punctuation sign that can be closed.
+	var copen byte
+	switch data[linkEnd-1] {
+	case '"':
+		copen = '"'
+	case '\'':
+		copen = '\''
+	case ')':
+		copen = '('
+	case ']':
+		copen = '['
+	case '}':
+		copen = '{'
+	default:
+		copen = 0
+	}
+
+	if copen != 0 {
+		bufEnd := offset - rewind + linkEnd - 2
+
+		openDelim := 1
+
+		/* Try to close the final punctuation sign in this same line;
+		 * if we managed to close it outside of the URL, that means that it's
+		 * not part of the URL. If it closes inside the URL, that means it
+		 * is part of the URL.
+		 *
+		 * Examples:
+		 *
+		 *      foo http://www.pokemon.com/Pikachu_(Electric) bar
+		 *              => http://www.pokemon.com/Pikachu_(Electric)
+		 *
+		 *      foo (http://www.pokemon.com/Pikachu_(Electric)) bar
+		 *              => http://www.pokemon.com/Pikachu_(Electric)
+		 *
+		 *      foo http://www.pokemon.com/Pikachu_(Electric)) bar
+		 *              => http://www.pokemon.com/Pikachu_(Electric))
+		 *
+		 *      (foo http://www.pokemon.com/Pikachu_(Electric)) bar
+		 *              => foo http://www.pokemon.com/Pikachu_(Electric)
+		 */
+
+		for bufEnd >= 0 && origData[bufEnd] != '\n' && openDelim != 0 {
+			if origData[bufEnd] == data[linkEnd-1] {
+				openDelim++
+			}
+
+			if origData[bufEnd] == copen {
+				openDelim--
+			}
+
+			bufEnd--
+		}
+
+		if openDelim == 0 {
+			linkEnd--
+		}
+	}
+
+	var uLink bytes.Buffer
+	unescapeText(&uLink, data[:linkEnd])
+
+	if uLink.Len() > 0 {
+		node := NewNode(Link)
+		node.Destination = uLink.Bytes()
+		node.AppendChild(text(uLink.Bytes()))
+		return linkEnd, node
+	}
+
+	return linkEnd, nil
+}
+
+func isEndOfLink(char byte) bool {
+	return isspace(char) || char == '<'
+}
+
+var validUris = [][]byte{[]byte("http://"), []byte("https://"), []byte("ftp://"), []byte("mailto://")}
+var validPaths = [][]byte{[]byte("/"), []byte("./"), []byte("../")}
+
+func isSafeLink(link []byte) bool {
+	for _, path := range validPaths {
+		if len(link) >= len(path) && bytes.Equal(link[:len(path)], path) {
+			if len(link) == len(path) {
+				return true
+			} else if isalnum(link[len(path)]) {
+				return true
+			}
+		}
+	}
+
+	for _, prefix := range validUris {
+		// TODO: handle unicode here
+		// case-insensitive prefix test
+		if len(link) > len(prefix) && bytes.Equal(bytes.ToLower(link[:len(prefix)]), prefix) && isalnum(link[len(prefix)]) {
+			return true
+		}
+	}
+
+	return false
+}
+
+// return the length of the given tag, or 0 is it's not valid
+func tagLength(data []byte) (autolink autolinkType, end int) {
+	var i, j int
+
+	// a valid tag can't be shorter than 3 chars
+	if len(data) < 3 {
+		return notAutolink, 0
+	}
+
+	// begins with a '<' optionally followed by '/', followed by letter or number
+	if data[0] != '<' {
+		return notAutolink, 0
+	}
+	if data[1] == '/' {
+		i = 2
+	} else {
+		i = 1
+	}
+
+	if !isalnum(data[i]) {
+		return notAutolink, 0
+	}
+
+	// scheme test
+	autolink = notAutolink
+
+	// try to find the beginning of an URI
+	for i < len(data) && (isalnum(data[i]) || data[i] == '.' || data[i] == '+' || data[i] == '-') {
+		i++
+	}
+
+	if i > 1 && i < len(data) && data[i] == '@' {
+		if j = isMailtoAutoLink(data[i:]); j != 0 {
+			return emailAutolink, i + j
+		}
+	}
+
+	if i > 2 && i < len(data) && data[i] == ':' {
+		autolink = normalAutolink
+		i++
+	}
+
+	// complete autolink test: no whitespace or ' or "
+	switch {
+	case i >= len(data):
+		autolink = notAutolink
+	case autolink != notAutolink:
+		j = i
+
+		for i < len(data) {
+			if data[i] == '\\' {
+				i += 2
+			} else if data[i] == '>' || data[i] == '\'' || data[i] == '"' || isspace(data[i]) {
+				break
+			} else {
+				i++
+			}
+
+		}
+
+		if i >= len(data) {
+			return autolink, 0
+		}
+		if i > j && data[i] == '>' {
+			return autolink, i + 1
+		}
+
+		// one of the forbidden chars has been found
+		autolink = notAutolink
+	}
+	i += bytes.IndexByte(data[i:], '>')
+	if i < 0 {
+		return autolink, 0
+	}
+	return autolink, i + 1
+}
+
+// look for the address part of a mail autolink and '>'
+// this is less strict than the original markdown e-mail address matching
+func isMailtoAutoLink(data []byte) int {
+	nb := 0
+
+	// address is assumed to be: [-@._a-zA-Z0-9]+ with exactly one '@'
+	for i := 0; i < len(data); i++ {
+		if isalnum(data[i]) {
+			continue
+		}
+
+		switch data[i] {
+		case '@':
+			nb++
+
+		case '-', '.', '_':
+			break
+
+		case '>':
+			if nb == 1 {
+				return i + 1
+			}
+			return 0
+		default:
+			return 0
+		}
+	}
+
+	return 0
+}
+
+// look for the next emph char, skipping other constructs
+func helperFindEmphChar(data []byte, c byte) int {
+	i := 0
+
+	for i < len(data) {
+		for i < len(data) && data[i] != c && data[i] != '`' && data[i] != '[' {
+			i++
+		}
+		if i >= len(data) {
+			return 0
+		}
+		// do not count escaped chars
+		if i != 0 && data[i-1] == '\\' {
+			i++
+			continue
+		}
+		if data[i] == c {
+			return i
+		}
+
+		if data[i] == '`' {
+			// skip a code span
+			tmpI := 0
+			i++
+			for i < len(data) && data[i] != '`' {
+				if tmpI == 0 && data[i] == c {
+					tmpI = i
+				}
+				i++
+			}
+			if i >= len(data) {
+				return tmpI
+			}
+			i++
+		} else if data[i] == '[' {
+			// skip a link
+			tmpI := 0
+			i++
+			for i < len(data) && data[i] != ']' {
+				if tmpI == 0 && data[i] == c {
+					tmpI = i
+				}
+				i++
+			}
+			i++
+			for i < len(data) && (data[i] == ' ' || data[i] == '\n') {
+				i++
+			}
+			if i >= len(data) {
+				return tmpI
+			}
+			if data[i] != '[' && data[i] != '(' { // not a link
+				if tmpI > 0 {
+					return tmpI
+				}
+				continue
+			}
+			cc := data[i]
+			i++
+			for i < len(data) && data[i] != cc {
+				if tmpI == 0 && data[i] == c {
+					return i
+				}
+				i++
+			}
+			if i >= len(data) {
+				return tmpI
+			}
+			i++
+		}
+	}
+	return 0
+}
+
+func helperEmphasis(p *Markdown, data []byte, c byte) (int, *Node) {
+	i := 0
+
+	// skip one symbol if coming from emph3
+	if len(data) > 1 && data[0] == c && data[1] == c {
+		i = 1
+	}
+
+	for i < len(data) {
+		length := helperFindEmphChar(data[i:], c)
+		if length == 0 {
+			return 0, nil
+		}
+		i += length
+		if i >= len(data) {
+			return 0, nil
+		}
+
+		if i+1 < len(data) && data[i+1] == c {
+			i++
+			continue
+		}
+
+		if data[i] == c && !isspace(data[i-1]) {
+
+			if p.extensions&NoIntraEmphasis != 0 {
+				if !(i+1 == len(data) || isspace(data[i+1]) || ispunct(data[i+1])) {
+					continue
+				}
+			}
+
+			emph := NewNode(Emph)
+			p.inline(emph, data[:i])
+			return i + 1, emph
+		}
+	}
+
+	return 0, nil
+}
+
+func helperDoubleEmphasis(p *Markdown, data []byte, c byte) (int, *Node) {
+	i := 0
+
+	for i < len(data) {
+		length := helperFindEmphChar(data[i:], c)
+		if length == 0 {
+			return 0, nil
+		}
+		i += length
+
+		if i+1 < len(data) && data[i] == c && data[i+1] == c && i > 0 && !isspace(data[i-1]) {
+			nodeType := Strong
+			if c == '~' {
+				nodeType = Del
+			}
+			node := NewNode(nodeType)
+			p.inline(node, data[:i])
+			return i + 2, node
+		}
+		i++
+	}
+	return 0, nil
+}
+
+func helperTripleEmphasis(p *Markdown, data []byte, offset int, c byte) (int, *Node) {
+	i := 0
+	origData := data
+	data = data[offset:]
+
+	for i < len(data) {
+		length := helperFindEmphChar(data[i:], c)
+		if length == 0 {
+			return 0, nil
+		}
+		i += length
+
+		// skip whitespace preceded symbols
+		if data[i] != c || isspace(data[i-1]) {
+			continue
+		}
+
+		switch {
+		case i+2 < len(data) && data[i+1] == c && data[i+2] == c:
+			// triple symbol found
+			strong := NewNode(Strong)
+			em := NewNode(Emph)
+			strong.AppendChild(em)
+			p.inline(em, data[:i])
+			return i + 3, strong
+		case (i+1 < len(data) && data[i+1] == c):
+			// double symbol found, hand over to emph1
+			length, node := helperEmphasis(p, origData[offset-2:], c)
+			if length == 0 {
+				return 0, nil
+			}
+			return length - 2, node
+		default:
+			// single symbol found, hand over to emph2
+			length, node := helperDoubleEmphasis(p, origData[offset-1:], c)
+			if length == 0 {
+				return 0, nil
+			}
+			return length - 1, node
+		}
+	}
+	return 0, nil
+}
+
+func text(s []byte) *Node {
+	node := NewNode(Text)
+	node.Literal = s
+	return node
+}
+
+func normalizeURI(s []byte) []byte {
+	return s // TODO: implement
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/markdown.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/markdown.go
new file mode 100644
index 00000000..ff61cb05
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/markdown.go
@@ -0,0 +1,940 @@
+// Blackfriday Markdown Processor
+// Available at http://github.com/russross/blackfriday
+//
+// Copyright © 2011 Russ Ross <russ@russross.com>.
+// Distributed under the Simplified BSD License.
+// See README.md for details.
+
+package blackfriday
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"strings"
+	"unicode/utf8"
+)
+
+//
+// Markdown parsing and processing
+//
+
+// Version string of the package. Appears in the rendered document when
+// CompletePage flag is on.
+const Version = "2.0"
+
+// Extensions is a bitwise or'ed collection of enabled Blackfriday's
+// extensions.
+type Extensions int
+
+// These are the supported markdown parsing extensions.
+// OR these values together to select multiple extensions.
+const (
+	NoExtensions           Extensions = 0
+	NoIntraEmphasis        Extensions = 1 << iota // Ignore emphasis markers inside words
+	Tables                                        // Render tables
+	FencedCode                                    // Render fenced code blocks
+	Autolink                                      // Detect embedded URLs that are not explicitly marked
+	Strikethrough                                 // Strikethrough text using ~~test~~
+	LaxHTMLBlocks                                 // Loosen up HTML block parsing rules
+	SpaceHeadings                                 // Be strict about prefix heading rules
+	HardLineBreak                                 // Translate newlines into line breaks
+	TabSizeEight                                  // Expand tabs to eight spaces instead of four
+	Footnotes                                     // Pandoc-style footnotes
+	NoEmptyLineBeforeBlock                        // No need to insert an empty line to start a (code, quote, ordered list, unordered list) block
+	HeadingIDs                                    // specify heading IDs  with {#id}
+	Titleblock                                    // Titleblock ala pandoc
+	AutoHeadingIDs                                // Create the heading ID from the text
+	BackslashLineBreak                            // Translate trailing backslashes into line breaks
+	DefinitionLists                               // Render definition lists
+
+	CommonHTMLFlags HTMLFlags = UseXHTML | Smartypants |
+		SmartypantsFractions | SmartypantsDashes | SmartypantsLatexDashes
+
+	CommonExtensions Extensions = NoIntraEmphasis | Tables | FencedCode |
+		Autolink | Strikethrough | SpaceHeadings | HeadingIDs |
+		BackslashLineBreak | DefinitionLists
+)
+
+// ListType contains bitwise or'ed flags for list and list item objects.
+type ListType int
+
+// These are the possible flag values for the ListItem renderer.
+// Multiple flag values may be ORed together.
+// These are mostly of interest if you are writing a new output format.
+const (
+	ListTypeOrdered ListType = 1 << iota
+	ListTypeDefinition
+	ListTypeTerm
+
+	ListItemContainsBlock
+	ListItemBeginningOfList // TODO: figure out if this is of any use now
+	ListItemEndOfList
+)
+
+// CellAlignFlags holds a type of alignment in a table cell.
+type CellAlignFlags int
+
+// These are the possible flag values for the table cell renderer.
+// Only a single one of these values will be used; they are not ORed together.
+// These are mostly of interest if you are writing a new output format.
+const (
+	TableAlignmentLeft CellAlignFlags = 1 << iota
+	TableAlignmentRight
+	TableAlignmentCenter = (TableAlignmentLeft | TableAlignmentRight)
+)
+
+// The size of a tab stop.
+const (
+	TabSizeDefault = 4
+	TabSizeDouble  = 8
+)
+
+// blockTags is a set of tags that are recognized as HTML block tags.
+// Any of these can be included in markdown text without special escaping.
+var blockTags = map[string]struct{}{
+	"blockquote": struct{}{},
+	"del":        struct{}{},
+	"div":        struct{}{},
+	"dl":         struct{}{},
+	"fieldset":   struct{}{},
+	"form":       struct{}{},
+	"h1":         struct{}{},
+	"h2":         struct{}{},
+	"h3":         struct{}{},
+	"h4":         struct{}{},
+	"h5":         struct{}{},
+	"h6":         struct{}{},
+	"iframe":     struct{}{},
+	"ins":        struct{}{},
+	"math":       struct{}{},
+	"noscript":   struct{}{},
+	"ol":         struct{}{},
+	"pre":        struct{}{},
+	"p":          struct{}{},
+	"script":     struct{}{},
+	"style":      struct{}{},
+	"table":      struct{}{},
+	"ul":         struct{}{},
+
+	// HTML5
+	"address":    struct{}{},
+	"article":    struct{}{},
+	"aside":      struct{}{},
+	"canvas":     struct{}{},
+	"figcaption": struct{}{},
+	"figure":     struct{}{},
+	"footer":     struct{}{},
+	"header":     struct{}{},
+	"hgroup":     struct{}{},
+	"main":       struct{}{},
+	"nav":        struct{}{},
+	"output":     struct{}{},
+	"progress":   struct{}{},
+	"section":    struct{}{},
+	"video":      struct{}{},
+}
+
+// Renderer is the rendering interface. This is mostly of interest if you are
+// implementing a new rendering format.
+//
+// Only an HTML implementation is provided in this repository, see the README
+// for external implementations.
+type Renderer interface {
+	// RenderNode is the main rendering method. It will be called once for
+	// every leaf node and twice for every non-leaf node (first with
+	// entering=true, then with entering=false). The method should write its
+	// rendition of the node to the supplied writer w.
+	RenderNode(w io.Writer, node *Node, entering bool) WalkStatus
+
+	// RenderHeader is a method that allows the renderer to produce some
+	// content preceding the main body of the output document. The header is
+	// understood in the broad sense here. For example, the default HTML
+	// renderer will write not only the HTML document preamble, but also the
+	// table of contents if it was requested.
+	//
+	// The method will be passed an entire document tree, in case a particular
+	// implementation needs to inspect it to produce output.
+	//
+	// The output should be written to the supplied writer w. If your
+	// implementation has no header to write, supply an empty implementation.
+	RenderHeader(w io.Writer, ast *Node)
+
+	// RenderFooter is a symmetric counterpart of RenderHeader.
+	RenderFooter(w io.Writer, ast *Node)
+}
+
+// Callback functions for inline parsing. One such function is defined
+// for each character that triggers a response when parsing inline data.
+type inlineParser func(p *Markdown, data []byte, offset int) (int, *Node)
+
+// Markdown is a type that holds extensions and the runtime state used by
+// Parse, and the renderer. You can not use it directly, construct it with New.
+type Markdown struct {
+	renderer          Renderer
+	referenceOverride ReferenceOverrideFunc
+	refs              map[string]*reference
+	inlineCallback    [256]inlineParser
+	extensions        Extensions
+	nesting           int
+	maxNesting        int
+	insideLink        bool
+
+	// Footnotes need to be ordered as well as available to quickly check for
+	// presence. If a ref is also a footnote, it's stored both in refs and here
+	// in notes. Slice is nil if footnotes not enabled.
+	notes []*reference
+
+	doc                  *Node
+	tip                  *Node // = doc
+	oldTip               *Node
+	lastMatchedContainer *Node // = doc
+	allClosed            bool
+}
+
+func (p *Markdown) getRef(refid string) (ref *reference, found bool) {
+	if p.referenceOverride != nil {
+		r, overridden := p.referenceOverride(refid)
+		if overridden {
+			if r == nil {
+				return nil, false
+			}
+			return &reference{
+				link:     []byte(r.Link),
+				title:    []byte(r.Title),
+				noteID:   0,
+				hasBlock: false,
+				text:     []byte(r.Text)}, true
+		}
+	}
+	// refs are case insensitive
+	ref, found = p.refs[strings.ToLower(refid)]
+	return ref, found
+}
+
+func (p *Markdown) finalize(block *Node) {
+	above := block.Parent
+	block.open = false
+	p.tip = above
+}
+
+func (p *Markdown) addChild(node NodeType, offset uint32) *Node {
+	return p.addExistingChild(NewNode(node), offset)
+}
+
+func (p *Markdown) addExistingChild(node *Node, offset uint32) *Node {
+	for !p.tip.canContain(node.Type) {
+		p.finalize(p.tip)
+	}
+	p.tip.AppendChild(node)
+	p.tip = node
+	return node
+}
+
+func (p *Markdown) closeUnmatchedBlocks() {
+	if !p.allClosed {
+		for p.oldTip != p.lastMatchedContainer {
+			parent := p.oldTip.Parent
+			p.finalize(p.oldTip)
+			p.oldTip = parent
+		}
+		p.allClosed = true
+	}
+}
+
+//
+//
+// Public interface
+//
+//
+
+// Reference represents the details of a link.
+// See the documentation in Options for more details on use-case.
+type Reference struct {
+	// Link is usually the URL the reference points to.
+	Link string
+	// Title is the alternate text describing the link in more detail.
+	Title string
+	// Text is the optional text to override the ref with if the syntax used was
+	// [refid][]
+	Text string
+}
+
+// ReferenceOverrideFunc is expected to be called with a reference string and
+// return either a valid Reference type that the reference string maps to or
+// nil. If overridden is false, the default reference logic will be executed.
+// See the documentation in Options for more details on use-case.
+type ReferenceOverrideFunc func(reference string) (ref *Reference, overridden bool)
+
+// New constructs a Markdown processor. You can use the same With* functions as
+// for Run() to customize parser's behavior and the renderer.
+func New(opts ...Option) *Markdown {
+	var p Markdown
+	for _, opt := range opts {
+		opt(&p)
+	}
+	p.refs = make(map[string]*reference)
+	p.maxNesting = 16
+	p.insideLink = false
+	docNode := NewNode(Document)
+	p.doc = docNode
+	p.tip = docNode
+	p.oldTip = docNode
+	p.lastMatchedContainer = docNode
+	p.allClosed = true
+	// register inline parsers
+	p.inlineCallback[' '] = maybeLineBreak
+	p.inlineCallback['*'] = emphasis
+	p.inlineCallback['_'] = emphasis
+	if p.extensions&Strikethrough != 0 {
+		p.inlineCallback['~'] = emphasis
+	}
+	p.inlineCallback['`'] = codeSpan
+	p.inlineCallback['\n'] = lineBreak
+	p.inlineCallback['['] = link
+	p.inlineCallback['<'] = leftAngle
+	p.inlineCallback['\\'] = escape
+	p.inlineCallback['&'] = entity
+	p.inlineCallback['!'] = maybeImage
+	p.inlineCallback['^'] = maybeInlineFootnote
+	if p.extensions&Autolink != 0 {
+		p.inlineCallback['h'] = maybeAutoLink
+		p.inlineCallback['m'] = maybeAutoLink
+		p.inlineCallback['f'] = maybeAutoLink
+		p.inlineCallback['H'] = maybeAutoLink
+		p.inlineCallback['M'] = maybeAutoLink
+		p.inlineCallback['F'] = maybeAutoLink
+	}
+	if p.extensions&Footnotes != 0 {
+		p.notes = make([]*reference, 0)
+	}
+	return &p
+}
+
+// Option customizes the Markdown processor's default behavior.
+type Option func(*Markdown)
+
+// WithRenderer allows you to override the default renderer.
+func WithRenderer(r Renderer) Option {
+	return func(p *Markdown) {
+		p.renderer = r
+	}
+}
+
+// WithExtensions allows you to pick some of the many extensions provided by
+// Blackfriday. You can bitwise OR them.
+func WithExtensions(e Extensions) Option {
+	return func(p *Markdown) {
+		p.extensions = e
+	}
+}
+
+// WithNoExtensions turns off all extensions and custom behavior.
+func WithNoExtensions() Option {
+	return func(p *Markdown) {
+		p.extensions = NoExtensions
+		p.renderer = NewHTMLRenderer(HTMLRendererParameters{
+			Flags: HTMLFlagsNone,
+		})
+	}
+}
+
+// WithRefOverride sets an optional function callback that is called every
+// time a reference is resolved.
+//
+// In Markdown, the link reference syntax can be made to resolve a link to
+// a reference instead of an inline URL, in one of the following ways:
+//
+//  * [link text][refid]
+//  * [refid][]
+//
+// Usually, the refid is defined at the bottom of the Markdown document. If
+// this override function is provided, the refid is passed to the override
+// function first, before consulting the defined refids at the bottom. If
+// the override function indicates an override did not occur, the refids at
+// the bottom will be used to fill in the link details.
+func WithRefOverride(o ReferenceOverrideFunc) Option {
+	return func(p *Markdown) {
+		p.referenceOverride = o
+	}
+}
+
+// Run is the main entry point to Blackfriday. It parses and renders a
+// block of markdown-encoded text.
+//
+// The simplest invocation of Run takes one argument, input:
+//     output := Run(input)
+// This will parse the input with CommonExtensions enabled and render it with
+// the default HTMLRenderer (with CommonHTMLFlags).
+//
+// Variadic arguments opts can customize the default behavior. Since Markdown
+// type does not contain exported fields, you can not use it directly. Instead,
+// use the With* functions. For example, this will call the most basic
+// functionality, with no extensions:
+//     output := Run(input, WithNoExtensions())
+//
+// You can use any number of With* arguments, even contradicting ones. They
+// will be applied in order of appearance and the latter will override the
+// former:
+//     output := Run(input, WithNoExtensions(), WithExtensions(exts),
+//         WithRenderer(yourRenderer))
+func Run(input []byte, opts ...Option) []byte {
+	r := NewHTMLRenderer(HTMLRendererParameters{
+		Flags: CommonHTMLFlags,
+	})
+	optList := []Option{WithRenderer(r), WithExtensions(CommonExtensions)}
+	optList = append(optList, opts...)
+	parser := New(optList...)
+	ast := parser.Parse(input)
+	var buf bytes.Buffer
+	parser.renderer.RenderHeader(&buf, ast)
+	ast.Walk(func(node *Node, entering bool) WalkStatus {
+		return parser.renderer.RenderNode(&buf, node, entering)
+	})
+	parser.renderer.RenderFooter(&buf, ast)
+	return buf.Bytes()
+}
+
+// Parse is an entry point to the parsing part of Blackfriday. It takes an
+// input markdown document and produces a syntax tree for its contents. This
+// tree can then be rendered with a default or custom renderer, or
+// analyzed/transformed by the caller to whatever non-standard needs they have.
+// The return value is the root node of the syntax tree.
+func (p *Markdown) Parse(input []byte) *Node {
+	p.block(input)
+	// Walk the tree and finish up some of unfinished blocks
+	for p.tip != nil {
+		p.finalize(p.tip)
+	}
+	// Walk the tree again and process inline markdown in each block
+	p.doc.Walk(func(node *Node, entering bool) WalkStatus {
+		if node.Type == Paragraph || node.Type == Heading || node.Type == TableCell {
+			p.inline(node, node.content)
+			node.content = nil
+		}
+		return GoToNext
+	})
+	p.parseRefsToAST()
+	return p.doc
+}
+
+func (p *Markdown) parseRefsToAST() {
+	if p.extensions&Footnotes == 0 || len(p.notes) == 0 {
+		return
+	}
+	p.tip = p.doc
+	block := p.addBlock(List, nil)
+	block.IsFootnotesList = true
+	block.ListFlags = ListTypeOrdered
+	flags := ListItemBeginningOfList
+	// Note: this loop is intentionally explicit, not range-form. This is
+	// because the body of the loop will append nested footnotes to p.notes and
+	// we need to process those late additions. Range form would only walk over
+	// the fixed initial set.
+	for i := 0; i < len(p.notes); i++ {
+		ref := p.notes[i]
+		p.addExistingChild(ref.footnote, 0)
+		block := ref.footnote
+		block.ListFlags = flags | ListTypeOrdered
+		block.RefLink = ref.link
+		if ref.hasBlock {
+			flags |= ListItemContainsBlock
+			p.block(ref.title)
+		} else {
+			p.inline(block, ref.title)
+		}
+		flags &^= ListItemBeginningOfList | ListItemContainsBlock
+	}
+	above := block.Parent
+	finalizeList(block)
+	p.tip = above
+	block.Walk(func(node *Node, entering bool) WalkStatus {
+		if node.Type == Paragraph || node.Type == Heading {
+			p.inline(node, node.content)
+			node.content = nil
+		}
+		return GoToNext
+	})
+}
+
+//
+// Link references
+//
+// This section implements support for references that (usually) appear
+// as footnotes in a document, and can be referenced anywhere in the document.
+// The basic format is:
+//
+//    [1]: http://www.google.com/ "Google"
+//    [2]: http://www.github.com/ "Github"
+//
+// Anywhere in the document, the reference can be linked by referring to its
+// label, i.e., 1 and 2 in this example, as in:
+//
+//    This library is hosted on [Github][2], a git hosting site.
+//
+// Actual footnotes as specified in Pandoc and supported by some other Markdown
+// libraries such as php-markdown are also taken care of. They look like this:
+//
+//    This sentence needs a bit of further explanation.[^note]
+//
+//    [^note]: This is the explanation.
+//
+// Footnotes should be placed at the end of the document in an ordered list.
+// Inline footnotes such as:
+//
+//    Inline footnotes^[Not supported.] also exist.
+//
+// are not yet supported.
+
+// reference holds all information necessary for a reference-style links or
+// footnotes.
+//
+// Consider this markdown with reference-style links:
+//
+//     [link][ref]
+//
+//     [ref]: /url/ "tooltip title"
+//
+// It will be ultimately converted to this HTML:
+//
+//     <p><a href=\"/url/\" title=\"title\">link</a></p>
+//
+// And a reference structure will be populated as follows:
+//
+//     p.refs["ref"] = &reference{
+//         link: "/url/",
+//         title: "tooltip title",
+//     }
+//
+// Alternatively, reference can contain information about a footnote. Consider
+// this markdown:
+//
+//     Text needing a footnote.[^a]
+//
+//     [^a]: This is the note
+//
+// A reference structure will be populated as follows:
+//
+//     p.refs["a"] = &reference{
+//         link: "a",
+//         title: "This is the note",
+//         noteID: <some positive int>,
+//     }
+//
+// TODO: As you can see, it begs for splitting into two dedicated structures
+// for refs and for footnotes.
+type reference struct {
+	link     []byte
+	title    []byte
+	noteID   int // 0 if not a footnote ref
+	hasBlock bool
+	footnote *Node // a link to the Item node within a list of footnotes
+
+	text []byte // only gets populated by refOverride feature with Reference.Text
+}
+
+func (r *reference) String() string {
+	return fmt.Sprintf("{link: %q, title: %q, text: %q, noteID: %d, hasBlock: %v}",
+		r.link, r.title, r.text, r.noteID, r.hasBlock)
+}
+
+// Check whether or not data starts with a reference link.
+// If so, it is parsed and stored in the list of references
+// (in the render struct).
+// Returns the number of bytes to skip to move past it,
+// or zero if the first line is not a reference.
+func isReference(p *Markdown, data []byte, tabSize int) int {
+	// up to 3 optional leading spaces
+	if len(data) < 4 {
+		return 0
+	}
+	i := 0
+	for i < 3 && data[i] == ' ' {
+		i++
+	}
+
+	noteID := 0
+
+	// id part: anything but a newline between brackets
+	if data[i] != '[' {
+		return 0
+	}
+	i++
+	if p.extensions&Footnotes != 0 {
+		if i < len(data) && data[i] == '^' {
+			// we can set it to anything here because the proper noteIds will
+			// be assigned later during the second pass. It just has to be != 0
+			noteID = 1
+			i++
+		}
+	}
+	idOffset := i
+	for i < len(data) && data[i] != '\n' && data[i] != '\r' && data[i] != ']' {
+		i++
+	}
+	if i >= len(data) || data[i] != ']' {
+		return 0
+	}
+	idEnd := i
+	// footnotes can have empty ID, like this: [^], but a reference can not be
+	// empty like this: []. Break early if it's not a footnote and there's no ID
+	if noteID == 0 && idOffset == idEnd {
+		return 0
+	}
+	// spacer: colon (space | tab)* newline? (space | tab)*
+	i++
+	if i >= len(data) || data[i] != ':' {
+		return 0
+	}
+	i++
+	for i < len(data) && (data[i] == ' ' || data[i] == '\t') {
+		i++
+	}
+	if i < len(data) && (data[i] == '\n' || data[i] == '\r') {
+		i++
+		if i < len(data) && data[i] == '\n' && data[i-1] == '\r' {
+			i++
+		}
+	}
+	for i < len(data) && (data[i] == ' ' || data[i] == '\t') {
+		i++
+	}
+	if i >= len(data) {
+		return 0
+	}
+
+	var (
+		linkOffset, linkEnd   int
+		titleOffset, titleEnd int
+		lineEnd               int
+		raw                   []byte
+		hasBlock              bool
+	)
+
+	if p.extensions&Footnotes != 0 && noteID != 0 {
+		linkOffset, linkEnd, raw, hasBlock = scanFootnote(p, data, i, tabSize)
+		lineEnd = linkEnd
+	} else {
+		linkOffset, linkEnd, titleOffset, titleEnd, lineEnd = scanLinkRef(p, data, i)
+	}
+	if lineEnd == 0 {
+		return 0
+	}
+
+	// a valid ref has been found
+
+	ref := &reference{
+		noteID:   noteID,
+		hasBlock: hasBlock,
+	}
+
+	if noteID > 0 {
+		// reusing the link field for the id since footnotes don't have links
+		ref.link = data[idOffset:idEnd]
+		// if footnote, it's not really a title, it's the contained text
+		ref.title = raw
+	} else {
+		ref.link = data[linkOffset:linkEnd]
+		ref.title = data[titleOffset:titleEnd]
+	}
+
+	// id matches are case-insensitive
+	id := string(bytes.ToLower(data[idOffset:idEnd]))
+
+	p.refs[id] = ref
+
+	return lineEnd
+}
+
+func scanLinkRef(p *Markdown, data []byte, i int) (linkOffset, linkEnd, titleOffset, titleEnd, lineEnd int) {
+	// link: whitespace-free sequence, optionally between angle brackets
+	if data[i] == '<' {
+		i++
+	}
+	linkOffset = i
+	for i < len(data) && data[i] != ' ' && data[i] != '\t' && data[i] != '\n' && data[i] != '\r' {
+		i++
+	}
+	linkEnd = i
+	if data[linkOffset] == '<' && data[linkEnd-1] == '>' {
+		linkOffset++
+		linkEnd--
+	}
+
+	// optional spacer: (space | tab)* (newline | '\'' | '"' | '(' )
+	for i < len(data) && (data[i] == ' ' || data[i] == '\t') {
+		i++
+	}
+	if i < len(data) && data[i] != '\n' && data[i] != '\r' && data[i] != '\'' && data[i] != '"' && data[i] != '(' {
+		return
+	}
+
+	// compute end-of-line
+	if i >= len(data) || data[i] == '\r' || data[i] == '\n' {
+		lineEnd = i
+	}
+	if i+1 < len(data) && data[i] == '\r' && data[i+1] == '\n' {
+		lineEnd++
+	}
+
+	// optional (space|tab)* spacer after a newline
+	if lineEnd > 0 {
+		i = lineEnd + 1
+		for i < len(data) && (data[i] == ' ' || data[i] == '\t') {
+			i++
+		}
+	}
+
+	// optional title: any non-newline sequence enclosed in '"() alone on its line
+	if i+1 < len(data) && (data[i] == '\'' || data[i] == '"' || data[i] == '(') {
+		i++
+		titleOffset = i
+
+		// look for EOL
+		for i < len(data) && data[i] != '\n' && data[i] != '\r' {
+			i++
+		}
+		if i+1 < len(data) && data[i] == '\n' && data[i+1] == '\r' {
+			titleEnd = i + 1
+		} else {
+			titleEnd = i
+		}
+
+		// step back
+		i--
+		for i > titleOffset && (data[i] == ' ' || data[i] == '\t') {
+			i--
+		}
+		if i > titleOffset && (data[i] == '\'' || data[i] == '"' || data[i] == ')') {
+			lineEnd = titleEnd
+			titleEnd = i
+		}
+	}
+
+	return
+}
+
+// The first bit of this logic is the same as Parser.listItem, but the rest
+// is much simpler. This function simply finds the entire block and shifts it
+// over by one tab if it is indeed a block (just returns the line if it's not).
+// blockEnd is the end of the section in the input buffer, and contents is the
+// extracted text that was shifted over one tab. It will need to be rendered at
+// the end of the document.
+func scanFootnote(p *Markdown, data []byte, i, indentSize int) (blockStart, blockEnd int, contents []byte, hasBlock bool) {
+	if i == 0 || len(data) == 0 {
+		return
+	}
+
+	// skip leading whitespace on first line
+	for i < len(data) && data[i] == ' ' {
+		i++
+	}
+
+	blockStart = i
+
+	// find the end of the line
+	blockEnd = i
+	for i < len(data) && data[i-1] != '\n' {
+		i++
+	}
+
+	// get working buffer
+	var raw bytes.Buffer
+
+	// put the first line into the working buffer
+	raw.Write(data[blockEnd:i])
+	blockEnd = i
+
+	// process the following lines
+	containsBlankLine := false
+
+gatherLines:
+	for blockEnd < len(data) {
+		i++
+
+		// find the end of this line
+		for i < len(data) && data[i-1] != '\n' {
+			i++
+		}
+
+		// if it is an empty line, guess that it is part of this item
+		// and move on to the next line
+		if p.isEmpty(data[blockEnd:i]) > 0 {
+			containsBlankLine = true
+			blockEnd = i
+			continue
+		}
+
+		n := 0
+		if n = isIndented(data[blockEnd:i], indentSize); n == 0 {
+			// this is the end of the block.
+			// we don't want to include this last line in the index.
+			break gatherLines
+		}
+
+		// if there were blank lines before this one, insert a new one now
+		if containsBlankLine {
+			raw.WriteByte('\n')
+			containsBlankLine = false
+		}
+
+		// get rid of that first tab, write to buffer
+		raw.Write(data[blockEnd+n : i])
+		hasBlock = true
+
+		blockEnd = i
+	}
+
+	if data[blockEnd-1] != '\n' {
+		raw.WriteByte('\n')
+	}
+
+	contents = raw.Bytes()
+
+	return
+}
+
+//
+//
+// Miscellaneous helper functions
+//
+//
+
+// Test if a character is a punctuation symbol.
+// Taken from a private function in regexp in the stdlib.
+func ispunct(c byte) bool {
+	for _, r := range []byte("!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~") {
+		if c == r {
+			return true
+		}
+	}
+	return false
+}
+
+// Test if a character is a whitespace character.
+func isspace(c byte) bool {
+	return c == ' ' || c == '\t' || c == '\n' || c == '\r' || c == '\f' || c == '\v'
+}
+
+// Test if a character is letter.
+func isletter(c byte) bool {
+	return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
+}
+
+// Test if a character is a letter or a digit.
+// TODO: check when this is looking for ASCII alnum and when it should use unicode
+func isalnum(c byte) bool {
+	return (c >= '0' && c <= '9') || isletter(c)
+}
+
+// Replace tab characters with spaces, aligning to the next TAB_SIZE column.
+// always ends output with a newline
+func expandTabs(out *bytes.Buffer, line []byte, tabSize int) {
+	// first, check for common cases: no tabs, or only tabs at beginning of line
+	i, prefix := 0, 0
+	slowcase := false
+	for i = 0; i < len(line); i++ {
+		if line[i] == '\t' {
+			if prefix == i {
+				prefix++
+			} else {
+				slowcase = true
+				break
+			}
+		}
+	}
+
+	// no need to decode runes if all tabs are at the beginning of the line
+	if !slowcase {
+		for i = 0; i < prefix*tabSize; i++ {
+			out.WriteByte(' ')
+		}
+		out.Write(line[prefix:])
+		return
+	}
+
+	// the slow case: we need to count runes to figure out how
+	// many spaces to insert for each tab
+	column := 0
+	i = 0
+	for i < len(line) {
+		start := i
+		for i < len(line) && line[i] != '\t' {
+			_, size := utf8.DecodeRune(line[i:])
+			i += size
+			column++
+		}
+
+		if i > start {
+			out.Write(line[start:i])
+		}
+
+		if i >= len(line) {
+			break
+		}
+
+		for {
+			out.WriteByte(' ')
+			column++
+			if column%tabSize == 0 {
+				break
+			}
+		}
+
+		i++
+	}
+}
+
+// Find if a line counts as indented or not.
+// Returns number of characters the indent is (0 = not indented).
+func isIndented(data []byte, indentSize int) int {
+	if len(data) == 0 {
+		return 0
+	}
+	if data[0] == '\t' {
+		return 1
+	}
+	if len(data) < indentSize {
+		return 0
+	}
+	for i := 0; i < indentSize; i++ {
+		if data[i] != ' ' {
+			return 0
+		}
+	}
+	return indentSize
+}
+
+// Create a url-safe slug for fragments
+func slugify(in []byte) []byte {
+	if len(in) == 0 {
+		return in
+	}
+	out := make([]byte, 0, len(in))
+	sym := false
+
+	for _, ch := range in {
+		if isalnum(ch) {
+			sym = false
+			out = append(out, ch)
+		} else if sym {
+			continue
+		} else {
+			out = append(out, '-')
+			sym = true
+		}
+	}
+	var a, b int
+	var ch byte
+	for a, ch = range out {
+		if ch != '-' {
+			break
+		}
+	}
+	for b = len(out) - 1; b > 0; b-- {
+		if out[b] != '-' {
+			break
+		}
+	}
+	return out[a : b+1]
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/node.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/node.go
new file mode 100644
index 00000000..51b9e8c1
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/node.go
@@ -0,0 +1,354 @@
+package blackfriday
+
+import (
+	"bytes"
+	"fmt"
+)
+
+// NodeType specifies a type of a single node of a syntax tree. Usually one
+// node (and its type) corresponds to a single markdown feature, e.g. emphasis
+// or code block.
+type NodeType int
+
+// Constants for identifying different types of nodes. See NodeType.
+const (
+	Document NodeType = iota
+	BlockQuote
+	List
+	Item
+	Paragraph
+	Heading
+	HorizontalRule
+	Emph
+	Strong
+	Del
+	Link
+	Image
+	Text
+	HTMLBlock
+	CodeBlock
+	Softbreak
+	Hardbreak
+	Code
+	HTMLSpan
+	Table
+	TableCell
+	TableHead
+	TableBody
+	TableRow
+)
+
+var nodeTypeNames = []string{
+	Document:       "Document",
+	BlockQuote:     "BlockQuote",
+	List:           "List",
+	Item:           "Item",
+	Paragraph:      "Paragraph",
+	Heading:        "Heading",
+	HorizontalRule: "HorizontalRule",
+	Emph:           "Emph",
+	Strong:         "Strong",
+	Del:            "Del",
+	Link:           "Link",
+	Image:          "Image",
+	Text:           "Text",
+	HTMLBlock:      "HTMLBlock",
+	CodeBlock:      "CodeBlock",
+	Softbreak:      "Softbreak",
+	Hardbreak:      "Hardbreak",
+	Code:           "Code",
+	HTMLSpan:       "HTMLSpan",
+	Table:          "Table",
+	TableCell:      "TableCell",
+	TableHead:      "TableHead",
+	TableBody:      "TableBody",
+	TableRow:       "TableRow",
+}
+
+func (t NodeType) String() string {
+	return nodeTypeNames[t]
+}
+
+// ListData contains fields relevant to a List and Item node type.
+type ListData struct {
+	ListFlags       ListType
+	Tight           bool   // Skip <p>s around list item data if true
+	BulletChar      byte   // '*', '+' or '-' in bullet lists
+	Delimiter       byte   // '.' or ')' after the number in ordered lists
+	RefLink         []byte // If not nil, turns this list item into a footnote item and triggers different rendering
+	IsFootnotesList bool   // This is a list of footnotes
+}
+
+// LinkData contains fields relevant to a Link node type.
+type LinkData struct {
+	Destination []byte // Destination is what goes into a href
+	Title       []byte // Title is the tooltip thing that goes in a title attribute
+	NoteID      int    // NoteID contains a serial number of a footnote, zero if it's not a footnote
+	Footnote    *Node  // If it's a footnote, this is a direct link to the footnote Node. Otherwise nil.
+}
+
+// CodeBlockData contains fields relevant to a CodeBlock node type.
+type CodeBlockData struct {
+	IsFenced    bool   // Specifies whether it's a fenced code block or an indented one
+	Info        []byte // This holds the info string
+	FenceChar   byte
+	FenceLength int
+	FenceOffset int
+}
+
+// TableCellData contains fields relevant to a TableCell node type.
+type TableCellData struct {
+	IsHeader bool           // This tells if it's under the header row
+	Align    CellAlignFlags // This holds the value for align attribute
+}
+
+// HeadingData contains fields relevant to a Heading node type.
+type HeadingData struct {
+	Level        int    // This holds the heading level number
+	HeadingID    string // This might hold heading ID, if present
+	IsTitleblock bool   // Specifies whether it's a title block
+}
+
+// Node is a single element in the abstract syntax tree of the parsed document.
+// It holds connections to the structurally neighboring nodes and, for certain
+// types of nodes, additional information that might be needed when rendering.
+type Node struct {
+	Type       NodeType // Determines the type of the node
+	Parent     *Node    // Points to the parent
+	FirstChild *Node    // Points to the first child, if any
+	LastChild  *Node    // Points to the last child, if any
+	Prev       *Node    // Previous sibling; nil if it's the first child
+	Next       *Node    // Next sibling; nil if it's the last child
+
+	Literal []byte // Text contents of the leaf nodes
+
+	HeadingData   // Populated if Type is Heading
+	ListData      // Populated if Type is List
+	CodeBlockData // Populated if Type is CodeBlock
+	LinkData      // Populated if Type is Link
+	TableCellData // Populated if Type is TableCell
+
+	content []byte // Markdown content of the block nodes
+	open    bool   // Specifies an open block node that has not been finished to process yet
+}
+
+// NewNode allocates a node of a specified type.
+func NewNode(typ NodeType) *Node {
+	return &Node{
+		Type: typ,
+		open: true,
+	}
+}
+
+func (n *Node) String() string {
+	ellipsis := ""
+	snippet := n.Literal
+	if len(snippet) > 16 {
+		snippet = snippet[:16]
+		ellipsis = "..."
+	}
+	return fmt.Sprintf("%s: '%s%s'", n.Type, snippet, ellipsis)
+}
+
+// Unlink removes node 'n' from the tree.
+// It panics if the node is nil.
+func (n *Node) Unlink() {
+	if n.Prev != nil {
+		n.Prev.Next = n.Next
+	} else if n.Parent != nil {
+		n.Parent.FirstChild = n.Next
+	}
+	if n.Next != nil {
+		n.Next.Prev = n.Prev
+	} else if n.Parent != nil {
+		n.Parent.LastChild = n.Prev
+	}
+	n.Parent = nil
+	n.Next = nil
+	n.Prev = nil
+}
+
+// AppendChild adds a node 'child' as a child of 'n'.
+// It panics if either node is nil.
+func (n *Node) AppendChild(child *Node) {
+	child.Unlink()
+	child.Parent = n
+	if n.LastChild != nil {
+		n.LastChild.Next = child
+		child.Prev = n.LastChild
+		n.LastChild = child
+	} else {
+		n.FirstChild = child
+		n.LastChild = child
+	}
+}
+
+// InsertBefore inserts 'sibling' immediately before 'n'.
+// It panics if either node is nil.
+func (n *Node) InsertBefore(sibling *Node) {
+	sibling.Unlink()
+	sibling.Prev = n.Prev
+	if sibling.Prev != nil {
+		sibling.Prev.Next = sibling
+	}
+	sibling.Next = n
+	n.Prev = sibling
+	sibling.Parent = n.Parent
+	if sibling.Prev == nil {
+		sibling.Parent.FirstChild = sibling
+	}
+}
+
+func (n *Node) isContainer() bool {
+	switch n.Type {
+	case Document:
+		fallthrough
+	case BlockQuote:
+		fallthrough
+	case List:
+		fallthrough
+	case Item:
+		fallthrough
+	case Paragraph:
+		fallthrough
+	case Heading:
+		fallthrough
+	case Emph:
+		fallthrough
+	case Strong:
+		fallthrough
+	case Del:
+		fallthrough
+	case Link:
+		fallthrough
+	case Image:
+		fallthrough
+	case Table:
+		fallthrough
+	case TableHead:
+		fallthrough
+	case TableBody:
+		fallthrough
+	case TableRow:
+		fallthrough
+	case TableCell:
+		return true
+	default:
+		return false
+	}
+}
+
+func (n *Node) canContain(t NodeType) bool {
+	if n.Type == List {
+		return t == Item
+	}
+	if n.Type == Document || n.Type == BlockQuote || n.Type == Item {
+		return t != Item
+	}
+	if n.Type == Table {
+		return t == TableHead || t == TableBody
+	}
+	if n.Type == TableHead || n.Type == TableBody {
+		return t == TableRow
+	}
+	if n.Type == TableRow {
+		return t == TableCell
+	}
+	return false
+}
+
+// WalkStatus allows NodeVisitor to have some control over the tree traversal.
+// It is returned from NodeVisitor and different values allow Node.Walk to
+// decide which node to go to next.
+type WalkStatus int
+
+const (
+	// GoToNext is the default traversal of every node.
+	GoToNext WalkStatus = iota
+	// SkipChildren tells walker to skip all children of current node.
+	SkipChildren
+	// Terminate tells walker to terminate the traversal.
+	Terminate
+)
+
+// NodeVisitor is a callback to be called when traversing the syntax tree.
+// Called twice for every node: once with entering=true when the branch is
+// first visited, then with entering=false after all the children are done.
+type NodeVisitor func(node *Node, entering bool) WalkStatus
+
+// Walk is a convenience method that instantiates a walker and starts a
+// traversal of subtree rooted at n.
+func (n *Node) Walk(visitor NodeVisitor) {
+	w := newNodeWalker(n)
+	for w.current != nil {
+		status := visitor(w.current, w.entering)
+		switch status {
+		case GoToNext:
+			w.next()
+		case SkipChildren:
+			w.entering = false
+			w.next()
+		case Terminate:
+			return
+		}
+	}
+}
+
+type nodeWalker struct {
+	current  *Node
+	root     *Node
+	entering bool
+}
+
+func newNodeWalker(root *Node) *nodeWalker {
+	return &nodeWalker{
+		current:  root,
+		root:     root,
+		entering: true,
+	}
+}
+
+func (nw *nodeWalker) next() {
+	if (!nw.current.isContainer() || !nw.entering) && nw.current == nw.root {
+		nw.current = nil
+		return
+	}
+	if nw.entering && nw.current.isContainer() {
+		if nw.current.FirstChild != nil {
+			nw.current = nw.current.FirstChild
+			nw.entering = true
+		} else {
+			nw.entering = false
+		}
+	} else if nw.current.Next == nil {
+		nw.current = nw.current.Parent
+		nw.entering = false
+	} else {
+		nw.current = nw.current.Next
+		nw.entering = true
+	}
+}
+
+func dump(ast *Node) {
+	fmt.Println(dumpString(ast))
+}
+
+func dumpR(ast *Node, depth int) string {
+	if ast == nil {
+		return ""
+	}
+	indent := bytes.Repeat([]byte("\t"), depth)
+	content := ast.Literal
+	if content == nil {
+		content = ast.content
+	}
+	result := fmt.Sprintf("%s%s(%q)\n", indent, ast.Type, content)
+	for n := ast.FirstChild; n != nil; n = n.Next {
+		result += dumpR(n, depth+1)
+	}
+	return result
+}
+
+func dumpString(ast *Node) string {
+	return dumpR(ast, 0)
+}
diff --git a/backend/vendor/gopkg.in/russross/blackfriday.v2/smartypants.go b/backend/vendor/gopkg.in/russross/blackfriday.v2/smartypants.go
new file mode 100644
index 00000000..3a220e94
--- /dev/null
+++ b/backend/vendor/gopkg.in/russross/blackfriday.v2/smartypants.go
@@ -0,0 +1,457 @@
+//
+// Blackfriday Markdown Processor
+// Available at http://github.com/russross/blackfriday
+//
+// Copyright © 2011 Russ Ross <russ@russross.com>.
+// Distributed under the Simplified BSD License.
+// See README.md for details.
+//
+
+//
+//
+// SmartyPants rendering
+//
+//
+
+package blackfriday
+
+import (
+	"bytes"
+	"io"
+)
+
+// SPRenderer is a struct containing state of a Smartypants renderer.
+type SPRenderer struct {
+	inSingleQuote bool
+	inDoubleQuote bool
+	callbacks     [256]smartCallback
+}
+
+func wordBoundary(c byte) bool {
+	return c == 0 || isspace(c) || ispunct(c)
+}
+
+func tolower(c byte) byte {
+	if c >= 'A' && c <= 'Z' {
+		return c - 'A' + 'a'
+	}
+	return c
+}
+
+func isdigit(c byte) bool {
+	return c >= '0' && c <= '9'
+}
+
+func smartQuoteHelper(out *bytes.Buffer, previousChar byte, nextChar byte, quote byte, isOpen *bool, addNBSP bool) bool {
+	// edge of the buffer is likely to be a tag that we don't get to see,
+	// so we treat it like text sometimes
+
+	// enumerate all sixteen possibilities for (previousChar, nextChar)
+	// each can be one of {0, space, punct, other}
+	switch {
+	case previousChar == 0 && nextChar == 0:
+		// context is not any help here, so toggle
+		*isOpen = !*isOpen
+	case isspace(previousChar) && nextChar == 0:
+		// [ "] might be [ "<code>foo...]
+		*isOpen = true
+	case ispunct(previousChar) && nextChar == 0:
+		// [!"] hmm... could be [Run!"] or [("<code>...]
+		*isOpen = false
+	case /* isnormal(previousChar) && */ nextChar == 0:
+		// [a"] is probably a close
+		*isOpen = false
+	case previousChar == 0 && isspace(nextChar):
+		// [" ] might be [...foo</code>" ]
+		*isOpen = false
+	case isspace(previousChar) && isspace(nextChar):
+		// [ " ] context is not any help here, so toggle
+		*isOpen = !*isOpen
+	case ispunct(previousChar) && isspace(nextChar):
+		// [!" ] is probably a close
+		*isOpen = false
+	case /* isnormal(previousChar) && */ isspace(nextChar):
+		// [a" ] this is one of the easy cases
+		*isOpen = false
+	case previousChar == 0 && ispunct(nextChar):
+		// ["!] hmm... could be ["$1.95] or [</code>"!...]
+		*isOpen = false
+	case isspace(previousChar) && ispunct(nextChar):
+		// [ "!] looks more like [ "$1.95]
+		*isOpen = true
+	case ispunct(previousChar) && ispunct(nextChar):
+		// [!"!] context is not any help here, so toggle
+		*isOpen = !*isOpen
+	case /* isnormal(previousChar) && */ ispunct(nextChar):
+		// [a"!] is probably a close
+		*isOpen = false
+	case previousChar == 0 /* && isnormal(nextChar) */ :
+		// ["a] is probably an open
+		*isOpen = true
+	case isspace(previousChar) /* && isnormal(nextChar) */ :
+		// [ "a] this is one of the easy cases
+		*isOpen = true
+	case ispunct(previousChar) /* && isnormal(nextChar) */ :
+		// [!"a] is probably an open
+		*isOpen = true
+	default:
+		// [a'b] maybe a contraction?
+		*isOpen = false
+	}
+
+	// Note that with the limited lookahead, this non-breaking
+	// space will also be appended to single double quotes.
+	if addNBSP && !*isOpen {
+		out.WriteString("&nbsp;")
+	}
+
+	out.WriteByte('&')
+	if *isOpen {
+		out.WriteByte('l')
+	} else {
+		out.WriteByte('r')
+	}
+	out.WriteByte(quote)
+	out.WriteString("quo;")
+
+	if addNBSP && *isOpen {
+		out.WriteString("&nbsp;")
+	}
+
+	return true
+}
+
+func (r *SPRenderer) smartSingleQuote(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 2 {
+		t1 := tolower(text[1])
+
+		if t1 == '\'' {
+			nextChar := byte(0)
+			if len(text) >= 3 {
+				nextChar = text[2]
+			}
+			if smartQuoteHelper(out, previousChar, nextChar, 'd', &r.inDoubleQuote, false) {
+				return 1
+			}
+		}
+
+		if (t1 == 's' || t1 == 't' || t1 == 'm' || t1 == 'd') && (len(text) < 3 || wordBoundary(text[2])) {
+			out.WriteString("&rsquo;")
+			return 0
+		}
+
+		if len(text) >= 3 {
+			t2 := tolower(text[2])
+
+			if ((t1 == 'r' && t2 == 'e') || (t1 == 'l' && t2 == 'l') || (t1 == 'v' && t2 == 'e')) &&
+				(len(text) < 4 || wordBoundary(text[3])) {
+				out.WriteString("&rsquo;")
+				return 0
+			}
+		}
+	}
+
+	nextChar := byte(0)
+	if len(text) > 1 {
+		nextChar = text[1]
+	}
+	if smartQuoteHelper(out, previousChar, nextChar, 's', &r.inSingleQuote, false) {
+		return 0
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartParens(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 3 {
+		t1 := tolower(text[1])
+		t2 := tolower(text[2])
+
+		if t1 == 'c' && t2 == ')' {
+			out.WriteString("&copy;")
+			return 2
+		}
+
+		if t1 == 'r' && t2 == ')' {
+			out.WriteString("&reg;")
+			return 2
+		}
+
+		if len(text) >= 4 && t1 == 't' && t2 == 'm' && text[3] == ')' {
+			out.WriteString("&trade;")
+			return 3
+		}
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartDash(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 2 {
+		if text[1] == '-' {
+			out.WriteString("&mdash;")
+			return 1
+		}
+
+		if wordBoundary(previousChar) && wordBoundary(text[1]) {
+			out.WriteString("&ndash;")
+			return 0
+		}
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartDashLatex(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 3 && text[1] == '-' && text[2] == '-' {
+		out.WriteString("&mdash;")
+		return 2
+	}
+	if len(text) >= 2 && text[1] == '-' {
+		out.WriteString("&ndash;")
+		return 1
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartAmpVariant(out *bytes.Buffer, previousChar byte, text []byte, quote byte, addNBSP bool) int {
+	if bytes.HasPrefix(text, []byte("&quot;")) {
+		nextChar := byte(0)
+		if len(text) >= 7 {
+			nextChar = text[6]
+		}
+		if smartQuoteHelper(out, previousChar, nextChar, quote, &r.inDoubleQuote, addNBSP) {
+			return 5
+		}
+	}
+
+	if bytes.HasPrefix(text, []byte("&#0;")) {
+		return 3
+	}
+
+	out.WriteByte('&')
+	return 0
+}
+
+func (r *SPRenderer) smartAmp(angledQuotes, addNBSP bool) func(*bytes.Buffer, byte, []byte) int {
+	var quote byte = 'd'
+	if angledQuotes {
+		quote = 'a'
+	}
+
+	return func(out *bytes.Buffer, previousChar byte, text []byte) int {
+		return r.smartAmpVariant(out, previousChar, text, quote, addNBSP)
+	}
+}
+
+func (r *SPRenderer) smartPeriod(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 3 && text[1] == '.' && text[2] == '.' {
+		out.WriteString("&hellip;")
+		return 2
+	}
+
+	if len(text) >= 5 && text[1] == ' ' && text[2] == '.' && text[3] == ' ' && text[4] == '.' {
+		out.WriteString("&hellip;")
+		return 4
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartBacktick(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if len(text) >= 2 && text[1] == '`' {
+		nextChar := byte(0)
+		if len(text) >= 3 {
+			nextChar = text[2]
+		}
+		if smartQuoteHelper(out, previousChar, nextChar, 'd', &r.inDoubleQuote, false) {
+			return 1
+		}
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartNumberGeneric(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if wordBoundary(previousChar) && previousChar != '/' && len(text) >= 3 {
+		// is it of the form digits/digits(word boundary)?, i.e., \d+/\d+\b
+		// note: check for regular slash (/) or fraction slash (⁄, 0x2044, or 0xe2 81 84 in utf-8)
+		//       and avoid changing dates like 1/23/2005 into fractions.
+		numEnd := 0
+		for len(text) > numEnd && isdigit(text[numEnd]) {
+			numEnd++
+		}
+		if numEnd == 0 {
+			out.WriteByte(text[0])
+			return 0
+		}
+		denStart := numEnd + 1
+		if len(text) > numEnd+3 && text[numEnd] == 0xe2 && text[numEnd+1] == 0x81 && text[numEnd+2] == 0x84 {
+			denStart = numEnd + 3
+		} else if len(text) < numEnd+2 || text[numEnd] != '/' {
+			out.WriteByte(text[0])
+			return 0
+		}
+		denEnd := denStart
+		for len(text) > denEnd && isdigit(text[denEnd]) {
+			denEnd++
+		}
+		if denEnd == denStart {
+			out.WriteByte(text[0])
+			return 0
+		}
+		if len(text) == denEnd || wordBoundary(text[denEnd]) && text[denEnd] != '/' {
+			out.WriteString("<sup>")
+			out.Write(text[:numEnd])
+			out.WriteString("</sup>&frasl;<sub>")
+			out.Write(text[denStart:denEnd])
+			out.WriteString("</sub>")
+			return denEnd - 1
+		}
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartNumber(out *bytes.Buffer, previousChar byte, text []byte) int {
+	if wordBoundary(previousChar) && previousChar != '/' && len(text) >= 3 {
+		if text[0] == '1' && text[1] == '/' && text[2] == '2' {
+			if len(text) < 4 || wordBoundary(text[3]) && text[3] != '/' {
+				out.WriteString("&frac12;")
+				return 2
+			}
+		}
+
+		if text[0] == '1' && text[1] == '/' && text[2] == '4' {
+			if len(text) < 4 || wordBoundary(text[3]) && text[3] != '/' || (len(text) >= 5 && tolower(text[3]) == 't' && tolower(text[4]) == 'h') {
+				out.WriteString("&frac14;")
+				return 2
+			}
+		}
+
+		if text[0] == '3' && text[1] == '/' && text[2] == '4' {
+			if len(text) < 4 || wordBoundary(text[3]) && text[3] != '/' || (len(text) >= 6 && tolower(text[3]) == 't' && tolower(text[4]) == 'h' && tolower(text[5]) == 's') {
+				out.WriteString("&frac34;")
+				return 2
+			}
+		}
+	}
+
+	out.WriteByte(text[0])
+	return 0
+}
+
+func (r *SPRenderer) smartDoubleQuoteVariant(out *bytes.Buffer, previousChar byte, text []byte, quote byte) int {
+	nextChar := byte(0)
+	if len(text) > 1 {
+		nextChar = text[1]
+	}
+	if !smartQuoteHelper(out, previousChar, nextChar, quote, &r.inDoubleQuote, false) {
+		out.WriteString("&quot;")
+	}
+
+	return 0
+}
+
+func (r *SPRenderer) smartDoubleQuote(out *bytes.Buffer, previousChar byte, text []byte) int {
+	return r.smartDoubleQuoteVariant(out, previousChar, text, 'd')
+}
+
+func (r *SPRenderer) smartAngledDoubleQuote(out *bytes.Buffer, previousChar byte, text []byte) int {
+	return r.smartDoubleQuoteVariant(out, previousChar, text, 'a')
+}
+
+func (r *SPRenderer) smartLeftAngle(out *bytes.Buffer, previousChar byte, text []byte) int {
+	i := 0
+
+	for i < len(text) && text[i] != '>' {
+		i++
+	}
+
+	out.Write(text[:i+1])
+	return i
+}
+
+type smartCallback func(out *bytes.Buffer, previousChar byte, text []byte) int
+
+// NewSmartypantsRenderer constructs a Smartypants renderer object.
+func NewSmartypantsRenderer(flags HTMLFlags) *SPRenderer {
+	var (
+		r SPRenderer
+
+		smartAmpAngled      = r.smartAmp(true, false)
+		smartAmpAngledNBSP  = r.smartAmp(true, true)
+		smartAmpRegular     = r.smartAmp(false, false)
+		smartAmpRegularNBSP = r.smartAmp(false, true)
+
+		addNBSP = flags&SmartypantsQuotesNBSP != 0
+	)
+
+	if flags&SmartypantsAngledQuotes == 0 {
+		r.callbacks['"'] = r.smartDoubleQuote
+		if !addNBSP {
+			r.callbacks['&'] = smartAmpRegular
+		} else {
+			r.callbacks['&'] = smartAmpRegularNBSP
+		}
+	} else {
+		r.callbacks['"'] = r.smartAngledDoubleQuote
+		if !addNBSP {
+			r.callbacks['&'] = smartAmpAngled
+		} else {
+			r.callbacks['&'] = smartAmpAngledNBSP
+		}
+	}
+	r.callbacks['\''] = r.smartSingleQuote
+	r.callbacks['('] = r.smartParens
+	if flags&SmartypantsDashes != 0 {
+		if flags&SmartypantsLatexDashes == 0 {
+			r.callbacks['-'] = r.smartDash
+		} else {
+			r.callbacks['-'] = r.smartDashLatex
+		}
+	}
+	r.callbacks['.'] = r.smartPeriod
+	if flags&SmartypantsFractions == 0 {
+		r.callbacks['1'] = r.smartNumber
+		r.callbacks['3'] = r.smartNumber
+	} else {
+		for ch := '1'; ch <= '9'; ch++ {
+			r.callbacks[ch] = r.smartNumberGeneric
+		}
+	}
+	r.callbacks['<'] = r.smartLeftAngle
+	r.callbacks['`'] = r.smartBacktick
+	return &r
+}
+
+// Process is the entry point of the Smartypants renderer.
+func (r *SPRenderer) Process(w io.Writer, text []byte) {
+	mark := 0
+	for i := 0; i < len(text); i++ {
+		if action := r.callbacks[text[i]]; action != nil {
+			if i > mark {
+				w.Write(text[mark:i])
+			}
+			previousChar := byte(0)
+			if i > 0 {
+				previousChar = text[i-1]
+			}
+			var tmp bytes.Buffer
+			i += action(&tmp, previousChar, text[i:])
+			w.Write(tmp.Bytes())
+			mark = i + 1
+		}
+	}
+	if mark < len(text) {
+		w.Write(text[mark:])
+	}
+}
diff --git a/backend/vendor/modules.txt b/backend/vendor/modules.txt
index c6a30b46..b8784ecb 100644
--- a/backend/vendor/modules.txt
+++ b/backend/vendor/modules.txt
@@ -1,3 +1,9 @@
+# github.com/Masterminds/semver v1.4.2
+github.com/Masterminds/semver
+# github.com/Masterminds/sprig v2.16.0+incompatible
+github.com/Masterminds/sprig
+# github.com/aokoli/goutils v1.0.1
+github.com/aokoli/goutils
 # github.com/apex/log v1.1.1
 github.com/apex/log
 # github.com/dgrijalva/jwt-go v3.2.0+incompatible
@@ -27,6 +33,8 @@ github.com/golang/protobuf/proto
 # github.com/gomodule/redigo v2.0.0+incompatible
 github.com/gomodule/redigo/internal
 github.com/gomodule/redigo/redis
+# github.com/google/uuid v1.1.1
+github.com/google/uuid
 # github.com/gopherjs/gopherjs v0.0.0-20181017120253-0766667cb4d1
 github.com/gopherjs/gopherjs/js
 # github.com/hashicorp/hcl v1.0.0
@@ -40,8 +48,14 @@ github.com/hashicorp/hcl/hcl/token
 github.com/hashicorp/hcl/json/parser
 github.com/hashicorp/hcl/json/scanner
 github.com/hashicorp/hcl/json/token
+# github.com/huandu/xstrings v1.2.0
+github.com/huandu/xstrings
+# github.com/imdario/mergo v0.3.6
+github.com/imdario/mergo
 # github.com/imroc/req v0.2.4
 github.com/imroc/req
+# github.com/jaytaylor/html2text v0.0.0-20180606194806-57d518f124b0
+github.com/jaytaylor/html2text
 # github.com/json-iterator/go v1.1.6
 github.com/json-iterator/go
 # github.com/jtolds/gls v4.20.0+incompatible
@@ -50,20 +64,28 @@ github.com/jtolds/gls
 github.com/leodido/go-urn
 # github.com/magiconair/properties v1.8.0
 github.com/magiconair/properties
+# github.com/matcornic/hermes v1.2.0
+github.com/matcornic/hermes
 # github.com/mattn/go-isatty v0.0.8
 github.com/mattn/go-isatty
+# github.com/mattn/go-runewidth v0.0.3
+github.com/mattn/go-runewidth
 # github.com/mitchellh/mapstructure v1.1.2
 github.com/mitchellh/mapstructure
 # github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd
 github.com/modern-go/concurrent
 # github.com/modern-go/reflect2 v1.0.1
 github.com/modern-go/reflect2
+# github.com/olekukonko/tablewriter v0.0.1
+github.com/olekukonko/tablewriter
 # github.com/pelletier/go-toml v1.2.0
 github.com/pelletier/go-toml
 # github.com/pkg/errors v0.8.1
 github.com/pkg/errors
 # github.com/satori/go.uuid v1.2.0
 github.com/satori/go.uuid
+# github.com/shurcooL/sanitized_anchor_name v1.0.0
+github.com/shurcooL/sanitized_anchor_name
 # github.com/smartystreets/assertions v1.0.0
 github.com/smartystreets/assertions
 github.com/smartystreets/assertions/internal/go-diff/diffmatchpatch
@@ -84,8 +106,16 @@ github.com/spf13/jwalterweatherman
 github.com/spf13/pflag
 # github.com/spf13/viper v1.4.0
 github.com/spf13/viper
+# github.com/ssor/bom v0.0.0-20170718123548-6386211fdfcf
+github.com/ssor/bom
 # github.com/ugorji/go v1.1.4
 github.com/ugorji/go/codec
+# golang.org/x/crypto v0.0.0-20190426145343-a29dc8fdc734
+golang.org/x/crypto/pbkdf2
+golang.org/x/crypto/scrypt
+# golang.org/x/net v0.0.0-20190620200207-3b0461eec859
+golang.org/x/net/html
+golang.org/x/net/html/atom
 # golang.org/x/sys v0.0.0-20190412213103-97732733099d
 golang.org/x/sys/unix
 # golang.org/x/text v0.3.0
@@ -95,5 +125,7 @@ golang.org/x/text/unicode/norm
 gopkg.in/go-playground/validator.v8
 # gopkg.in/go-playground/validator.v9 v9.29.1
 gopkg.in/go-playground/validator.v9
+# gopkg.in/russross/blackfriday.v2 v2.0.0
+gopkg.in/russross/blackfriday.v2
 # gopkg.in/yaml.v2 v2.2.2
 gopkg.in/yaml.v2

From c67c1be52ac8f88992e20d3f988037bb207fdfdb Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 19:56:13 +0800
Subject: [PATCH 05/42] =?UTF-8?q?=E5=8A=A0=E5=85=A5=E7=94=A8=E6=88=B7?=
 =?UTF-8?q?=E8=AE=BE=E7=BD=AE=E4=B8=8E=E9=82=AE=E4=BB=B6=E9=80=9A=E7=9F=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/conf/config.yml                |  11 ++-
 backend/constants/notification.go      |   7 ++
 backend/go.mod                         |   2 +
 backend/go.sum                         |   4 +
 backend/main.go                        |   1 +
 backend/model/schedule.go              |  33 +------
 backend/model/task.go                  |   1 +
 backend/model/user.go                  |   6 ++
 backend/routes/schedule.go             |   3 +
 backend/routes/task.go                 |   3 +
 backend/routes/user.go                 |  38 ++++++++
 backend/services/notification/mail.go  | 130 ++++++++++++++++++++++++-
 backend/services/schedule.go           |   3 +
 backend/services/task.go               |  27 +++++
 backend/services/user.go               |   6 ++
 backend/vendor/modules.txt             |   4 +
 frontend/src/i18n/zh.js                |  12 +++
 frontend/src/router/index.js           |  18 ++++
 frontend/src/store/modules/user.js     |   5 +
 frontend/src/views/login/index.vue     |  11 ++-
 frontend/src/views/setting/Setting.vue | 102 +++++++++++++++++++
 frontend/src/views/user/UserList.vue   |  16 ++-
 22 files changed, 405 insertions(+), 38 deletions(-)
 create mode 100644 backend/constants/notification.go
 create mode 100644 frontend/src/views/setting/Setting.vue

diff --git a/backend/conf/config.yml b/backend/conf/config.yml
index a51a34c9..6e3e0611 100644
--- a/backend/conf/config.yml
+++ b/backend/conf/config.yml
@@ -37,4 +37,13 @@ other:
   tmppath: "/tmp"
 version: 0.4.3
 setting:
-  allowRegister: "N"
\ No newline at end of file
+  allowRegister: "N"
+notification:
+  mail:
+    server: ''
+    port: ''
+    senderEmail: ''
+    senderIdentity: ''
+    smtp:
+      user: ''
+      password: ''
diff --git a/backend/constants/notification.go b/backend/constants/notification.go
new file mode 100644
index 00000000..cad3b19b
--- /dev/null
+++ b/backend/constants/notification.go
@@ -0,0 +1,7 @@
+package constants
+
+const (
+	NotificationTriggerOnTaskEnd   = "notification_trigger_on_task_end"
+	NotificationTriggerOnTaskError = "notification_trigger_on_task_error"
+	NotificationTriggerNever       = "notification_trigger_never"
+)
diff --git a/backend/go.mod b/backend/go.mod
index 497c5b4d..791c6771 100644
--- a/backend/go.mod
+++ b/backend/go.mod
@@ -19,7 +19,9 @@ require (
 	github.com/satori/go.uuid v1.2.0
 	github.com/smartystreets/goconvey v0.0.0-20190731233626-505e41936337
 	github.com/spf13/viper v1.4.0
+	gopkg.in/alexcesaro/quotedprintable.v3 v3.0.0-20150716171945-2caba252f4dc // indirect
 	gopkg.in/go-playground/validator.v9 v9.29.1
+	gopkg.in/gomail.v2 v2.0.0-20150902115704-41f357289737
 	gopkg.in/russross/blackfriday.v2 v2.0.0 // indirect
 	gopkg.in/yaml.v2 v2.2.2
 )
diff --git a/backend/go.sum b/backend/go.sum
index 5a186edd..37fe6119 100644
--- a/backend/go.sum
+++ b/backend/go.sum
@@ -235,6 +235,8 @@ google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoA
 google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
 google.golang.org/grpc v1.21.0/go.mod h1:oYelfM1adQP15Ek0mdvEgi9Df8B9CZIaU1084ijfRaM=
 gopkg.in/alecthomas/kingpin.v2 v2.2.6/go.mod h1:FMv+mEhP44yOT+4EoQTLFTRgOQ1FBLkstjWtayDeSgw=
+gopkg.in/alexcesaro/quotedprintable.v3 v3.0.0-20150716171945-2caba252f4dc h1:2gGKlE2+asNV9m7xrywl36YYNnBG5ZQ0r/BOOxqPpmk=
+gopkg.in/alexcesaro/quotedprintable.v3 v3.0.0-20150716171945-2caba252f4dc/go.mod h1:m7x9LTH6d71AHyAX77c9yqWCCa3UKHcVEj9y7hAtKDk=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127 h1:qIbj1fsPNlZgppZ+VLlY7N33q108Sa+fhmuc+sWQYwY=
 gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
@@ -245,6 +247,8 @@ gopkg.in/go-playground/validator.v8 v8.18.2 h1:lFB4DoMU6B626w8ny76MV7VX6W2VHct2G
 gopkg.in/go-playground/validator.v8 v8.18.2/go.mod h1:RX2a/7Ha8BgOhfk7j780h4/u/RRjR0eouCJSH80/M2Y=
 gopkg.in/go-playground/validator.v9 v9.29.1 h1:SvGtYmN60a5CVKTOzMSyfzWDeZRxRuGvRQyEAKbw1xc=
 gopkg.in/go-playground/validator.v9 v9.29.1/go.mod h1:+c9/zcJMFNgbLvly1L1V+PpxWdVbfP1avr/N00E2vyQ=
+gopkg.in/gomail.v2 v2.0.0-20150902115704-41f357289737 h1:NvePS/smRcFQ4bMtTddFtknbGCtoBkJxGmpSpVRafCc=
+gopkg.in/gomail.v2 v2.0.0-20150902115704-41f357289737/go.mod h1:LRQQ+SO6ZHR7tOkpBDuZnXENFzX8qRjMDMyPD6BRkCw=
 gopkg.in/resty.v1 v1.12.0/go.mod h1:mDo4pnntr5jdWRML875a/NmxYqAlA73dVijT2AXvQQo=
 gopkg.in/russross/blackfriday.v2 v2.0.0 h1:+FlnIV8DSQnT7NZ43hcVKcdJdzZoeCmJj4Ql8gq5keA=
 gopkg.in/russross/blackfriday.v2 v2.0.0/go.mod h1:6sSBNz/GtOm/pJTuh5UmBK2ZHfmnxGbl2NZg1UliSOI=
diff --git a/backend/main.go b/backend/main.go
index 08cdf70f..2b052171 100644
--- a/backend/main.go
+++ b/backend/main.go
@@ -204,6 +204,7 @@ func main() {
 			authGroup.POST("/users/:id", routes.PostUser)     // 更改用户
 			authGroup.DELETE("/users/:id", routes.DeleteUser) // 删除用户
 			authGroup.GET("/me", routes.GetMe)                // 获取自己账户
+			authGroup.POST("/me", routes.PostMe)              // 修改自己账户
 			// release版本
 			authGroup.GET("/version", routes.GetVersion) // 获取发布的版本
 			// 系统
diff --git a/backend/model/schedule.go b/backend/model/schedule.go
index 3b654b74..d98dabf6 100644
--- a/backend/model/schedule.go
+++ b/backend/model/schedule.go
@@ -23,6 +23,7 @@ type Schedule struct {
 	NodeIds     []bson.ObjectId `json:"node_ids" bson:"node_ids"`
 	Status      string          `json:"status" bson:"status"`
 	Enabled     bool            `json:"enabled" bson:"enabled"`
+	UserId      bson.ObjectId   `json:"user_id" bson:"user_id"`
 
 	// 前端展示
 	SpiderName string `json:"spider_name" bson:"spider_name"`
@@ -49,27 +50,6 @@ func (sch *Schedule) Delete() error {
 	return c.RemoveId(sch.Id)
 }
 
-//func (sch *Schedule) SyncNodeIdAndSpiderId(node Node, spider Spider) {
-//	sch.syncNodeId(node)
-//	sch.syncSpiderId(spider)
-//}
-
-//func (sch *Schedule) syncNodeId(node Node) {
-//	if node.Id.Hex() == sch.NodeId.Hex() {
-//		return
-//	}
-//	sch.NodeId = node.Id
-//	_ = sch.Save()
-//}
-
-//func (sch *Schedule) syncSpiderId(spider Spider) {
-//	if spider.Id.Hex() == sch.SpiderId.Hex() {
-//		return
-//	}
-//	sch.SpiderId = spider.Id
-//	_ = sch.Save()
-//}
-
 func GetScheduleList(filter interface{}) ([]Schedule, error) {
 	s, c := database.GetCol("schedules")
 	defer s.Close()
@@ -125,13 +105,8 @@ func UpdateSchedule(id bson.ObjectId, item Schedule) error {
 	if err := c.FindId(id).One(&result); err != nil {
 		return err
 	}
-	//node, err := GetNode(item.NodeId)
-	//if err != nil {
-	//	return err
-	//}
 
 	item.UpdateTs = time.Now()
-	//item.NodeKey = node.Key
 	if err := item.Save(); err != nil {
 		return err
 	}
@@ -142,15 +117,9 @@ func AddSchedule(item Schedule) error {
 	s, c := database.GetCol("schedules")
 	defer s.Close()
 
-	//node, err := GetNode(item.NodeId)
-	//if err != nil {
-	//	return err
-	//}
-
 	item.Id = bson.NewObjectId()
 	item.CreateTs = time.Now()
 	item.UpdateTs = time.Now()
-	//item.NodeKey = node.Key
 
 	if err := c.Insert(&item); err != nil {
 		debug.PrintStack()
diff --git a/backend/model/task.go b/backend/model/task.go
index 6762bd54..a2792061 100644
--- a/backend/model/task.go
+++ b/backend/model/task.go
@@ -25,6 +25,7 @@ type Task struct {
 	RuntimeDuration float64       `json:"runtime_duration" bson:"runtime_duration"`
 	TotalDuration   float64       `json:"total_duration" bson:"total_duration"`
 	Pid             int           `json:"pid" bson:"pid"`
+	UserId          bson.ObjectId `json:"user_id" bson:"user_id"`
 
 	// 前端数据
 	SpiderName string `json:"spider_name"`
diff --git a/backend/model/user.go b/backend/model/user.go
index 19313e97..246fa448 100644
--- a/backend/model/user.go
+++ b/backend/model/user.go
@@ -16,11 +16,17 @@ type User struct {
 	Username string        `json:"username" bson:"username"`
 	Password string        `json:"password" bson:"password"`
 	Role     string        `json:"role" bson:"role"`
+	Email    string        `json:"email" bson:"email"`
+	Setting  UserSetting   `json:"setting" bson:"setting"`
 
 	CreateTs time.Time `json:"create_ts" bson:"create_ts"`
 	UpdateTs time.Time `json:"update_ts" bson:"update_ts"`
 }
 
+type UserSetting struct {
+	NotificationTrigger string `json:"notification_trigger" bson:"notification_trigger"`
+}
+
 func (user *User) Save() error {
 	s, c := database.GetCol("users")
 	defer s.Close()
diff --git a/backend/routes/schedule.go b/backend/routes/schedule.go
index c7ef474a..3776019c 100644
--- a/backend/routes/schedule.go
+++ b/backend/routes/schedule.go
@@ -76,6 +76,9 @@ func PutSchedule(c *gin.Context) {
 		return
 	}
 
+	// 加入用户ID
+	item.UserId = services.GetCurrentUser(c).Id
+
 	// 更新数据库
 	if err := model.AddSchedule(item); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
diff --git a/backend/routes/task.go b/backend/routes/task.go
index d5e3cacc..07105f2d 100644
--- a/backend/routes/task.go
+++ b/backend/routes/task.go
@@ -112,6 +112,7 @@ func PutTask(c *gin.Context) {
 				SpiderId: reqBody.SpiderId,
 				NodeId:   node.Id,
 				Param:    reqBody.Param,
+				UserId:   services.GetCurrentUser(c).Id,
 			}
 
 			if err := services.AddTask(t); err != nil {
@@ -124,6 +125,7 @@ func PutTask(c *gin.Context) {
 		t := model.Task{
 			SpiderId: reqBody.SpiderId,
 			Param:    reqBody.Param,
+			UserId:   services.GetCurrentUser(c).Id,
 		}
 		if err := services.AddTask(t); err != nil {
 			HandleError(http.StatusInternalServerError, c, err)
@@ -136,6 +138,7 @@ func PutTask(c *gin.Context) {
 				SpiderId: reqBody.SpiderId,
 				NodeId:   nodeId,
 				Param:    reqBody.Param,
+				UserId:   services.GetCurrentUser(c).Id,
 			}
 
 			if err := services.AddTask(t); err != nil {
diff --git a/backend/routes/user.go b/backend/routes/user.go
index 33b6a958..3d636f41 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -22,6 +22,7 @@ type UserRequestData struct {
 	Username string `json:"username"`
 	Password string `json:"password"`
 	Role     string `json:"role"`
+	Email    string `json:"email"`
 }
 
 func GetUser(c *gin.Context) {
@@ -99,6 +100,7 @@ func PutUser(c *gin.Context) {
 		Username: strings.ToLower(reqData.Username),
 		Password: utils.EncryptPassword(reqData.Password),
 		Role:     reqData.Role,
+		Email:    reqData.Email,
 	}
 	if err := user.Add(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
@@ -205,3 +207,39 @@ func GetMe(c *gin.Context) {
 		User: user,
 	}, nil)
 }
+
+func PostMe(c *gin.Context) {
+	type ReqBody struct {
+		Email               string `json:"email"`
+		Password            string `json:"password"`
+		NotificationTrigger string `json:"notification_trigger"`
+	}
+	ctx := context.WithGinContext(c)
+	user := ctx.User()
+	if user == nil {
+		ctx.FailedWithError(constants.ErrorUserNotFound, http.StatusUnauthorized)
+		return
+	}
+	var reqBody ReqBody
+	if err := c.ShouldBindJSON(&reqBody); err != nil {
+		HandleErrorF(http.StatusBadRequest, c, "invalid request")
+		return
+	}
+	if reqBody.Email != "" {
+		user.Email = reqBody.Email
+	}
+	if reqBody.Password != "" {
+		user.Password = utils.EncryptPassword(reqBody.Password)
+	}
+	if reqBody.NotificationTrigger != "" {
+		user.Setting.NotificationTrigger = reqBody.NotificationTrigger
+	}
+	if err := user.Save(); err != nil {
+		HandleError(http.StatusInternalServerError, c, err)
+		return
+	}
+	c.JSON(http.StatusOK, Response{
+		Status:  "ok",
+		Message: "success",
+	})
+}
diff --git a/backend/services/notification/mail.go b/backend/services/notification/mail.go
index dd1490dc..48c86351 100644
--- a/backend/services/notification/mail.go
+++ b/backend/services/notification/mail.go
@@ -1,7 +1,131 @@
 package notification
 
-import "github.com/matcornic/hermes"
+import (
+	"errors"
+	"github.com/apex/log"
+	"github.com/matcornic/hermes"
+	"gopkg.in/gomail.v2"
+	"net/mail"
+	"os"
+	"runtime/debug"
+	"strconv"
+)
 
-func SendMail() error {
-	hermes
+func SendMail(toEmail string, subject string, content string) error {
+	// hermes instance
+	h := hermes.Hermes{
+		Product: hermes.Product{
+			Name: "Hermes",
+			Link: "https://example-hermes.com/",
+			Logo: "http://www.duchess-france.org/wp-content/uploads/2016/01/gopher.png",
+		},
+	}
+
+	// config
+	port, _ := strconv.Atoi(os.Getenv("CRAWLAB_NOTIFICATION_MAIL_PORT"))
+	password := os.Getenv("CRAWLAB_NOTIFICATION_MAIL_SMTP_PASSWORD")
+	SMTPUser := os.Getenv("CRAWLAB_NOTIFICATION_MAIL_SMTP_USER")
+	smtpConfig := smtpAuthentication{
+		Server:         os.Getenv("CRAWLAB_NOTIFICATION_MAIL_SERVER"),
+		Port:           port,
+		SenderEmail:    os.Getenv("CRAWLAB_NOTIFICATION_MAIL_SENDEREMAIL"),
+		SenderIdentity: os.Getenv("CRAWLAB_NOTIFICATION_MAIL_SENDERIDENTITY"),
+		SMTPPassword:   password,
+		SMTPUser:       SMTPUser,
+	}
+	options := sendOptions{
+		To:      toEmail,
+		Subject: subject,
+	}
+
+	// email instance
+	email := hermes.Email{
+		Body: hermes.Body{
+			FreeMarkdown: hermes.Markdown(content),
+		},
+	}
+
+	// generate html
+	html, err := h.GenerateHTML(email)
+	if err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+		return err
+	}
+
+	// generate text
+	text, err := h.GeneratePlainText(email)
+	if err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+		return err
+	}
+
+	// send the email
+	if err := send(smtpConfig, options, html, text); err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+		return err
+	}
+
+	return nil
+}
+
+type smtpAuthentication struct {
+	Server         string
+	Port           int
+	SenderEmail    string
+	SenderIdentity string
+	SMTPUser       string
+	SMTPPassword   string
+}
+
+// sendOptions are options for sending an email
+type sendOptions struct {
+	To      string
+	Subject string
+}
+
+// send sends the email
+func send(smtpConfig smtpAuthentication, options sendOptions, htmlBody string, txtBody string) error {
+
+	if smtpConfig.Server == "" {
+		return errors.New("SMTP server config is empty")
+	}
+	if smtpConfig.Port == 0 {
+		return errors.New("SMTP port config is empty")
+	}
+
+	if smtpConfig.SMTPUser == "" {
+		return errors.New("SMTP user is empty")
+	}
+
+	if smtpConfig.SenderIdentity == "" {
+		return errors.New("SMTP sender identity is empty")
+	}
+
+	if smtpConfig.SenderEmail == "" {
+		return errors.New("SMTP sender email is empty")
+	}
+
+	if options.To == "" {
+		return errors.New("no receiver emails configured")
+	}
+
+	from := mail.Address{
+		Name:    smtpConfig.SenderIdentity,
+		Address: smtpConfig.SenderEmail,
+	}
+
+	m := gomail.NewMessage()
+	m.SetHeader("From", from.String())
+	m.SetHeader("To", options.To)
+	m.SetHeader("Subject", options.Subject)
+
+	m.SetBody("text/plain", txtBody)
+	m.AddAlternative("text/html", htmlBody)
+
+	d := gomail.NewPlainDialer(smtpConfig.Server, smtpConfig.Port, smtpConfig.SMTPUser, smtpConfig.SMTPPassword)
+
+	return d.DialAndSend(m)
 }
diff --git a/backend/services/schedule.go b/backend/services/schedule.go
index c321c393..b5268a88 100644
--- a/backend/services/schedule.go
+++ b/backend/services/schedule.go
@@ -34,6 +34,7 @@ func AddScheduleTask(s model.Schedule) func() {
 					SpiderId: s.SpiderId,
 					NodeId:   node.Id,
 					Param:    s.Param,
+					UserId:   s.UserId,
 				}
 
 				if err := AddTask(t); err != nil {
@@ -46,6 +47,7 @@ func AddScheduleTask(s model.Schedule) func() {
 				Id:       id.String(),
 				SpiderId: s.SpiderId,
 				Param:    s.Param,
+				UserId:   s.UserId,
 			}
 			if err := AddTask(t); err != nil {
 				log.Errorf(err.Error())
@@ -60,6 +62,7 @@ func AddScheduleTask(s model.Schedule) func() {
 					SpiderId: s.SpiderId,
 					NodeId:   nodeId,
 					Param:    s.Param,
+					UserId:   s.UserId,
 				}
 
 				if err := AddTask(t); err != nil {
diff --git a/backend/services/task.go b/backend/services/task.go
index 7a2eed2a..ea7c47c3 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -6,9 +6,11 @@ import (
 	"crawlab/entity"
 	"crawlab/lib/cron"
 	"crawlab/model"
+	"crawlab/services/notification"
 	"crawlab/utils"
 	"encoding/json"
 	"errors"
+	"fmt"
 	"github.com/apex/log"
 	"github.com/globalsign/mgo/bson"
 	uuid "github.com/satori/go.uuid"
@@ -499,6 +501,31 @@ func ExecuteTask(id int) {
 	t.RuntimeDuration = t.FinishTs.Sub(t.StartTs).Seconds() // 运行时长
 	t.TotalDuration = t.FinishTs.Sub(t.CreateTs).Seconds()  // 总时长
 
+	// 获得触发任务用户
+	user, err := model.GetUser(t.UserId)
+	if err != nil {
+		log.Errorf(GetWorkerPrefix(id) + err.Error())
+		return
+	}
+
+	// 如果是任务结束时发送通知，则发送通知
+	if user.Email != "" &&
+		user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
+		emailContent := fmt.Sprintf(`
+# Your task has finished
+
+The stats of the task are as below.
+
+Metric | Value
+--- | ---
+Create Time | %s
+Start Time | %s
+Finish Time | %s
+Total Duration | %.1f
+`, t.CreateTs, t.StartTs, t.FinishTs, t.TotalDuration)
+		_ = notification.SendMail(user.Email, fmt.Sprintf("[%s] Crawlab Task Finished", spider.Name), emailContent)
+	}
+
 	// 保存任务
 	if err := t.Save(); err != nil {
 		log.Errorf(GetWorkerPrefix(id) + err.Error())
diff --git a/backend/services/user.go b/backend/services/user.go
index 61fd952e..f20b9e7a 100644
--- a/backend/services/user.go
+++ b/backend/services/user.go
@@ -6,6 +6,7 @@ import (
 	"crawlab/utils"
 	"errors"
 	"github.com/dgrijalva/jwt-go"
+	"github.com/gin-gonic/gin"
 	"github.com/globalsign/mgo/bson"
 	"github.com/spf13/viper"
 	"time"
@@ -91,3 +92,8 @@ func CheckToken(tokenStr string) (user model.User, err error) {
 
 	return
 }
+
+func GetCurrentUser(c *gin.Context) *model.User {
+	data, _ := c.Get("currentUser")
+	return data.(*model.User)
+}
diff --git a/backend/vendor/modules.txt b/backend/vendor/modules.txt
index b8784ecb..a73e7de6 100644
--- a/backend/vendor/modules.txt
+++ b/backend/vendor/modules.txt
@@ -121,10 +121,14 @@ golang.org/x/sys/unix
 # golang.org/x/text v0.3.0
 golang.org/x/text/transform
 golang.org/x/text/unicode/norm
+# gopkg.in/alexcesaro/quotedprintable.v3 v3.0.0-20150716171945-2caba252f4dc
+gopkg.in/alexcesaro/quotedprintable.v3
 # gopkg.in/go-playground/validator.v8 v8.18.2
 gopkg.in/go-playground/validator.v8
 # gopkg.in/go-playground/validator.v9 v9.29.1
 gopkg.in/go-playground/validator.v9
+# gopkg.in/gomail.v2 v2.0.0-20150902115704-41f357289737
+gopkg.in/gomail.v2
 # gopkg.in/russross/blackfriday.v2 v2.0.0
 gopkg.in/russross/blackfriday.v2
 # gopkg.in/yaml.v2 v2.2.2
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index e739b979..841316fb 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -11,6 +11,7 @@ export default {
   'Schedules': '定时任务',
   'Deploys': '部署',
   'Sites': '网站',
+  'Setting': '设置',
 
   // 标签
   'Overview': '概览',
@@ -323,6 +324,7 @@ export default {
   'The schedule has been removed': '已删除定时任务',
   'The schedule has been added': '已添加定时任务',
   'The schedule has been saved': '已保存定时任务',
+  'Email format invalid': '邮箱地址格式不正确',
 
   // 登录
   'Sign in': '登录',
@@ -343,6 +345,16 @@ export default {
   'Role': '角色',
   'Edit User': '更改用户',
   'Users': '用户',
+  'Email': '邮箱',
+  'Optional': '可选',
+
+  // 设置
+  'Notification Trigger': '通知触发',
+  'On Task End': '当任务结束',
+  'On Task Error': '当任务发生错误',
+  'Never': '从不',
+
+  // 其他
   tagsView: {
     closeOthers: '关闭其他',
     close: '关闭',
diff --git a/frontend/src/router/index.js b/frontend/src/router/index.js
index fdb7c39f..c019836a 100644
--- a/frontend/src/router/index.js
+++ b/frontend/src/router/index.js
@@ -210,6 +210,24 @@ export const constantRouterMap = [
       }
     ]
   },
+  {
+    path: '/setting',
+    component: Layout,
+    meta: {
+      title: 'Setting',
+      icon: 'fa fa-gear'
+    },
+    children: [
+      {
+        path: '',
+        component: () => import('../views/setting/Setting'),
+        meta: {
+          title: 'Setting',
+          icon: 'fa fa-gear'
+        }
+      }
+    ]
+  },
 
   { path: '*', redirect: '/404', hidden: true }
 ]
diff --git a/frontend/src/store/modules/user.js b/frontend/src/store/modules/user.js
index 2b3a79d7..af778816 100644
--- a/frontend/src/store/modules/user.js
+++ b/frontend/src/store/modules/user.js
@@ -91,6 +91,11 @@ const user = {
         })
     },
 
+    // 修改用户信息
+    postInfo ({ commit }, form) {
+      return request.post('/me', form)
+    },
+
     // 注册
     register ({ dispatch, commit, state }, userInfo) {
       return new Promise((resolve, reject) => {
diff --git a/frontend/src/views/login/index.vue b/frontend/src/views/login/index.vue
index 664d05e5..5bad9935 100644
--- a/frontend/src/views/login/index.vue
+++ b/frontend/src/views/login/index.vue
@@ -34,6 +34,14 @@
           @keyup.enter.native="onKeyEnter"
         />
       </el-form-item>
+      <el-form-item v-if="isSignUp" prop="email" style="margin-bottom: 28px;">
+        <el-input
+          v-model="loginForm.email"
+          name="email"
+          :placeholder="$t('Email')"
+          @keyup.enter.native="onKeyEnter"
+        />
+      </el-form-item>
       <el-form-item style="border: none">
         <el-button v-if="isSignUp" :loading="loading" type="primary" style="width:100%;"
                    @click.native.prevent="handleSignup">
@@ -104,7 +112,8 @@ export default {
       loginForm: {
         username: '',
         password: '',
-        confirmPassword: ''
+        confirmPassword: '',
+        email: ''
       },
       loginRules: {
         username: [{ required: true, trigger: 'blur', validator: validateUsername }],
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
new file mode 100644
index 00000000..83695f59
--- /dev/null
+++ b/frontend/src/views/setting/Setting.vue
@@ -0,0 +1,102 @@
+<template>
+  <div class="app-container">
+    <el-form :model="userInfo" class="setting-form" ref="setting-form" label-width="150px" :rules="rules"
+             inline-message>
+      <el-form-item prop="username" :label="$t('Username')">
+        <el-input v-model="userInfo.username" disabled></el-input>
+      </el-form-item>
+      <el-form-item prop="password" :label="$t('Password')">
+        <el-input v-model="userInfo.password" type="password" :placeholder="$t('Password')"></el-input>
+      </el-form-item>
+      <el-form-item prop="email" :label="$t('Email')">
+        <el-input v-model="userInfo.email" :placeholder="$t('Email')"></el-input>
+      </el-form-item>
+      <el-form-item :label="$t('Notification Trigger')">
+        <el-radio-group v-model="userInfo.setting.notification_trigger">
+          <el-radio label="notification_trigger_on_task_end">
+            {{$t('On Task End')}}
+          </el-radio>
+          <el-radio label="notification_trigger_on_task_error">
+            {{$t('On Task Error')}}
+          </el-radio>
+          <el-radio label="notification_trigger_never">
+            {{$t('Never')}}
+          </el-radio>
+        </el-radio-group>
+      </el-form-item>
+      <el-form-item>
+        <div class="buttons">
+          <el-button type="success" @click="saveUserInfo">{{$t('Save')}}</el-button>
+        </div>
+      </el-form-item>
+    </el-form>
+  </div>
+</template>
+
+<script>
+export default {
+  name: 'Setting',
+  data () {
+    const validatePass = (rule, value, callback) => {
+      if (!value) return callback()
+      if (value.length < 5) {
+        callback(new Error(this.$t('Password length should be no shorter than 5')))
+      } else {
+        callback()
+      }
+    }
+    const validateEmail = (rule, value, callback) => {
+      if (!value) return callback()
+      if (!value.match(/.+@.+/i)) {
+        callback(new Error(this.$t('Email format invalid')))
+      } else {
+        callback()
+      }
+    }
+    return {
+      userInfo: { setting: {} },
+      rules: {
+        password: [{ trigger: 'blur', validator: validatePass }],
+        email: [{ trigger: 'blur', validator: validateEmail }]
+      }
+    }
+  },
+  methods: {
+    getUserInfo () {
+      const data = localStorage.getItem('user_info')
+      if (!data) return {}
+      this.userInfo = JSON.parse(data)
+      if (!this.userInfo.setting) this.userInfo.setting = {}
+    },
+    saveUserInfo () {
+      this.$refs['setting-form'].validate(async valid => {
+        if (!valid) return
+        const res = await this.$store.dispatch('user/postInfo', {
+          password: this.userInfo.password,
+          email: this.userInfo.email,
+          notification_trigger: this.userInfo.setting.notification_trigger
+        })
+        if (!res || res.error) {
+          this.$message.error(res.error)
+        } else {
+          this.$message.success(this.$t('Saved successfully'))
+        }
+      })
+    }
+  },
+  async created () {
+    await this.$store.dispatch('user/getInfo')
+    this.getUserInfo()
+  }
+}
+</script>
+
+<style scoped>
+  .setting-form {
+    width: 600px;
+  }
+
+  .setting-form .buttons {
+    text-align: right;
+  }
+</style>
diff --git a/frontend/src/views/user/UserList.vue b/frontend/src/views/user/UserList.vue
index 26cbedea..9389f08c 100644
--- a/frontend/src/views/user/UserList.vue
+++ b/frontend/src/views/user/UserList.vue
@@ -15,6 +15,9 @@
             <el-option value="normal" :label="$t('normal')"></el-option>
           </el-select>
         </el-form-item>
+        <el-form-item prop="email" :label="$t('Email')">
+          <el-input v-model="userForm.email" :placeholder="$t('Email')"/>
+        </el-form-item>
       </el-form>
       <template slot="footer">
         <el-button size="small" @click="dialogVisible=false">{{$t('Cancel')}}</el-button>
@@ -107,11 +110,20 @@ export default {
         callback()
       }
     }
+    const validateEmail = (rule, value, callback) => {
+      if (!value) return callback()
+      if (!value.match(/.+@.+/i)) {
+        callback(new Error(this.$t('Email format invalid')))
+      } else {
+        callback()
+      }
+    }
     return {
       dialogVisible: false,
       isAdd: false,
       rules: {
-        password: [{ validator: validatePass }]
+        password: [{ validator: validatePass }],
+        email: [{ validator: validateEmail }]
       }
     }
   },
@@ -205,6 +217,8 @@ export default {
       this.isAdd = true
       this.$store.commit('user/SET_USER_FORM', {})
       this.dialogVisible = true
+    },
+    onValidateEmail (value) {
     }
   },
   created () {

From b2f3578aea43e27309c5d23a0899dc3c57de6ed0 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Mon, 13 Jan 2020 19:56:34 +0800
Subject: [PATCH 06/42] added vendor

---
 .../alexcesaro/quotedprintable.v3/LICENSE     |  20 ++
 .../alexcesaro/quotedprintable.v3/README.md   |  16 +
 .../quotedprintable.v3/encodedword.go         | 279 ++++++++++++++++
 .../alexcesaro/quotedprintable.v3/pool.go     |  26 ++
 .../quotedprintable.v3/pool_go12.go           |  24 ++
 .../alexcesaro/quotedprintable.v3/reader.go   | 121 +++++++
 .../alexcesaro/quotedprintable.v3/writer.go   | 166 ++++++++++
 backend/vendor/gopkg.in/gomail.v2/.travis.yml |   8 +
 .../vendor/gopkg.in/gomail.v2/CHANGELOG.md    |  20 ++
 .../vendor/gopkg.in/gomail.v2/CONTRIBUTING.md |  20 ++
 backend/vendor/gopkg.in/gomail.v2/LICENSE     |  20 ++
 backend/vendor/gopkg.in/gomail.v2/README.md   |  97 ++++++
 backend/vendor/gopkg.in/gomail.v2/auth.go     |  67 ++++
 backend/vendor/gopkg.in/gomail.v2/doc.go      |   5 +
 backend/vendor/gopkg.in/gomail.v2/message.go  | 302 ++++++++++++++++++
 backend/vendor/gopkg.in/gomail.v2/mime.go     |  19 ++
 .../vendor/gopkg.in/gomail.v2/mime_go14.go    |  16 +
 backend/vendor/gopkg.in/gomail.v2/send.go     | 117 +++++++
 backend/vendor/gopkg.in/gomail.v2/smtp.go     | 175 ++++++++++
 backend/vendor/gopkg.in/gomail.v2/writeto.go  | 242 ++++++++++++++
 20 files changed, 1760 insertions(+)
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/LICENSE
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/README.md
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/encodedword.go
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool.go
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool_go12.go
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/reader.go
 create mode 100644 backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/writer.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/.travis.yml
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/CHANGELOG.md
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/CONTRIBUTING.md
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/LICENSE
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/README.md
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/auth.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/doc.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/message.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/mime.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/mime_go14.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/send.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/smtp.go
 create mode 100644 backend/vendor/gopkg.in/gomail.v2/writeto.go

diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/LICENSE b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/LICENSE
new file mode 100644
index 00000000..5f5c12af
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/LICENSE
@@ -0,0 +1,20 @@
+The MIT License (MIT)
+
+Copyright (c) 2014 Alexandre Cesaro
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/README.md b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/README.md
new file mode 100644
index 00000000..98ddf829
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/README.md
@@ -0,0 +1,16 @@
+# quotedprintable
+
+## Introduction
+
+Package quotedprintable implements quoted-printable and message header encoding
+as specified by RFC 2045 and RFC 2047.
+
+It is a copy of the Go 1.5 package `mime/quotedprintable`. It also includes
+the new functions of package `mime` concerning RFC 2047.
+
+This code has minor changes with the standard library code in order to work
+with Go 1.0 and newer. 
+
+## Documentation
+
+https://godoc.org/gopkg.in/alexcesaro/quotedprintable.v3
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/encodedword.go b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/encodedword.go
new file mode 100644
index 00000000..cfd02617
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/encodedword.go
@@ -0,0 +1,279 @@
+package quotedprintable
+
+import (
+	"bytes"
+	"encoding/base64"
+	"errors"
+	"fmt"
+	"io"
+	"strings"
+	"unicode"
+	"unicode/utf8"
+)
+
+// A WordEncoder is a RFC 2047 encoded-word encoder.
+type WordEncoder byte
+
+const (
+	// BEncoding represents Base64 encoding scheme as defined by RFC 2045.
+	BEncoding = WordEncoder('b')
+	// QEncoding represents the Q-encoding scheme as defined by RFC 2047.
+	QEncoding = WordEncoder('q')
+)
+
+var (
+	errInvalidWord = errors.New("mime: invalid RFC 2047 encoded-word")
+)
+
+// Encode returns the encoded-word form of s. If s is ASCII without special
+// characters, it is returned unchanged. The provided charset is the IANA
+// charset name of s. It is case insensitive.
+func (e WordEncoder) Encode(charset, s string) string {
+	if !needsEncoding(s) {
+		return s
+	}
+	return e.encodeWord(charset, s)
+}
+
+func needsEncoding(s string) bool {
+	for _, b := range s {
+		if (b < ' ' || b > '~') && b != '\t' {
+			return true
+		}
+	}
+	return false
+}
+
+// encodeWord encodes a string into an encoded-word.
+func (e WordEncoder) encodeWord(charset, s string) string {
+	buf := getBuffer()
+	defer putBuffer(buf)
+
+	buf.WriteString("=?")
+	buf.WriteString(charset)
+	buf.WriteByte('?')
+	buf.WriteByte(byte(e))
+	buf.WriteByte('?')
+
+	if e == BEncoding {
+		w := base64.NewEncoder(base64.StdEncoding, buf)
+		io.WriteString(w, s)
+		w.Close()
+	} else {
+		enc := make([]byte, 3)
+		for i := 0; i < len(s); i++ {
+			b := s[i]
+			switch {
+			case b == ' ':
+				buf.WriteByte('_')
+			case b <= '~' && b >= '!' && b != '=' && b != '?' && b != '_':
+				buf.WriteByte(b)
+			default:
+				enc[0] = '='
+				enc[1] = upperhex[b>>4]
+				enc[2] = upperhex[b&0x0f]
+				buf.Write(enc)
+			}
+		}
+	}
+	buf.WriteString("?=")
+	return buf.String()
+}
+
+const upperhex = "0123456789ABCDEF"
+
+// A WordDecoder decodes MIME headers containing RFC 2047 encoded-words.
+type WordDecoder struct {
+	// CharsetReader, if non-nil, defines a function to generate
+	// charset-conversion readers, converting from the provided
+	// charset into UTF-8.
+	// Charsets are always lower-case. utf-8, iso-8859-1 and us-ascii charsets
+	// are handled by default.
+	// One of the the CharsetReader's result values must be non-nil.
+	CharsetReader func(charset string, input io.Reader) (io.Reader, error)
+}
+
+// Decode decodes an encoded-word. If word is not a valid RFC 2047 encoded-word,
+// word is returned unchanged.
+func (d *WordDecoder) Decode(word string) (string, error) {
+	fields := strings.Split(word, "?") // TODO: remove allocation?
+	if len(fields) != 5 || fields[0] != "=" || fields[4] != "=" || len(fields[2]) != 1 {
+		return "", errInvalidWord
+	}
+
+	content, err := decode(fields[2][0], fields[3])
+	if err != nil {
+		return "", err
+	}
+
+	buf := getBuffer()
+	defer putBuffer(buf)
+
+	if err := d.convert(buf, fields[1], content); err != nil {
+		return "", err
+	}
+
+	return buf.String(), nil
+}
+
+// DecodeHeader decodes all encoded-words of the given string. It returns an
+// error if and only if CharsetReader of d returns an error.
+func (d *WordDecoder) DecodeHeader(header string) (string, error) {
+	// If there is no encoded-word, returns before creating a buffer.
+	i := strings.Index(header, "=?")
+	if i == -1 {
+		return header, nil
+	}
+
+	buf := getBuffer()
+	defer putBuffer(buf)
+
+	buf.WriteString(header[:i])
+	header = header[i:]
+
+	betweenWords := false
+	for {
+		start := strings.Index(header, "=?")
+		if start == -1 {
+			break
+		}
+		cur := start + len("=?")
+
+		i := strings.Index(header[cur:], "?")
+		if i == -1 {
+			break
+		}
+		charset := header[cur : cur+i]
+		cur += i + len("?")
+
+		if len(header) < cur+len("Q??=") {
+			break
+		}
+		encoding := header[cur]
+		cur++
+
+		if header[cur] != '?' {
+			break
+		}
+		cur++
+
+		j := strings.Index(header[cur:], "?=")
+		if j == -1 {
+			break
+		}
+		text := header[cur : cur+j]
+		end := cur + j + len("?=")
+
+		content, err := decode(encoding, text)
+		if err != nil {
+			betweenWords = false
+			buf.WriteString(header[:start+2])
+			header = header[start+2:]
+			continue
+		}
+
+		// Write characters before the encoded-word. White-space and newline
+		// characters separating two encoded-words must be deleted.
+		if start > 0 && (!betweenWords || hasNonWhitespace(header[:start])) {
+			buf.WriteString(header[:start])
+		}
+
+		if err := d.convert(buf, charset, content); err != nil {
+			return "", err
+		}
+
+		header = header[end:]
+		betweenWords = true
+	}
+
+	if len(header) > 0 {
+		buf.WriteString(header)
+	}
+
+	return buf.String(), nil
+}
+
+func decode(encoding byte, text string) ([]byte, error) {
+	switch encoding {
+	case 'B', 'b':
+		return base64.StdEncoding.DecodeString(text)
+	case 'Q', 'q':
+		return qDecode(text)
+	}
+	return nil, errInvalidWord
+}
+
+func (d *WordDecoder) convert(buf *bytes.Buffer, charset string, content []byte) error {
+	switch {
+	case strings.EqualFold("utf-8", charset):
+		buf.Write(content)
+	case strings.EqualFold("iso-8859-1", charset):
+		for _, c := range content {
+			buf.WriteRune(rune(c))
+		}
+	case strings.EqualFold("us-ascii", charset):
+		for _, c := range content {
+			if c >= utf8.RuneSelf {
+				buf.WriteRune(unicode.ReplacementChar)
+			} else {
+				buf.WriteByte(c)
+			}
+		}
+	default:
+		if d.CharsetReader == nil {
+			return fmt.Errorf("mime: unhandled charset %q", charset)
+		}
+		r, err := d.CharsetReader(strings.ToLower(charset), bytes.NewReader(content))
+		if err != nil {
+			return err
+		}
+		if _, err = buf.ReadFrom(r); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+// hasNonWhitespace reports whether s (assumed to be ASCII) contains at least
+// one byte of non-whitespace.
+func hasNonWhitespace(s string) bool {
+	for _, b := range s {
+		switch b {
+		// Encoded-words can only be separated by linear white spaces which does
+		// not include vertical tabs (\v).
+		case ' ', '\t', '\n', '\r':
+		default:
+			return true
+		}
+	}
+	return false
+}
+
+// qDecode decodes a Q encoded string.
+func qDecode(s string) ([]byte, error) {
+	dec := make([]byte, len(s))
+	n := 0
+	for i := 0; i < len(s); i++ {
+		switch c := s[i]; {
+		case c == '_':
+			dec[n] = ' '
+		case c == '=':
+			if i+2 >= len(s) {
+				return nil, errInvalidWord
+			}
+			b, err := readHexByte(s[i+1], s[i+2])
+			if err != nil {
+				return nil, err
+			}
+			dec[n] = b
+			i += 2
+		case (c <= '~' && c >= ' ') || c == '\n' || c == '\r' || c == '\t':
+			dec[n] = c
+		default:
+			return nil, errInvalidWord
+		}
+		n++
+	}
+
+	return dec[:n], nil
+}
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool.go b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool.go
new file mode 100644
index 00000000..24283c52
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool.go
@@ -0,0 +1,26 @@
+// +build go1.3
+
+package quotedprintable
+
+import (
+	"bytes"
+	"sync"
+)
+
+var bufPool = sync.Pool{
+	New: func() interface{} {
+		return new(bytes.Buffer)
+	},
+}
+
+func getBuffer() *bytes.Buffer {
+	return bufPool.Get().(*bytes.Buffer)
+}
+
+func putBuffer(buf *bytes.Buffer) {
+	if buf.Len() > 1024 {
+		return
+	}
+	buf.Reset()
+	bufPool.Put(buf)
+}
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool_go12.go b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool_go12.go
new file mode 100644
index 00000000..d335b4ab
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/pool_go12.go
@@ -0,0 +1,24 @@
+// +build !go1.3
+
+package quotedprintable
+
+import "bytes"
+
+var ch = make(chan *bytes.Buffer, 32)
+
+func getBuffer() *bytes.Buffer {
+	select {
+	case buf := <-ch:
+		return buf
+	default:
+	}
+	return new(bytes.Buffer)
+}
+
+func putBuffer(buf *bytes.Buffer) {
+	buf.Reset()
+	select {
+	case ch <- buf:
+	default:
+	}
+}
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/reader.go b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/reader.go
new file mode 100644
index 00000000..955edca2
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/reader.go
@@ -0,0 +1,121 @@
+// Package quotedprintable implements quoted-printable encoding as specified by
+// RFC 2045.
+package quotedprintable
+
+import (
+	"bufio"
+	"bytes"
+	"fmt"
+	"io"
+)
+
+// Reader is a quoted-printable decoder.
+type Reader struct {
+	br   *bufio.Reader
+	rerr error  // last read error
+	line []byte // to be consumed before more of br
+}
+
+// NewReader returns a quoted-printable reader, decoding from r.
+func NewReader(r io.Reader) *Reader {
+	return &Reader{
+		br: bufio.NewReader(r),
+	}
+}
+
+func fromHex(b byte) (byte, error) {
+	switch {
+	case b >= '0' && b <= '9':
+		return b - '0', nil
+	case b >= 'A' && b <= 'F':
+		return b - 'A' + 10, nil
+	// Accept badly encoded bytes.
+	case b >= 'a' && b <= 'f':
+		return b - 'a' + 10, nil
+	}
+	return 0, fmt.Errorf("quotedprintable: invalid hex byte 0x%02x", b)
+}
+
+func readHexByte(a, b byte) (byte, error) {
+	var hb, lb byte
+	var err error
+	if hb, err = fromHex(a); err != nil {
+		return 0, err
+	}
+	if lb, err = fromHex(b); err != nil {
+		return 0, err
+	}
+	return hb<<4 | lb, nil
+}
+
+func isQPDiscardWhitespace(r rune) bool {
+	switch r {
+	case '\n', '\r', ' ', '\t':
+		return true
+	}
+	return false
+}
+
+var (
+	crlf       = []byte("\r\n")
+	lf         = []byte("\n")
+	softSuffix = []byte("=")
+)
+
+// Read reads and decodes quoted-printable data from the underlying reader.
+func (r *Reader) Read(p []byte) (n int, err error) {
+	// Deviations from RFC 2045:
+	// 1. in addition to "=\r\n", "=\n" is also treated as soft line break.
+	// 2. it will pass through a '\r' or '\n' not preceded by '=', consistent
+	//    with other broken QP encoders & decoders.
+	for len(p) > 0 {
+		if len(r.line) == 0 {
+			if r.rerr != nil {
+				return n, r.rerr
+			}
+			r.line, r.rerr = r.br.ReadSlice('\n')
+
+			// Does the line end in CRLF instead of just LF?
+			hasLF := bytes.HasSuffix(r.line, lf)
+			hasCR := bytes.HasSuffix(r.line, crlf)
+			wholeLine := r.line
+			r.line = bytes.TrimRightFunc(wholeLine, isQPDiscardWhitespace)
+			if bytes.HasSuffix(r.line, softSuffix) {
+				rightStripped := wholeLine[len(r.line):]
+				r.line = r.line[:len(r.line)-1]
+				if !bytes.HasPrefix(rightStripped, lf) && !bytes.HasPrefix(rightStripped, crlf) {
+					r.rerr = fmt.Errorf("quotedprintable: invalid bytes after =: %q", rightStripped)
+				}
+			} else if hasLF {
+				if hasCR {
+					r.line = append(r.line, '\r', '\n')
+				} else {
+					r.line = append(r.line, '\n')
+				}
+			}
+			continue
+		}
+		b := r.line[0]
+
+		switch {
+		case b == '=':
+			if len(r.line[1:]) < 2 {
+				return n, io.ErrUnexpectedEOF
+			}
+			b, err = readHexByte(r.line[1], r.line[2])
+			if err != nil {
+				return n, err
+			}
+			r.line = r.line[2:] // 2 of the 3; other 1 is done below
+		case b == '\t' || b == '\r' || b == '\n':
+			break
+		case b < ' ' || b > '~':
+			return n, fmt.Errorf("quotedprintable: invalid unescaped byte 0x%02x in body", b)
+		}
+		p[0] = b
+		p = p[1:]
+		r.line = r.line[1:]
+		n++
+	}
+	return n, nil
+}
diff --git a/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/writer.go b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/writer.go
new file mode 100644
index 00000000..43359d51
--- /dev/null
+++ b/backend/vendor/gopkg.in/alexcesaro/quotedprintable.v3/writer.go
@@ -0,0 +1,166 @@
+package quotedprintable
+
+import "io"
+
+const lineMaxLen = 76
+
+// A Writer is a quoted-printable writer that implements io.WriteCloser.
+type Writer struct {
+	// Binary mode treats the writer's input as pure binary and processes end of
+	// line bytes as binary data.
+	Binary bool
+
+	w    io.Writer
+	i    int
+	line [78]byte
+	cr   bool
+}
+
+// NewWriter returns a new Writer that writes to w.
+func NewWriter(w io.Writer) *Writer {
+	return &Writer{w: w}
+}
+
+// Write encodes p using quoted-printable encoding and writes it to the
+// underlying io.Writer. It limits line length to 76 characters. The encoded
+// bytes are not necessarily flushed until the Writer is closed.
+func (w *Writer) Write(p []byte) (n int, err error) {
+	for i, b := range p {
+		switch {
+		// Simple writes are done in batch.
+		case b >= '!' && b <= '~' && b != '=':
+			continue
+		case isWhitespace(b) || !w.Binary && (b == '\n' || b == '\r'):
+			continue
+		}
+
+		if i > n {
+			if err := w.write(p[n:i]); err != nil {
+				return n, err
+			}
+			n = i
+		}
+
+		if err := w.encode(b); err != nil {
+			return n, err
+		}
+		n++
+	}
+
+	if n == len(p) {
+		return n, nil
+	}
+
+	if err := w.write(p[n:]); err != nil {
+		return n, err
+	}
+
+	return len(p), nil
+}
+
+// Close closes the Writer, flushing any unwritten data to the underlying
+// io.Writer, but does not close the underlying io.Writer.
+func (w *Writer) Close() error {
+	if err := w.checkLastByte(); err != nil {
+		return err
+	}
+
+	return w.flush()
+}
+
+// write limits text encoded in quoted-printable to 76 characters per line.
+func (w *Writer) write(p []byte) error {
+	for _, b := range p {
+		if b == '\n' || b == '\r' {
+			// If the previous byte was \r, the CRLF has already been inserted.
+			if w.cr && b == '\n' {
+				w.cr = false
+				continue
+			}
+
+			if b == '\r' {
+				w.cr = true
+			}
+
+			if err := w.checkLastByte(); err != nil {
+				return err
+			}
+			if err := w.insertCRLF(); err != nil {
+				return err
+			}
+			continue
+		}
+
+		if w.i == lineMaxLen-1 {
+			if err := w.insertSoftLineBreak(); err != nil {
+				return err
+			}
+		}
+
+		w.line[w.i] = b
+		w.i++
+		w.cr = false
+	}
+
+	return nil
+}
+
+func (w *Writer) encode(b byte) error {
+	if lineMaxLen-1-w.i < 3 {
+		if err := w.insertSoftLineBreak(); err != nil {
+			return err
+		}
+	}
+
+	w.line[w.i] = '='
+	w.line[w.i+1] = upperhex[b>>4]
+	w.line[w.i+2] = upperhex[b&0x0f]
+	w.i += 3
+
+	return nil
+}
+
+// checkLastByte encodes the last buffered byte if it is a space or a tab.
+func (w *Writer) checkLastByte() error {
+	if w.i == 0 {
+		return nil
+	}
+
+	b := w.line[w.i-1]
+	if isWhitespace(b) {
+		w.i--
+		if err := w.encode(b); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+
+func (w *Writer) insertSoftLineBreak() error {
+	w.line[w.i] = '='
+	w.i++
+
+	return w.insertCRLF()
+}
+
+func (w *Writer) insertCRLF() error {
+	w.line[w.i] = '\r'
+	w.line[w.i+1] = '\n'
+	w.i += 2
+
+	return w.flush()
+}
+
+func (w *Writer) flush() error {
+	if _, err := w.w.Write(w.line[:w.i]); err != nil {
+		return err
+	}
+
+	w.i = 0
+	return nil
+}
+
+func isWhitespace(b byte) bool {
+	return b == ' ' || b == '\t'
+}
diff --git a/backend/vendor/gopkg.in/gomail.v2/.travis.yml b/backend/vendor/gopkg.in/gomail.v2/.travis.yml
new file mode 100644
index 00000000..24edf22c
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/.travis.yml
@@ -0,0 +1,8 @@
+language: go
+
+go:
+  - 1.2
+  - 1.3
+  - 1.4
+  - 1.5
+  - tip
diff --git a/backend/vendor/gopkg.in/gomail.v2/CHANGELOG.md b/backend/vendor/gopkg.in/gomail.v2/CHANGELOG.md
new file mode 100644
index 00000000..a797ab4c
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/CHANGELOG.md
@@ -0,0 +1,20 @@
+# Change Log
+All notable changes to this project will be documented in this file.
+This project adheres to [Semantic Versioning](http://semver.org/).
+
+## [2.0.0] - 2015-09-02
+
+- Mailer has been removed. It has been replaced by Dialer and Sender.
+- `File` type and the `CreateFile` and `OpenFile` functions have been removed.
+- `Message.Attach` and `Message.Embed` have a new signature.
+- `Message.GetBodyWriter` has been removed. Use `Message.AddAlternativeWriter`
+instead.
+- `Message.Export` has been removed. `Message.WriteTo` can be used instead.
+- `Message.DelHeader` has been removed.
+- The `Bcc` header field is no longer sent. It is far more simpler and
+efficient: the same message is sent to all recipients instead of sending a
+different email to each Bcc address.
+- LoginAuth has been removed. `NewPlainDialer` now implements the LOGIN
+authentication mechanism when needed.
+- Go 1.2 is now required instead of Go 1.3. No external dependency are used when
+using Go 1.5.
diff --git a/backend/vendor/gopkg.in/gomail.v2/CONTRIBUTING.md b/backend/vendor/gopkg.in/gomail.v2/CONTRIBUTING.md
new file mode 100644
index 00000000..d5601c25
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/CONTRIBUTING.md
@@ -0,0 +1,20 @@
+Thank you for contributing to Gomail! Here are a few guidelines:
+
+## Bugs
+
+If you think you found a bug, create an issue and supply the minimum amount
+of code triggering the bug so it can be reproduced.
+
+
+## Fixing a bug
+
+If you want to fix a bug, you can send a pull request. It should contains a
+new test or update an existing one to cover that bug.
+
+
+## New feature proposal
+
+If you think Gomail lacks a feature, you can open an issue or send a pull
+request. I want to keep Gomail code and API as simple as possible so please
+describe your needs so we can discuss whether this feature should be added to
+Gomail or not.
diff --git a/backend/vendor/gopkg.in/gomail.v2/LICENSE b/backend/vendor/gopkg.in/gomail.v2/LICENSE
new file mode 100644
index 00000000..5f5c12af
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/LICENSE
@@ -0,0 +1,20 @@
+The MIT License (MIT)
+
+Copyright (c) 2014 Alexandre Cesaro
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
+the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
+IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/backend/vendor/gopkg.in/gomail.v2/README.md b/backend/vendor/gopkg.in/gomail.v2/README.md
new file mode 100644
index 00000000..18cb8813
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/README.md
@@ -0,0 +1,97 @@
+# Gomail
+[![Build Status](https://travis-ci.org/go-gomail/gomail.svg?branch=v2)](https://travis-ci.org/go-gomail/gomail) [![Code Coverage](http://gocover.io/_badge/gopkg.in/gomail.v2)](http://gocover.io/gopkg.in/gomail.v2) [![Documentation](https://godoc.org/gopkg.in/gomail.v2?status.svg)](https://godoc.org/gopkg.in/gomail.v2)
+
+## Introduction
+
+Gomail is a simple and efficient package to send emails. It is well tested and
+documented.
+
+It is versioned using [gopkg.in](https://gopkg.in) so I promise
+they will never be backward incompatible changes within each version.
+
+It requires Go 1.2 or newer. With Go 1.5, no external dependencies are used.
+
+
+## Features
+
+Gomail supports:
+- Attachments
+- Embedded images
+- HTML and text templates
+- Automatic encoding of special characters
+- SSL and TLS
+- Sending multiple emails with the same SMTP connection
+- Any method to send emails: SMTP, postfix (not included but easily doable), etc
+
+
+## Documentation
+
+https://godoc.org/gopkg.in/gomail.v2
+
+
+## Download
+
+    go get gopkg.in/gomail.v2
+
+
+## Examples
+
+See the [examples in the documentation](https://godoc.org/gopkg.in/gomail.v2#example-package).
+
+
+## FAQ
+
+### x509: certificate signed by unknown authority
+
+If you get this error it means the certificate used by the SMTP server is not
+considered valid by the client running Gomail. As a quick workaround you can
+bypass the verification of the server's certificate chain and host name by using
+`SetTLSConfig`:
+
+    d := gomail.NewPlainDialer("smtp.example.com", "user", "123456", 587)
+    d.TLSConfig = &tls.Config{InsecureSkipVerify: true}
+
+Note, however, that this is insecure and should not be used in production.
+
+
+## Contribute
+
+Contributions are more than welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for
+more info.
+
+
+## Change log
+
+See [CHANGELOG.md](CHANGELOG.md).
+
+
+## License
+
+[MIT](LICENSE)
+
+
+## Contact
+
+You can ask questions on the [Gomail
+thread](https://groups.google.com/d/topic/golang-nuts/jMxZHzvvEVg/discussion)
+in the Go mailing-list.
+
+
+## Support
+
+If you want to support the development of Gomail, I gladly accept donations.
+
+I will give 100% of the money I receive to
+[Enfants, Espoir Du Monde](http://www.eedm.fr/).
+EEDM is a French NGO which helps children in Bangladesh, Cameroun, Haiti, India
+and Madagascar.
+
+All its members are volunteers so its operating costs are only
+1.9%. So your money will directly helps children of these countries.
+
+As an added bonus, your donations will also tip me by lowering my taxes :smile:
+
+I will send an email with the receipt of the donation to EEDM annually to all
+donors.
+
+[![Donate](https://www.paypalobjects.com/en_US/i/btn/btn_donate_LG.gif)](https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=PYQKC7VFVXCFG)
diff --git a/backend/vendor/gopkg.in/gomail.v2/auth.go b/backend/vendor/gopkg.in/gomail.v2/auth.go
new file mode 100644
index 00000000..4bcdd062
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/auth.go
@@ -0,0 +1,67 @@
+package gomail
+
+import (
+	"bytes"
+	"errors"
+	"fmt"
+	"net/smtp"
+)
+
+// plainAuth is an smtp.Auth that implements the PLAIN authentication mechanism.
+// It fallbacks to the LOGIN mechanism if it is the only mechanism advertised
+// by the server.
+type plainAuth struct {
+	username string
+	password string
+	host     string
+	login    bool
+}
+
+func (a *plainAuth) Start(server *smtp.ServerInfo) (string, []byte, error) {
+	if server.Name != a.host {
+		return "", nil, errors.New("gomail: wrong host name")
+	}
+
+	var plain, login bool
+	for _, a := range server.Auth {
+		switch a {
+		case "PLAIN":
+			plain = true
+		case "LOGIN":
+			login = true
+		}
+	}
+
+	if !server.TLS && !plain && !login {
+		return "", nil, errors.New("gomail: unencrypted connection")
+	}
+
+	if !plain && login {
+		a.login = true
+		return "LOGIN", nil, nil
+	}
+
+	return "PLAIN", []byte("\x00" + a.username + "\x00" + a.password), nil
+}
+
+func (a *plainAuth) Next(fromServer []byte, more bool) ([]byte, error) {
+	if !a.login {
+		if more {
+			return nil, errors.New("gomail: unexpected server challenge")
+		}
+		return nil, nil
+	}
+
+	if !more {
+		return nil, nil
+	}
+
+	switch {
+	case bytes.Equal(fromServer, []byte("Username:")):
+		return []byte(a.username), nil
+	case bytes.Equal(fromServer, []byte("Password:")):
+		return []byte(a.password), nil
+	default:
+		return nil, fmt.Errorf("gomail: unexpected server challenge: %s", fromServer)
+	}
+}
diff --git a/backend/vendor/gopkg.in/gomail.v2/doc.go b/backend/vendor/gopkg.in/gomail.v2/doc.go
new file mode 100644
index 00000000..a8f5091f
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/doc.go
@@ -0,0 +1,5 @@
+// Package gomail provides a simple interface to compose emails and to mail them
+// efficiently.
+//
+// More info on Github: https://github.com/go-gomail/gomail
+package gomail
diff --git a/backend/vendor/gopkg.in/gomail.v2/message.go b/backend/vendor/gopkg.in/gomail.v2/message.go
new file mode 100644
index 00000000..2f75368b
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/message.go
@@ -0,0 +1,302 @@
+package gomail
+
+import (
+	"bytes"
+	"io"
+	"os"
+	"path/filepath"
+	"time"
+)
+
+// Message represents an email.
+type Message struct {
+	header      header
+	parts       []part
+	attachments []*file
+	embedded    []*file
+	charset     string
+	encoding    Encoding
+	hEncoder    mimeEncoder
+	buf         bytes.Buffer
+}
+
+type header map[string][]string
+
+type part struct {
+	header header
+	copier func(io.Writer) error
+}
+
+// NewMessage creates a new message. It uses UTF-8 and quoted-printable encoding
+// by default.
+func NewMessage(settings ...MessageSetting) *Message {
+	m := &Message{
+		header:   make(header),
+		charset:  "UTF-8",
+		encoding: QuotedPrintable,
+	}
+
+	m.applySettings(settings)
+
+	if m.encoding == Base64 {
+		m.hEncoder = bEncoding
+	} else {
+		m.hEncoder = qEncoding
+	}
+
+	return m
+}
+
+// Reset resets the message so it can be reused. The message keeps its previous
+// settings so it is in the same state that after a call to NewMessage.
+func (m *Message) Reset() {
+	for k := range m.header {
+		delete(m.header, k)
+	}
+	m.parts = nil
+	m.attachments = nil
+	m.embedded = nil
+}
+
+func (m *Message) applySettings(settings []MessageSetting) {
+	for _, s := range settings {
+		s(m)
+	}
+}
+
+// A MessageSetting can be used as an argument in NewMessage to configure an
+// email.
+type MessageSetting func(m *Message)
+
+// SetCharset is a message setting to set the charset of the email.
+func SetCharset(charset string) MessageSetting {
+	return func(m *Message) {
+		m.charset = charset
+	}
+}
+
+// SetEncoding is a message setting to set the encoding of the email.
+func SetEncoding(enc Encoding) MessageSetting {
+	return func(m *Message) {
+		m.encoding = enc
+	}
+}
+
+// Encoding represents a MIME encoding scheme like quoted-printable or base64.
+type Encoding string
+
+const (
+	// QuotedPrintable represents the quoted-printable encoding as defined in
+	// RFC 2045.
+	QuotedPrintable Encoding = "quoted-printable"
+	// Base64 represents the base64 encoding as defined in RFC 2045.
+	Base64 Encoding = "base64"
+	// Unencoded can be used to avoid encoding the body of an email. The headers
+	// will still be encoded using quoted-printable encoding.
+	Unencoded Encoding = "8bit"
+)
+
+// SetHeader sets a value to the given header field.
+func (m *Message) SetHeader(field string, value ...string) {
+	m.encodeHeader(value)
+	m.header[field] = value
+}
+
+func (m *Message) encodeHeader(values []string) {
+	for i := range values {
+		values[i] = m.encodeString(values[i])
+	}
+}
+
+func (m *Message) encodeString(value string) string {
+	return m.hEncoder.Encode(m.charset, value)
+}
+
+// SetHeaders sets the message headers.
+func (m *Message) SetHeaders(h map[string][]string) {
+	for k, v := range h {
+		m.SetHeader(k, v...)
+	}
+}
+
+// SetAddressHeader sets an address to the given header field.
+func (m *Message) SetAddressHeader(field, address, name string) {
+	m.header[field] = []string{m.FormatAddress(address, name)}
+}
+
+// FormatAddress formats an address and a name as a valid RFC 5322 address.
+func (m *Message) FormatAddress(address, name string) string {
+	enc := m.encodeString(name)
+	if enc == name {
+		m.buf.WriteByte('"')
+		for i := 0; i < len(name); i++ {
+			b := name[i]
+			if b == '\\' || b == '"' {
+				m.buf.WriteByte('\\')
+			}
+			m.buf.WriteByte(b)
+		}
+		m.buf.WriteByte('"')
+	} else if hasSpecials(name) {
+		m.buf.WriteString(bEncoding.Encode(m.charset, name))
+	} else {
+		m.buf.WriteString(enc)
+	}
+	m.buf.WriteString(" <")
+	m.buf.WriteString(address)
+	m.buf.WriteByte('>')
+
+	addr := m.buf.String()
+	m.buf.Reset()
+	return addr
+}
+
+func hasSpecials(text string) bool {
+	for i := 0; i < len(text); i++ {
+		switch c := text[i]; c {
+		case '(', ')', '<', '>', '[', ']', ':', ';', '@', '\\', ',', '.', '"':
+			return true
+		}
+	}
+
+	return false
+}
+
+// SetDateHeader sets a date to the given header field.
+func (m *Message) SetDateHeader(field string, date time.Time) {
+	m.header[field] = []string{m.FormatDate(date)}
+}
+
+// FormatDate formats a date as a valid RFC 5322 date.
+func (m *Message) FormatDate(date time.Time) string {
+	return date.Format(time.RFC1123Z)
+}
+
+// GetHeader gets a header field.
+func (m *Message) GetHeader(field string) []string {
+	return m.header[field]
+}
+
+// SetBody sets the body of the message.
+func (m *Message) SetBody(contentType, body string) {
+	m.parts = []part{
+		{
+			header: m.getPartHeader(contentType),
+			copier: func(w io.Writer) error {
+				_, err := io.WriteString(w, body)
+				return err
+			},
+		},
+	}
+}
+
+// AddAlternative adds an alternative part to the message.
+//
+// It is commonly used to send HTML emails that default to the plain text
+// version for backward compatibility.
+//
+// More info: http://en.wikipedia.org/wiki/MIME#Alternative
+func (m *Message) AddAlternative(contentType, body string) {
+	m.parts = append(m.parts,
+		part{
+			header: m.getPartHeader(contentType),
+			copier: func(w io.Writer) error {
+				_, err := io.WriteString(w, body)
+				return err
+			},
+		},
+	)
+}
+
+// AddAlternativeWriter adds an alternative part to the message. It can be
+// useful with the text/template or html/template packages.
+func (m *Message) AddAlternativeWriter(contentType string, f func(io.Writer) error) {
+	m.parts = []part{
+		{
+			header: m.getPartHeader(contentType),
+			copier: f,
+		},
+	}
+}
+
+func (m *Message) getPartHeader(contentType string) header {
+	return map[string][]string{
+		"Content-Type":              {contentType + "; charset=" + m.charset},
+		"Content-Transfer-Encoding": {string(m.encoding)},
+	}
+}
+
+type file struct {
+	Name     string
+	Header   map[string][]string
+	CopyFunc func(w io.Writer) error
+}
+
+func (f *file) setHeader(field, value string) {
+	f.Header[field] = []string{value}
+}
+
+// A FileSetting can be used as an argument in Message.Attach or Message.Embed.
+type FileSetting func(*file)
+
+// SetHeader is a file setting to set the MIME header of the message part that
+// contains the file content.
+//
+// Mandatory headers are automatically added if they are not set when sending
+// the email.
+func SetHeader(h map[string][]string) FileSetting {
+	return func(f *file) {
+		for k, v := range h {
+			f.Header[k] = v
+		}
+	}
+}
+
+// SetCopyFunc is a file setting to replace the function that runs when the
+// message is sent. It should copy the content of the file to the io.Writer.
+//
+// The default copy function opens the file with the given filename, and copy
+// its content to the io.Writer.
+func SetCopyFunc(f func(io.Writer) error) FileSetting {
+	return func(fi *file) {
+		fi.CopyFunc = f
+	}
+}
+
+func (m *Message) appendFile(list []*file, name string, settings []FileSetting) []*file {
+	f := &file{
+		Name:   filepath.Base(name),
+		Header: make(map[string][]string),
+		CopyFunc: func(w io.Writer) error {
+			h, err := os.Open(name)
+			if err != nil {
+				return err
+			}
+			if _, err := io.Copy(w, h); err != nil {
+				h.Close()
+				return err
+			}
+			return h.Close()
+		},
+	}
+
+	for _, s := range settings {
+		s(f)
+	}
+
+	if list == nil {
+		return []*file{f}
+	}
+
+	return append(list, f)
+}
+
+// Attach attaches the files to the email.
+func (m *Message) Attach(filename string, settings ...FileSetting) {
+	m.attachments = m.appendFile(m.attachments, filename, settings)
+}
+
+// Embed embeds the images to the email.
+func (m *Message) Embed(filename string, settings ...FileSetting) {
+	m.embedded = m.appendFile(m.embedded, filename, settings)
+}
diff --git a/backend/vendor/gopkg.in/gomail.v2/mime.go b/backend/vendor/gopkg.in/gomail.v2/mime.go
new file mode 100644
index 00000000..51cba724
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/mime.go
@@ -0,0 +1,19 @@
+// +build go1.5
+
+package gomail
+
+import (
+	"mime"
+	"mime/quotedprintable"
+)
+
+var newQPWriter = quotedprintable.NewWriter
+
+type mimeEncoder struct {
+	mime.WordEncoder
+}
+
+var (
+	bEncoding = mimeEncoder{mime.BEncoding}
+	qEncoding = mimeEncoder{mime.QEncoding}
+)
diff --git a/backend/vendor/gopkg.in/gomail.v2/mime_go14.go b/backend/vendor/gopkg.in/gomail.v2/mime_go14.go
new file mode 100644
index 00000000..246e2e5e
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/mime_go14.go
@@ -0,0 +1,16 @@
+// +build !go1.5
+
+package gomail
+
+import "gopkg.in/alexcesaro/quotedprintable.v3"
+
+var newQPWriter = quotedprintable.NewWriter
+
+type mimeEncoder struct {
+	quotedprintable.WordEncoder
+}
+
+var (
+	bEncoding = mimeEncoder{quotedprintable.BEncoding}
+	qEncoding = mimeEncoder{quotedprintable.QEncoding}
+)
diff --git a/backend/vendor/gopkg.in/gomail.v2/send.go b/backend/vendor/gopkg.in/gomail.v2/send.go
new file mode 100644
index 00000000..3e672650
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/send.go
@@ -0,0 +1,117 @@
+package gomail
+
+import (
+	"errors"
+	"fmt"
+	"io"
+	"net/mail"
+)
+
+// Sender is the interface that wraps the Send method.
+//
+// Send sends an email to the given addresses.
+type Sender interface {
+	Send(from string, to []string, msg io.WriterTo) error
+}
+
+// SendCloser is the interface that groups the Send and Close methods.
+type SendCloser interface {
+	Sender
+	Close() error
+}
+
+// A SendFunc is a function that sends emails to the given adresses.
+//
+// The SendFunc type is an adapter to allow the use of ordinary functions as
+// email senders. If f is a function with the appropriate signature, SendFunc(f)
+// is a Sender object that calls f.
+type SendFunc func(from string, to []string, msg io.WriterTo) error
+
+// Send calls f(from, to, msg).
+func (f SendFunc) Send(from string, to []string, msg io.WriterTo) error {
+	return f(from, to, msg)
+}
+
+// Send sends emails using the given Sender.
+func Send(s Sender, msg ...*Message) error {
+	for i, m := range msg {
+		if err := send(s, m); err != nil {
+			return fmt.Errorf("gomail: could not send email %d: %v", i+1, err)
+		}
+	}
+
+	return nil
+}
+
+func send(s Sender, m *Message) error {
+	from, err := m.getFrom()
+	if err != nil {
+		return err
+	}
+
+	to, err := m.getRecipients()
+	if err != nil {
+		return err
+	}
+
+	if err := s.Send(from, to, m); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func (m *Message) getFrom() (string, error) {
+	from := m.header["Sender"]
+	if len(from) == 0 {
+		from = m.header["From"]
+		if len(from) == 0 {
+			return "", errors.New(`gomail: invalid message, "From" field is absent`)
+		}
+	}
+
+	return parseAddress(from[0])
+}
+
+func (m *Message) getRecipients() ([]string, error) {
+	n := 0
+	for _, field := range []string{"To", "Cc", "Bcc"} {
+		if addresses, ok := m.header[field]; ok {
+			n += len(addresses)
+		}
+	}
+	list := make([]string, 0, n)
+
+	for _, field := range []string{"To", "Cc", "Bcc"} {
+		if addresses, ok := m.header[field]; ok {
+			for _, a := range addresses {
+				addr, err := parseAddress(a)
+				if err != nil {
+					return nil, err
+				}
+				list = addAddress(list, addr)
+			}
+		}
+	}
+
+	return list, nil
+}
+
+func addAddress(list []string, addr string) []string {
+	for _, a := range list {
+		if addr == a {
+			return list
+		}
+	}
+
+	return append(list, addr)
+}
+
+func parseAddress(field string) (string, error) {
+	a, err := mail.ParseAddress(field)
+	if a == nil {
+		return "", err
+	}
+
+	return a.Address, err
+}
diff --git a/backend/vendor/gopkg.in/gomail.v2/smtp.go b/backend/vendor/gopkg.in/gomail.v2/smtp.go
new file mode 100644
index 00000000..cf773a10
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/smtp.go
@@ -0,0 +1,175 @@
+package gomail
+
+import (
+	"crypto/tls"
+	"fmt"
+	"io"
+	"net"
+	"net/smtp"
+)
+
+// A Dialer is a dialer to an SMTP server.
+type Dialer struct {
+	// Host represents the host of the SMTP server.
+	Host string
+	// Port represents the port of the SMTP server.
+	Port int
+	// Auth represents the authentication mechanism used to authenticate to the
+	// SMTP server.
+	Auth smtp.Auth
+	// SSL defines whether an SSL connection is used. It should be false in
+	// most cases since the authentication mechanism should use the STARTTLS
+	// extension instead.
+	SSL bool
+	// TSLConfig represents the TLS configuration used for the TLS (when the
+	// STARTTLS extension is used) or SSL connection.
+	TLSConfig *tls.Config
+}
+
+// NewPlainDialer returns a Dialer. The given parameters are used to connect to
+// the SMTP server via a PLAIN authentication mechanism.
+//
+// It fallbacks to the LOGIN mechanism if it is the only mechanism advertised by
+// the server.
+func NewPlainDialer(host string, port int, username, password string) *Dialer {
+	return &Dialer{
+		Host: host,
+		Port: port,
+		Auth: &plainAuth{
+			username: username,
+			password: password,
+			host:     host,
+		},
+		SSL: port == 465,
+	}
+}
+
+// Dial dials and authenticates to an SMTP server. The returned SendCloser
+// should be closed when done using it.
+func (d *Dialer) Dial() (SendCloser, error) {
+	c, err := d.dial()
+	if err != nil {
+		return nil, err
+	}
+
+	if d.Auth != nil {
+		if ok, _ := c.Extension("AUTH"); ok {
+			if err = c.Auth(d.Auth); err != nil {
+				c.Close()
+				return nil, err
+			}
+		}
+	}
+
+	return &smtpSender{c}, nil
+}
+
+func (d *Dialer) dial() (smtpClient, error) {
+	if d.SSL {
+		return d.sslDial()
+	}
+	return d.starttlsDial()
+}
+
+func (d *Dialer) starttlsDial() (smtpClient, error) {
+	c, err := smtpDial(addr(d.Host, d.Port))
+	if err != nil {
+		return nil, err
+	}
+
+	if ok, _ := c.Extension("STARTTLS"); ok {
+		if err := c.StartTLS(d.tlsConfig()); err != nil {
+			c.Close()
+			return nil, err
+		}
+	}
+
+	return c, nil
+}
+
+func (d *Dialer) sslDial() (smtpClient, error) {
+	conn, err := tlsDial("tcp", addr(d.Host, d.Port), d.tlsConfig())
+	if err != nil {
+		return nil, err
+	}
+
+	return newClient(conn, d.Host)
+}
+
+func (d *Dialer) tlsConfig() *tls.Config {
+	if d.TLSConfig == nil {
+		return &tls.Config{ServerName: d.Host}
+	}
+
+	return d.TLSConfig
+}
+
+func addr(host string, port int) string {
+	return fmt.Sprintf("%s:%d", host, port)
+}
+
+// DialAndSend opens a connection to the SMTP server, sends the given emails and
+// closes the connection.
+func (d *Dialer) DialAndSend(m ...*Message) error {
+	s, err := d.Dial()
+	if err != nil {
+		return err
+	}
+	defer s.Close()
+
+	return Send(s, m...)
+}
+
+type smtpSender struct {
+	smtpClient
+}
+
+func (c *smtpSender) Send(from string, to []string, msg io.WriterTo) error {
+	if err := c.Mail(from); err != nil {
+		return err
+	}
+
+	for _, addr := range to {
+		if err := c.Rcpt(addr); err != nil {
+			return err
+		}
+	}
+
+	w, err := c.Data()
+	if err != nil {
+		return err
+	}
+
+	if _, err = msg.WriteTo(w); err != nil {
+		w.Close()
+		return err
+	}
+
+	return w.Close()
+}
+
+func (c *smtpSender) Close() error {
+	return c.Quit()
+}
+
+// Stubbed out for tests.
+var (
+	smtpDial = func(addr string) (smtpClient, error) {
+		return smtp.Dial(addr)
+	}
+	tlsDial   = tls.Dial
+	newClient = func(conn net.Conn, host string) (smtpClient, error) {
+		return smtp.NewClient(conn, host)
+	}
+)
+
+type smtpClient interface {
+	Extension(string) (bool, string)
+	StartTLS(*tls.Config) error
+	Auth(smtp.Auth) error
+	Mail(string) error
+	Rcpt(string) error
+	Data() (io.WriteCloser, error)
+	Quit() error
+	Close() error
+}
diff --git a/backend/vendor/gopkg.in/gomail.v2/writeto.go b/backend/vendor/gopkg.in/gomail.v2/writeto.go
new file mode 100644
index 00000000..57a1dd7f
--- /dev/null
+++ b/backend/vendor/gopkg.in/gomail.v2/writeto.go
@@ -0,0 +1,242 @@
+package gomail
+
+import (
+	"encoding/base64"
+	"errors"
+	"io"
+	"mime"
+	"mime/multipart"
+	"path/filepath"
+	"time"
+)
+
+// WriteTo implements io.WriterTo. It dumps the whole message into w.
+func (m *Message) WriteTo(w io.Writer) (int64, error) {
+	mw := &messageWriter{w: w}
+	mw.writeMessage(m)
+	return mw.n, mw.err
+}
+
+func (w *messageWriter) writeMessage(m *Message) {
+	if _, ok := m.header["Mime-Version"]; !ok {
+		w.writeString("Mime-Version: 1.0\r\n")
+	}
+	if _, ok := m.header["Date"]; !ok {
+		w.writeHeader("Date", m.FormatDate(now()))
+	}
+	w.writeHeaders(m.header)
+
+	if m.hasMixedPart() {
+		w.openMultipart("mixed")
+	}
+
+	if m.hasRelatedPart() {
+		w.openMultipart("related")
+	}
+
+	if m.hasAlternativePart() {
+		w.openMultipart("alternative")
+	}
+	for _, part := range m.parts {
+		w.writeHeaders(part.header)
+		w.writeBody(part.copier, m.encoding)
+	}
+	if m.hasAlternativePart() {
+		w.closeMultipart()
+	}
+
+	w.addFiles(m.embedded, false)
+	if m.hasRelatedPart() {
+		w.closeMultipart()
+	}
+
+	w.addFiles(m.attachments, true)
+	if m.hasMixedPart() {
+		w.closeMultipart()
+	}
+}
+
+func (m *Message) hasMixedPart() bool {
+	return (len(m.parts) > 0 && len(m.attachments) > 0) || len(m.attachments) > 1
+}
+
+func (m *Message) hasRelatedPart() bool {
+	return (len(m.parts) > 0 && len(m.embedded) > 0) || len(m.embedded) > 1
+}
+
+func (m *Message) hasAlternativePart() bool {
+	return len(m.parts) > 1
+}
+
+type messageWriter struct {
+	w          io.Writer
+	n          int64
+	writers    [3]*multipart.Writer
+	partWriter io.Writer
+	depth      uint8
+	err        error
+}
+
+func (w *messageWriter) openMultipart(mimeType string) {
+	mw := multipart.NewWriter(w)
+	contentType := "multipart/" + mimeType + "; boundary=" + mw.Boundary()
+	w.writers[w.depth] = mw
+
+	if w.depth == 0 {
+		w.writeHeader("Content-Type", contentType)
+		w.writeString("\r\n")
+	} else {
+		w.createPart(map[string][]string{
+			"Content-Type": {contentType},
+		})
+	}
+	w.depth++
+}
+
+func (w *messageWriter) createPart(h map[string][]string) {
+	w.partWriter, w.err = w.writers[w.depth-1].CreatePart(h)
+}
+
+func (w *messageWriter) closeMultipart() {
+	if w.depth > 0 {
+		w.writers[w.depth-1].Close()
+		w.depth--
+	}
+}
+
+func (w *messageWriter) addFiles(files []*file, isAttachment bool) {
+	for _, f := range files {
+		if _, ok := f.Header["Content-Type"]; !ok {
+			mediaType := mime.TypeByExtension(filepath.Ext(f.Name))
+			if mediaType == "" {
+				mediaType = "application/octet-stream"
+			}
+			f.setHeader("Content-Type", mediaType+`; name="`+f.Name+`"`)
+		}
+
+		if _, ok := f.Header["Content-Transfer-Encoding"]; !ok {
+			f.setHeader("Content-Transfer-Encoding", string(Base64))
+		}
+
+		if _, ok := f.Header["Content-Disposition"]; !ok {
+			var disp string
+			if isAttachment {
+				disp = "attachment"
+			} else {
+				disp = "inline"
+			}
+			f.setHeader("Content-Disposition", disp+`; filename="`+f.Name+`"`)
+		}
+
+		if !isAttachment {
+			if _, ok := f.Header["Content-ID"]; !ok {
+				f.setHeader("Content-ID", "<"+f.Name+">")
+			}
+		}
+		w.writeHeaders(f.Header)
+		w.writeBody(f.CopyFunc, Base64)
+	}
+}
+
+func (w *messageWriter) Write(p []byte) (int, error) {
+	if w.err != nil {
+		return 0, errors.New("gomail: cannot write as writer is in error")
+	}
+
+	var n int
+	n, w.err = w.w.Write(p)
+	w.n += int64(n)
+	return n, w.err
+}
+
+func (w *messageWriter) writeString(s string) {
+	n, _ := io.WriteString(w.w, s)
+	w.n += int64(n)
+}
+
+func (w *messageWriter) writeStrings(a []string, sep string) {
+	if len(a) > 0 {
+		w.writeString(a[0])
+		if len(a) == 1 {
+			return
+		}
+	}
+	for _, s := range a[1:] {
+		w.writeString(sep)
+		w.writeString(s)
+	}
+}
+
+func (w *messageWriter) writeHeader(k string, v ...string) {
+	w.writeString(k)
+	w.writeString(": ")
+	w.writeStrings(v, ", ")
+	w.writeString("\r\n")
+}
+
+func (w *messageWriter) writeHeaders(h map[string][]string) {
+	if w.depth == 0 {
+		for k, v := range h {
+			if k != "Bcc" {
+				w.writeHeader(k, v...)
+			}
+		}
+	} else {
+		w.createPart(h)
+	}
+}
+
+func (w *messageWriter) writeBody(f func(io.Writer) error, enc Encoding) {
+	var subWriter io.Writer
+	if w.depth == 0 {
+		w.writeString("\r\n")
+		subWriter = w.w
+	} else {
+		subWriter = w.partWriter
+	}
+
+	if enc == Base64 {
+		wc := base64.NewEncoder(base64.StdEncoding, newBase64LineWriter(subWriter))
+		w.err = f(wc)
+		wc.Close()
+	} else if enc == Unencoded {
+		w.err = f(subWriter)
+	} else {
+		wc := newQPWriter(subWriter)
+		w.err = f(wc)
+		wc.Close()
+	}
+}
+
+// As required by RFC 2045, 6.7. (page 21) for quoted-printable, and
+// RFC 2045, 6.8. (page 25) for base64.
+const maxLineLen = 76
+
+// base64LineWriter limits text encoded in base64 to 76 characters per line
+type base64LineWriter struct {
+	w       io.Writer
+	lineLen int
+}
+
+func newBase64LineWriter(w io.Writer) *base64LineWriter {
+	return &base64LineWriter{w: w}
+}
+
+func (w *base64LineWriter) Write(p []byte) (int, error) {
+	n := 0
+	for len(p)+w.lineLen > maxLineLen {
+		w.w.Write(p[:maxLineLen-w.lineLen])
+		w.w.Write([]byte("\r\n"))
+		p = p[maxLineLen-w.lineLen:]
+		n += maxLineLen - w.lineLen
+		w.lineLen = 0
+	}
+
+	w.w.Write(p)
+	w.lineLen += len(p)
+
+	return n + len(p), nil
+}
+
+// Stubbed out for testing.
+var now = time.Now

From 875332dba76d38b83ea3a7b9c0d528f404fc28a2 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Tue, 14 Jan 2020 11:49:42 +0800
Subject: [PATCH 07/42] =?UTF-8?q?=E6=9B=B4=E6=96=B0=E9=82=AE=E4=BB=B6?=
 =?UTF-8?q?=E9=80=9A=E7=9F=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/services/notification/mail.go | 17 +++--
 backend/services/task.go              | 99 +++++++++++++++++++++------
 backend/utils/time.go                 | 16 +++++
 3 files changed, 107 insertions(+), 25 deletions(-)
 create mode 100644 backend/utils/time.go

diff --git a/backend/services/notification/mail.go b/backend/services/notification/mail.go
index 48c86351..2231151b 100644
--- a/backend/services/notification/mail.go
+++ b/backend/services/notification/mail.go
@@ -11,13 +11,13 @@ import (
 	"strconv"
 )
 
-func SendMail(toEmail string, subject string, content string) error {
+func SendMail(toEmail string, toName string, subject string, content string) error {
 	// hermes instance
 	h := hermes.Hermes{
+		Theme: new(hermes.Default),
 		Product: hermes.Product{
-			Name: "Hermes",
-			Link: "https://example-hermes.com/",
-			Logo: "http://www.duchess-france.org/wp-content/uploads/2016/01/gopher.png",
+			Name:      "Crawlab Team",
+			Copyright: "© 2019 Crawlab, Made by Crawlab-Team",
 		},
 	}
 
@@ -41,7 +41,8 @@ func SendMail(toEmail string, subject string, content string) error {
 	// email instance
 	email := hermes.Email{
 		Body: hermes.Body{
-			FreeMarkdown: hermes.Markdown(content),
+			Name:         toName,
+			FreeMarkdown: hermes.Markdown(content + GetFooter()),
 		},
 	}
 
@@ -129,3 +130,9 @@ func send(smtpConfig smtpAuthentication, options sendOptions, htmlBody string, t
 
 	return d.DialAndSend(m)
 }
+
+func GetFooter() string {
+	return `
+[Github](https://github.com/crawlab-team/crawlab) | [Documentation](http://docs.crawlab.cn) | [Docker](https://hub.docker.com/r/tikazyq/crawlab)
+`
+}
diff --git a/backend/services/task.go b/backend/services/task.go
index ea7c47c3..16445213 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -476,9 +476,23 @@ func ExecuteTask(id int) {
 		defer cronExec.Stop()
 	}
 
+	// 获得触发任务用户
+	user, err := model.GetUser(t.UserId)
+	if err != nil {
+		log.Errorf(GetWorkerPrefix(id) + err.Error())
+		return
+	}
+
 	// 执行Shell命令
 	if err := ExecuteShellCmd(cmd, cwd, t, spider); err != nil {
 		log.Errorf(GetWorkerPrefix(id) + err.Error())
+
+		// 如果发生错误，则发送通知
+		t, _ = model.GetTask(t.Id)
+		if user.Email != "" &&
+			(user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd || user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskError) {
+			SendTaskEmail(user, t, spider)
+		}
 		return
 	}
 
@@ -501,29 +515,10 @@ func ExecuteTask(id int) {
 	t.RuntimeDuration = t.FinishTs.Sub(t.StartTs).Seconds() // 运行时长
 	t.TotalDuration = t.FinishTs.Sub(t.CreateTs).Seconds()  // 总时长
 
-	// 获得触发任务用户
-	user, err := model.GetUser(t.UserId)
-	if err != nil {
-		log.Errorf(GetWorkerPrefix(id) + err.Error())
-		return
-	}
-
 	// 如果是任务结束时发送通知，则发送通知
 	if user.Email != "" &&
 		user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
-		emailContent := fmt.Sprintf(`
-# Your task has finished
-
-The stats of the task are as below.
-
-Metric | Value
---- | ---
-Create Time | %s
-Start Time | %s
-Finish Time | %s
-Total Duration | %.1f
-`, t.CreateTs, t.StartTs, t.FinishTs, t.TotalDuration)
-		_ = notification.SendMail(user.Email, fmt.Sprintf("[%s] Crawlab Task Finished", spider.Name), emailContent)
+		SendTaskEmail(user, t, spider)
 	}
 
 	// 保存任务
@@ -696,6 +691,70 @@ func HandleTaskError(t model.Task, err error) {
 	debug.PrintStack()
 }
 
+func GetTaskEmailContent(t model.Task, s model.Spider) string {
+	n, _ := model.GetNode(t.NodeId)
+	errMsg := ""
+	statusMsg := fmt.Sprintf(`<span style="color:green">%s</span>`, t.Status)
+	if t.Status == constants.StatusError {
+		errMsg = " with errors"
+		statusMsg = fmt.Sprintf(`<span style="color:red">%s</span>`, t.Status)
+	}
+	return fmt.Sprintf(`
+Your task has finished%s. Please find the task info below.
+
+ | 
+--: | :--
+**Task ID:** | %s
+**Task Status:** | %s
+**Task Param:** | %s
+**Spider ID:** | %s
+**Spider Name:** | %s
+**Node:** | %s
+**Create Time:** | %s
+**Start Time:** | %s
+**Finish Time:** | %s
+**Wait Duration:** | %.0f sec
+**Runtime Duration:** | %.0f sec
+**Total Duration:** | %.0f sec
+**Number of Results:** | %d
+**Error:** | <span style="color:red">%s</span>
+
+Please login to Crawlab to view the details.
+`,
+		errMsg,
+		t.Id,
+		statusMsg,
+		t.Param,
+		s.Id.Hex(),
+		s.Name,
+		n.Name,
+		utils.GetLocalTimeString(t.CreateTs),
+		utils.GetLocalTimeString(t.StartTs),
+		utils.GetLocalTimeString(t.FinishTs),
+		t.WaitDuration,
+		t.RuntimeDuration,
+		t.TotalDuration,
+		t.ResultCount,
+		t.Error,
+	)
+}
+
+func SendTaskEmail(u model.User, t model.Task, s model.Spider) {
+	statusMsg := "has finished"
+	if t.Status == constants.StatusError {
+		statusMsg = "has an error"
+	}
+	if err := notification.SendMail(
+		u.Email,
+		u.Username,
+		fmt.Sprintf("[Crawlab] Task for \"%s\" %s", s.Name, statusMsg),
+		GetTaskEmailContent(t, s),
+	); err != nil {
+		log.Errorf("mail error: " + err.Error())
+		debug.PrintStack()
+	}
+}
+
 func InitTaskExecutor() error {
 	c := cron.New(cron.WithSeconds())
 	Exec = &Executor{
diff --git a/backend/utils/time.go b/backend/utils/time.go
new file mode 100644
index 00000000..84b40f4e
--- /dev/null
+++ b/backend/utils/time.go
@@ -0,0 +1,16 @@
+package utils
+
+import "time"
+
+func GetLocalTime(t time.Time) time.Time {
+	return t.In(time.Local)
+}
+
+func GetTimeString(t time.Time) string {
+	return t.Format("2006-01-02 15:04:05")
+}
+
+func GetLocalTimeString(t time.Time) string {
+	t = GetLocalTime(t)
+	return GetTimeString(t)
+}

From c227cdeaba9b127ef3a2bcfcd2af3146f2d31e25 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Tue, 14 Jan 2020 15:42:51 +0800
Subject: [PATCH 08/42] =?UTF-8?q?=E9=BB=98=E8=AE=A4=E4=B8=BA=E4=B8=8D?=
 =?UTF-8?q?=E8=A7=A6=E5=8F=91=E9=80=9A=E7=9F=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/routes/user.go   | 3 +++
 backend/services/user.go | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/backend/routes/user.go b/backend/routes/user.go
index 3d636f41..8aae8b24 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -101,6 +101,9 @@ func PutUser(c *gin.Context) {
 		Password: utils.EncryptPassword(reqData.Password),
 		Role:     reqData.Role,
 		Email:    reqData.Email,
+		Setting: model.UserSetting{
+			NotificationTrigger: constants.NotificationTriggerNever,
+		},
 	}
 	if err := user.Add(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
diff --git a/backend/services/user.go b/backend/services/user.go
index f20b9e7a..483a7408 100644
--- a/backend/services/user.go
+++ b/backend/services/user.go
@@ -17,6 +17,9 @@ func InitUserService() error {
 		Username: "admin",
 		Password: utils.EncryptPassword("admin"),
 		Role:     constants.RoleAdmin,
+		Setting: model.UserSetting{
+			NotificationTrigger: constants.NotificationTriggerNever,
+		},
 	}
 	_ = adminUser.Add()
 	return nil

From c318f310f549147f3b27ba05aac671c86639b46d Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Tue, 14 Jan 2020 15:57:52 +0800
Subject: [PATCH 09/42] fixed https issue

---
 frontend/src/api/request.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/frontend/src/api/request.js b/frontend/src/api/request.js
index 3b13c014..8504f624 100644
--- a/frontend/src/api/request.js
+++ b/frontend/src/api/request.js
@@ -5,7 +5,7 @@ import { Message } from 'element-ui'
 // 根据 VUE_APP_BASE_URL 生成 baseUrl
 let baseUrl = process.env.VUE_APP_BASE_URL ? process.env.VUE_APP_BASE_URL : 'http://localhost:8000'
 if (!baseUrl.match(/^https?/i)) {
-  baseUrl = `http://${window.location.host}${process.env.VUE_APP_BASE_URL}`
+  baseUrl = `${window.location.protocol}//${window.location.host}${process.env.VUE_APP_BASE_URL}`
 }
 
 // 如果 Docker 中设置了 CRAWLAB_API_ADDRESS 这个环境变量，则会将 baseUrl 覆盖

From 720492ed38fc35115b49195817b751e4667c500f Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Tue, 14 Jan 2020 17:29:19 +0800
Subject: [PATCH 10/42] updated setting

---
 backend/model/user.go                  |  2 ++
 backend/routes/user.go                 |  8 ++++++++
 backend/utils/encrypt.go               | 16 ++++++++++++++++
 frontend/src/i18n/zh.js                |  2 ++
 frontend/src/views/setting/Setting.vue | 26 ++++++++++++++++++++++++--
 5 files changed, 52 insertions(+), 2 deletions(-)
 create mode 100644 backend/utils/encrypt.go

diff --git a/backend/model/user.go b/backend/model/user.go
index 246fa448..3e24e096 100644
--- a/backend/model/user.go
+++ b/backend/model/user.go
@@ -25,6 +25,8 @@ type User struct {
 
 type UserSetting struct {
 	NotificationTrigger string `json:"notification_trigger" bson:"notification_trigger"`
+	DingTalkAppKey      string `json:"ding_talk_app_key" bson:"ding_talk_app_key"`
+	DingTalkAppSecret   string `json:"ding_talk_app_secret" bson:"ding_talk_app_secret"`
 }
 
 func (user *User) Save() error {
diff --git a/backend/routes/user.go b/backend/routes/user.go
index 8aae8b24..9afddffc 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -216,6 +216,8 @@ func PostMe(c *gin.Context) {
 		Email               string `json:"email"`
 		Password            string `json:"password"`
 		NotificationTrigger string `json:"notification_trigger"`
+		DingTalkAppKey      string `json:"ding_talk_app_key"`
+		DingTalkAppSecret   string `json:"ding_talk_app_secret"`
 	}
 	ctx := context.WithGinContext(c)
 	user := ctx.User()
@@ -237,6 +239,12 @@ func PostMe(c *gin.Context) {
 	if reqBody.NotificationTrigger != "" {
 		user.Setting.NotificationTrigger = reqBody.NotificationTrigger
 	}
+	if reqBody.DingTalkAppKey != "" {
+		user.Setting.DingTalkAppKey = reqBody.DingTalkAppKey
+	}
+	if reqBody.DingTalkAppSecret != "" {
+		user.Setting.DingTalkAppSecret = reqBody.DingTalkAppSecret
+	}
 	if err := user.Save(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
 		return
diff --git a/backend/utils/encrypt.go b/backend/utils/encrypt.go
new file mode 100644
index 00000000..52013b9c
--- /dev/null
+++ b/backend/utils/encrypt.go
@@ -0,0 +1,16 @@
+package utils
+
+import (
+	"crypto/hmac"
+	"crypto/sha256"
+	"encoding/base64"
+	"encoding/hex"
+)
+
+func ComputeHmacSha256(message string, secret string) string {
+	key := []byte(secret)
+	h := hmac.New(sha256.New, key)
+	h.Write([]byte(message))
+	sha := hex.EncodeToString(h.Sum(nil))
+	return base64.StdEncoding.EncodeToString([]byte(sha))
+}
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index 841316fb..740408a2 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -353,6 +353,8 @@ export default {
   'On Task End': '当任务结束',
   'On Task Error': '当任务发生错误',
   'Never': '从不',
+  'DingTalk AppKey': '钉钉 AppKey',
+  'DingTalk AppSecret': '钉钉 AppSecret',
 
   // 其他
   tagsView: {
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
index 83695f59..49a38700 100644
--- a/frontend/src/views/setting/Setting.vue
+++ b/frontend/src/views/setting/Setting.vue
@@ -24,6 +24,25 @@
           </el-radio>
         </el-radio-group>
       </el-form-item>
+      <el-form-item prop="ding_talk_app_key" :label="$t('DingTalk AppKey')">
+        <el-input v-model="userInfo.setting.ding_talk_app_key" :placeholder="$t('DingTalk AppKey')"></el-input>
+      </el-form-item>
+      <el-form-item prop="ding_talk_app_secret" :label="$t('DingTalk AppSecret')">
+        <template v-if="isShowDingTalkAppSecret">
+          <el-input
+            v-model="userInfo.setting.ding_talk_app_secret"
+            :placeholder="$t('DingTalk AppSecret')"
+          />
+        </template>
+        <template v-else>
+          <el-input
+            type="password"
+            suffix-icon="el-icon-view"
+            v-model="userInfo.setting.ding_talk_app_secret"
+            :placeholder="$t('DingTalk AppSecret')"
+          />
+        </template>
+      </el-form-item>
       <el-form-item>
         <div class="buttons">
           <el-button type="success" @click="saveUserInfo">{{$t('Save')}}</el-button>
@@ -58,7 +77,8 @@ export default {
       rules: {
         password: [{ trigger: 'blur', validator: validatePass }],
         email: [{ trigger: 'blur', validator: validateEmail }]
-      }
+      },
+      isShowDingTalkAppSecret: false
     }
   },
   methods: {
@@ -74,7 +94,9 @@ export default {
         const res = await this.$store.dispatch('user/postInfo', {
           password: this.userInfo.password,
           email: this.userInfo.email,
-          notification_trigger: this.userInfo.setting.notification_trigger
+          notification_trigger: this.userInfo.setting.notification_trigger,
+          ding_talk_app_key: this.userInfo.setting.ding_talk_app_key,
+          ding_talk_app_secret: this.userInfo.setting.ding_talk_app_secret
         })
         if (!res || res.error) {
           this.$message.error(res.error)

From 8069000956ab4b29bc466207f9700a552e0224db Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Tue, 14 Jan 2020 20:03:04 +0800
Subject: [PATCH 11/42] added dingtalk

---
 backend/services/notification/ding_talk.go | 84 ++++++++++++++++++++++
 backend/services/task.go                   |  4 ++
 frontend/src/views/setting/Setting.vue     | 16 ++++-
 3 files changed, 102 insertions(+), 2 deletions(-)
 create mode 100644 backend/services/notification/ding_talk.go

diff --git a/backend/services/notification/ding_talk.go b/backend/services/notification/ding_talk.go
new file mode 100644
index 00000000..0a1dac14
--- /dev/null
+++ b/backend/services/notification/ding_talk.go
@@ -0,0 +1,84 @@
+package notification
+
+import (
+	"crawlab/model"
+	"crawlab/utils"
+	"errors"
+	"fmt"
+	"github.com/apex/log"
+	"github.com/imroc/req"
+	"runtime/debug"
+	"time"
+)
+
+func SendDingTalkNotification(t model.Task, s model.Spider) error {
+	// 获取用户
+	user, _ := model.GetUser(t.UserId)
+
+	// 如果AppKey或AppSecret未设置，则返回错误
+	if user.Setting.DingTalkAppKey == "" || user.Setting.DingTalkAppSecret == "" {
+		return errors.New("ding_talk_app_key or ding_talk_app_secret is empty")
+	}
+
+	// 获取access_token
+	accessToken, err := GetDingTalkAccessToken(user)
+	if err != nil {
+		return err
+	}
+
+	// 时间戳
+	timestamp := time.Now().Unix()
+
+	// 计算sign
+	signRawString := fmt.Sprintf("%d\n%s", timestamp, user.Setting.DingTalkAppSecret)
+	sign := utils.ComputeHmacSha256(signRawString, user.Setting.DingTalkAppSecret)
+
+	// 请求数据
+	url := fmt.Sprintf("https://oapi.dingtalk.com/robot/send?access_token=%s", accessToken)
+	header := req.Header{
+		"Content-Type": "application/json",
+		"timestamp":    fmt.Sprintf("%d000", timestamp),
+		"sign":         sign,
+	}
+	data := req.Param{
+		"msgtype": "text",
+		"text": req.Param{
+			"text": "it works",
+		},
+	}
+	res, err := req.Post(url, header, req.BodyJSON(&data))
+	if err != nil {
+		log.Errorf("dingtalk notification error: " + err.Error())
+		debug.PrintStack()
+		return err
+	}
+	log.Infof(fmt.Sprintf("%+v", res))
+	return nil
+}
+
+func GetDingTalkAccessToken(u model.User) (string, error) {
+	type ResBody struct {
+		ErrCode     int `json:"errcode"`
+		ErrMsg      string `json:"errmsg"`
+		AccessToken string `json:"access_token"`
+	}
+
+	// 请求数据
+	url := fmt.Sprintf("https://oapi.dingtalk.com/gettoken?appkey=%s&appsecret=%s", u.Setting.DingTalkAppKey, u.Setting.DingTalkAppSecret)
+	res, err := req.Get(url)
+	if err != nil {
+		log.Errorf("get dingtalk access_token error: " + err.Error())
+		debug.PrintStack()
+		return "", err
+	}
+
+	// 解析相应body
+	var resBody ResBody
+	if err := res.ToJSON(&resBody); err != nil {
+		log.Errorf("get dingtalk access_token error: " + err.Error())
+		debug.PrintStack()
+		return "", err
+	}
+
+	return resBody.AccessToken, nil
+}
diff --git a/backend/services/task.go b/backend/services/task.go
index 16445213..12c3580a 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -519,6 +519,10 @@ func ExecuteTask(id int) {
 	if user.Email != "" &&
 		user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
 		SendTaskEmail(user, t, spider)
+		if err := notification.SendDingTalkNotification(t, spider); err != nil {
+			log.Errorf(err.Error())
+			debug.PrintStack()
+		}
 	}
 
 	// 保存任务
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
index 49a38700..4ca476a5 100644
--- a/frontend/src/views/setting/Setting.vue
+++ b/frontend/src/views/setting/Setting.vue
@@ -33,14 +33,16 @@
             v-model="userInfo.setting.ding_talk_app_secret"
             :placeholder="$t('DingTalk AppSecret')"
           />
+
+          <i class="icon el-icon-view" @click="toggleDingTalkAppSecret"></i>
         </template>
         <template v-else>
           <el-input
-            type="password"
-            suffix-icon="el-icon-view"
             v-model="userInfo.setting.ding_talk_app_secret"
+            type="password"
             :placeholder="$t('DingTalk AppSecret')"
           />
+          <i class="icon el-icon-view" @click="toggleDingTalkAppSecret"></i>
         </template>
       </el-form-item>
       <el-form-item>
@@ -104,6 +106,9 @@ export default {
           this.$message.success(this.$t('Saved successfully'))
         }
       })
+    },
+    toggleDingTalkAppSecret () {
+      this.isShowDingTalkAppSecret = !this.isShowDingTalkAppSecret
     }
   },
   async created () {
@@ -121,4 +126,11 @@ export default {
   .setting-form .buttons {
     text-align: right;
   }
+
+  .setting-form .icon {
+    top: calc(50% - 14px / 2);
+    right: 14px;
+    position: absolute;
+    color: #DCDFE6;
+  }
 </style>

From 9e61e4c149b8d11ff531750e6d9891fe141840a4 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 10:31:58 +0800
Subject: [PATCH 12/42] added vendor

---
 backend/go.mod                                |   1 +
 backend/go.sum                                |   2 +
 backend/services/notification/ding_talk.go    |  53 ++++----
 .../vendor/github.com/royeo/dingrobot/LICENSE |  21 ++++
 .../github.com/royeo/dingrobot/README.md      | 110 ++++++++++++++++
 .../github.com/royeo/dingrobot/dingrobot.go   | 118 ++++++++++++++++++
 .../github.com/royeo/dingrobot/message.go     |  60 +++++++++
 backend/vendor/modules.txt                    |   2 +
 8 files changed, 346 insertions(+), 21 deletions(-)
 create mode 100644 backend/vendor/github.com/royeo/dingrobot/LICENSE
 create mode 100644 backend/vendor/github.com/royeo/dingrobot/README.md
 create mode 100644 backend/vendor/github.com/royeo/dingrobot/dingrobot.go
 create mode 100644 backend/vendor/github.com/royeo/dingrobot/message.go

diff --git a/backend/go.mod b/backend/go.mod
index 791c6771..cbc7d75b 100644
--- a/backend/go.mod
+++ b/backend/go.mod
@@ -16,6 +16,7 @@ require (
 	github.com/matcornic/hermes v1.2.0
 	github.com/matcornic/hermes/v2 v2.0.2 // indirect
 	github.com/pkg/errors v0.8.1
+	github.com/royeo/dingrobot v1.0.0
 	github.com/satori/go.uuid v1.2.0
 	github.com/smartystreets/goconvey v0.0.0-20190731233626-505e41936337
 	github.com/spf13/viper v1.4.0
diff --git a/backend/go.sum b/backend/go.sum
index 37fe6119..a55ef74e 100644
--- a/backend/go.sum
+++ b/backend/go.sum
@@ -146,6 +146,8 @@ github.com/prometheus/procfs v0.0.0-20190507164030-5867b95ac084/go.mod h1:TjEm7z
 github.com/prometheus/tsdb v0.7.1/go.mod h1:qhTCs0VvXwvX/y3TZrWD7rabWM+ijKTux40TwIPHuXU=
 github.com/rogpeppe/fastuuid v0.0.0-20150106093220-6724a57986af/go.mod h1:XWv6SoW27p1b0cqNHllgS5HIMJraePCO15w5zCzIWYg=
 github.com/rogpeppe/fastuuid v1.1.0/go.mod h1:jVj6XXZzXRy/MSR5jhDC/2q6DgLz+nrA6LYCDYWNEvQ=
+github.com/royeo/dingrobot v1.0.0 h1:K4GrF+fOecNX0yi+oBKpfh7z0XP/8TzaIIHu1B2kKUQ=
+github.com/royeo/dingrobot v1.0.0/go.mod h1:RqDM8E/hySCVwI2aUFRJAUGDcHHRnIhzNmbNG3bamQs=
 github.com/russross/blackfriday/v2 v2.0.1/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
 github.com/satori/go.uuid v1.2.0 h1:0uYX9dsZ2yD7q2RtLRtPSdGDWzjeM3TbMJP9utgA0ww=
 github.com/satori/go.uuid v1.2.0/go.mod h1:dA0hQrYB0VpLJoorglMZABFdXlWrHn1NEOzdhQKdks0=
diff --git a/backend/services/notification/ding_talk.go b/backend/services/notification/ding_talk.go
index 0a1dac14..ae1756aa 100644
--- a/backend/services/notification/ding_talk.go
+++ b/backend/services/notification/ding_talk.go
@@ -2,13 +2,12 @@ package notification
 
 import (
 	"crawlab/model"
-	"crawlab/utils"
 	"errors"
 	"fmt"
 	"github.com/apex/log"
 	"github.com/imroc/req"
+	"github.com/royeo/dingrobot"
 	"runtime/debug"
-	"time"
 )
 
 func SendDingTalkNotification(t model.Task, s model.Spider) error {
@@ -27,38 +26,50 @@ func SendDingTalkNotification(t model.Task, s model.Spider) error {
 	}
 
 	// 时间戳
-	timestamp := time.Now().Unix()
+	//timestamp := time.Now().Unix()
 
 	// 计算sign
-	signRawString := fmt.Sprintf("%d\n%s", timestamp, user.Setting.DingTalkAppSecret)
-	sign := utils.ComputeHmacSha256(signRawString, user.Setting.DingTalkAppSecret)
+	//signRawString := fmt.Sprintf("%d\n%s", timestamp, user.Setting.DingTalkAppSecret)
+	//sign := utils.ComputeHmacSha256(signRawString, user.Setting.DingTalkAppSecret)
 
 	// 请求数据
 	url := fmt.Sprintf("https://oapi.dingtalk.com/robot/send?access_token=%s", accessToken)
-	header := req.Header{
-		"Content-Type": "application/json",
-		"timestamp":    fmt.Sprintf("%d000", timestamp),
-		"sign":         sign,
-	}
-	data := req.Param{
-		"msgtype": "text",
-		"text": req.Param{
-			"text": "it works",
-		},
-	}
-	res, err := req.Post(url, header, req.BodyJSON(&data))
-	if err != nil {
-		log.Errorf("dingtalk notification error: " + err.Error())
+	robot := dingrobot.NewRobot(url)
+
+	text := "it works"
+	if err := robot.SendText(text, []string{}, false); err != nil {
+		log.Errorf(err.Error())
 		debug.PrintStack()
 		return err
 	}
-	log.Infof(fmt.Sprintf("%+v", res))
+	//header := req.Header{
+	//	"Content-Type": "application/json; charset=utf-8",
+	//	"timestamp":    fmt.Sprintf("%d000", timestamp),
+	//	"sign":         sign,
+	//}
+	//data := req.Param{
+	//	"msgtype": "text",
+	//	"text": req.Param{
+	//		"text": "it works",
+	//	},
+	//	"at": req.Param{
+	//		"atMobiles": []string{},
+	//		"isAtAll": false,
+	//	},
+	//}
+	//res, err := req.Post(url, header, req.BodyJSON(&data))
+	//if err != nil {
+	//	log.Errorf("dingtalk notification error: " + err.Error())
+	//	debug.PrintStack()
+	//	return err
+	//}
+	//log.Infof(fmt.Sprintf("%+v", res))
 	return nil
 }
 
 func GetDingTalkAccessToken(u model.User) (string, error) {
 	type ResBody struct {
-		ErrCode     int `json:"errcode"`
+		ErrCode     int    `json:"errcode"`
 		ErrMsg      string `json:"errmsg"`
 		AccessToken string `json:"access_token"`
 	}
diff --git a/backend/vendor/github.com/royeo/dingrobot/LICENSE b/backend/vendor/github.com/royeo/dingrobot/LICENSE
new file mode 100644
index 00000000..e2c20381
--- /dev/null
+++ b/backend/vendor/github.com/royeo/dingrobot/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2018 Royeo
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
\ No newline at end of file
diff --git a/backend/vendor/github.com/royeo/dingrobot/README.md b/backend/vendor/github.com/royeo/dingrobot/README.md
new file mode 100644
index 00000000..c76c6313
--- /dev/null
+++ b/backend/vendor/github.com/royeo/dingrobot/README.md
@@ -0,0 +1,110 @@
+# Dingrobot
+
+钉钉自定义机器人 Golang API.
+
+支持的消息类型：
+- 文本类型
+- link 类型
+- markdown 类型
+- 整体跳转 ActionCard 类型
+
+## Installation
+
+Install:
+
+```sh
+go get -u github.com/royeo/dingrobot
+```
+
+Import:
+
+```go
+import "github.com/royeo/dingrobot"
+```
+
+## Quick start
+
+发送文本类型的消息：
+
+```go
+func main() {
+	// You should replace the webhook here with your own.
+	webhook := "https://oapi.dingtalk.com/robot/send?access_token=xxx"
+	robot := dingrobot.NewRobot(webhook)
+
+	content := "我就是我,  @1825718XXXX 是不一样的烟火"
+	atMobiles := []string{"1825718XXXX"}
+	isAtAll := false
+
+	err := robot.SendText(content, atMobiles, isAtAll)
+	if err != nil {
+		log.Fatal(err)
+	}
+}
+```
+
+发送 link 类型的消息：
+
+```go
+func main() {
+	// You should replace the webhook here with your own.
+	webhook := "https://oapi.dingtalk.com/robot/send?access_token=xxx"
+	robot := dingrobot.NewRobot(webhook)
+
+	title := "自定义机器人协议"
+	text := "群机器人是钉钉群的高级扩展功能。群机器人可以将第三方服务的信息聚合到群聊中，实现自动化的信息同步。例如：通过聚合GitHub，GitLab等源码管理服务，实现源码更新同步；通过聚合Trello，JIRA等项目协调服务，实现项目信息同步。不仅如此，群机器人支持Webhook协议的自定义接入，支持更多可能性，例如：你可将运维报警提醒通过自定义机器人聚合到钉钉群。"
+	messageUrl := "https://open-doc.dingtalk.com/docs/doc.htm?spm=a219a.7629140.0.0.Rqyvqo&treeId=257&articleId=105735&docType=1"
+	picUrl := ""
+
+	err := robot.SendLink(title, text, messageUrl, picUrl)
+	if err != nil {
+		log.Fatal(err)
+	}
+}
+```
+
+发送 markdown 类型的消息：
+
+```go
+func main() {
+	// You should replace the webhook here with your own.
+	webhook := "https://oapi.dingtalk.com/robot/send?access_token=xxx"
+	robot := dingrobot.NewRobot(webhook)
+
+	title := "杭州天气"
+	text := "#### 杭州天气  \n > 9度，@1825718XXXX 西北风1级，空气良89，相对温度73%\n\n > ![screenshot](http://i01.lw.aliimg.com/media/lALPBbCc1ZhJGIvNAkzNBLA_1200_588.png)\n  > ###### 10点20分发布 [天气](http://www.thinkpage.cn/) "
+	atMobiles := []string{"1825718XXXX"}
+	isAtAll := false
+
+	err := robot.SendMarkdown(title, text, atMobiles, isAtAll)
+	if err != nil {
+		log.Fatal(err)
+	}
+}
+```
+
+发送整体跳转 ActionCard 类型的消息：
+
+```go
+func main() {
+	// You should replace the webhook here with your own.
+	webhook := "https://oapi.dingtalk.com/robot/send?access_token=xxx"
+	robot := dingrobot.NewRobot(webhook)
+
+	title := "乔布斯 20 年前想打造一间苹果咖啡厅，而它正是 Apple Store 的前身"
+	text := "![screenshot](@lADOpwk3K80C0M0FoA) \n #### 乔布斯 20 年前想打造的苹果咖啡厅 \n\n Apple Store 的设计正从原来满满的科技感走向生活化，而其生活化的走向其实可以追溯到 20 年前苹果一个建立咖啡馆的计划"
+	singleTitle := "阅读全文"
+	singleURL := "https://www.dingtalk.com/"
+	btnOrientation := "0"
+	hideAvatar := "0"
+
+	err := robot.SendActionCard(title, text, singleTitle, singleURL, btnOrientation, hideAvatar)
+	if err != nil {
+		log.Fatal(err)
+	}
+}
+```
+
+## License
+
+MIT Copyright (c) 2018 Royeo
diff --git a/backend/vendor/github.com/royeo/dingrobot/dingrobot.go b/backend/vendor/github.com/royeo/dingrobot/dingrobot.go
new file mode 100644
index 00000000..cd5b65b0
--- /dev/null
+++ b/backend/vendor/github.com/royeo/dingrobot/dingrobot.go
@@ -0,0 +1,118 @@
+package dingrobot
+
+import (
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"io/ioutil"
+	"net/http"
+)
+
+// Roboter is the interface implemented by Robot that can send multiple types of messages.
+type Roboter interface {
+	SendText(content string, atMobiles []string, isAtAll bool) error
+	SendLink(title, text, messageURL, picURL string) error
+	SendMarkdown(title, text string, atMobiles []string, isAtAll bool) error
+	SendActionCard(title, text, singleTitle, singleURL, btnOrientation, hideAvatar string) error
+}
+
+// Robot represents a dingtalk custom robot that can send messages to groups.
+type Robot struct {
+	Webhook string
+}
+
+// NewRobot returns a roboter that can send messages.
+func NewRobot(webhook string) Roboter {
+	return Robot{Webhook: webhook}
+}
+
+// SendText send a text type message.
+func (r Robot) SendText(content string, atMobiles []string, isAtAll bool) error {
+	return r.send(&textMessage{
+		MsgType: msgTypeText,
+		Text: textParams{
+			Content: content,
+		},
+		At: atParams{
+			AtMobiles: atMobiles,
+			IsAtAll:   isAtAll,
+		},
+	})
+}
+
+// SendLink send a link type message.
+func (r Robot) SendLink(title, text, messageURL, picURL string) error {
+	return r.send(&linkMessage{
+		MsgType: msgTypeLink,
+		Link: linkParams{
+			Title:      title,
+			Text:       text,
+			MessageURL: messageURL,
+			PicURL:     picURL,
+		},
+	})
+}
+
+// SendMarkdown send a markdown type message.
+func (r Robot) SendMarkdown(title, text string, atMobiles []string, isAtAll bool) error {
+	return r.send(&markdownMessage{
+		MsgType: msgTypeMarkdown,
+		Markdown: markdownParams{
+			Title: title,
+			Text:  text,
+		},
+		At: atParams{
+			AtMobiles: atMobiles,
+			IsAtAll:   isAtAll,
+		},
+	})
+}
+
+// SendActionCard send a action card type message.
+func (r Robot) SendActionCard(title, text, singleTitle, singleURL, btnOrientation, hideAvatar string) error {
+	return r.send(&actionCardMessage{
+		MsgType: msgTypeActionCard,
+		ActionCard: actionCardParams{
+			Title:          title,
+			Text:           text,
+			SingleTitle:    singleTitle,
+			SingleURL:      singleURL,
+			BtnOrientation: btnOrientation,
+			HideAvatar:     hideAvatar,
+		},
+	})
+}
+
+type dingResponse struct {
+	Errcode int
+	Errmsg  string
+}
+
+func (r Robot) send(msg interface{}) error {
+	m, err := json.Marshal(msg)
+	if err != nil {
+		return err
+	}
+
+	resp, err := http.Post(r.Webhook, "application/json", bytes.NewReader(m))
+	if err != nil {
+		return err
+	}
+	defer resp.Body.Close()
+
+	data, err := ioutil.ReadAll(resp.Body)
+	if err != nil {
+		return err
+	}
+
+	var dr dingResponse
+	err = json.Unmarshal(data, &dr)
+	if err != nil {
+		return err
+	}
+	if dr.Errcode != 0 {
+		return fmt.Errorf("dingrobot send failed: %v", dr.Errmsg)
+	}
+
+	return nil
+}
diff --git a/backend/vendor/github.com/royeo/dingrobot/message.go b/backend/vendor/github.com/royeo/dingrobot/message.go
new file mode 100644
index 00000000..82f286f8
--- /dev/null
+++ b/backend/vendor/github.com/royeo/dingrobot/message.go
@@ -0,0 +1,60 @@
+package dingrobot
+
+const (
+	msgTypeText       = "text"
+	msgTypeLink       = "link"
+	msgTypeMarkdown   = "markdown"
+	msgTypeActionCard = "actionCard"
+)
+
+type textMessage struct {
+	MsgType string     `json:"msgtype"`
+	Text    textParams `json:"text"`
+	At      atParams   `json:"at"`
+}
+
+type textParams struct {
+	Content string `json:"content"`
+}
+
+type atParams struct {
+	AtMobiles []string `json:"atMobiles,omitempty"`
+	IsAtAll   bool     `json:"isAtAll,omitempty"`
+}
+
+type linkMessage struct {
+	MsgType string     `json:"msgtype"`
+	Link    linkParams `json:"link"`
+}
+
+type linkParams struct {
+	Title      string `json:"title"`
+	Text       string `json:"text"`
+	MessageURL string `json:"messageUrl"`
+	PicURL     string `json:"picUrl,omitempty"`
+}
+
+type markdownMessage struct {
+	MsgType  string         `json:"msgtype"`
+	Markdown markdownParams `json:"markdown"`
+	At       atParams       `json:"at"`
+}
+
+type markdownParams struct {
+	Title string `json:"title"`
+	Text  string `json:"text"`
+}
+
+type actionCardMessage struct {
+	MsgType    string           `json:"msgtype"`
+	ActionCard actionCardParams `json:"actionCard"`
+}
+
+type actionCardParams struct {
+	Title          string `json:"title"`
+	Text           string `json:"text"`
+	SingleTitle    string `json:"singleTitle"`
+	SingleURL      string `json:"singleURL"`
+	BtnOrientation string `json:"btnOrientation,omitempty"`
+	HideAvatar     string `json:"hideAvatar,omitempty"`
+}
diff --git a/backend/vendor/modules.txt b/backend/vendor/modules.txt
index a73e7de6..dd2a41b3 100644
--- a/backend/vendor/modules.txt
+++ b/backend/vendor/modules.txt
@@ -82,6 +82,8 @@ github.com/olekukonko/tablewriter
 github.com/pelletier/go-toml
 # github.com/pkg/errors v0.8.1
 github.com/pkg/errors
+# github.com/royeo/dingrobot v1.0.0
+github.com/royeo/dingrobot
 # github.com/satori/go.uuid v1.2.0
 github.com/satori/go.uuid
 # github.com/shurcooL/sanitized_anchor_name v1.0.0

From 32bfb63d8f248d14de96198c2a347531e67a1270 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 13:03:43 +0800
Subject: [PATCH 13/42] =?UTF-8?q?=E5=AE=8C=E6=88=90=E9=92=89=E9=92=89?=
 =?UTF-8?q?=E6=9C=BA=E5=99=A8=E4=BA=BA=E6=B6=88=E6=81=AF=E9=80=9A=E7=9F=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/model/user.go                      |   5 +-
 backend/routes/user.go                     |  16 ++-
 backend/services/notification/ding_talk.go | 111 +++++++--------------
 backend/services/task.go                   |  90 ++++++++++++++---
 frontend/src/i18n/zh.js                    |   3 +-
 5 files changed, 124 insertions(+), 101 deletions(-)

diff --git a/backend/model/user.go b/backend/model/user.go
index 3e24e096..096f0ba3 100644
--- a/backend/model/user.go
+++ b/backend/model/user.go
@@ -24,9 +24,8 @@ type User struct {
 }
 
 type UserSetting struct {
-	NotificationTrigger string `json:"notification_trigger" bson:"notification_trigger"`
-	DingTalkAppKey      string `json:"ding_talk_app_key" bson:"ding_talk_app_key"`
-	DingTalkAppSecret   string `json:"ding_talk_app_secret" bson:"ding_talk_app_secret"`
+	NotificationTrigger  string `json:"notification_trigger" bson:"notification_trigger"`
+	DingTalkRobotWebhook string `json:"ding_talk_robot_webhook" bson:"ding_talk_robot_webhook"`
 }
 
 func (user *User) Save() error {
diff --git a/backend/routes/user.go b/backend/routes/user.go
index 9afddffc..95fd2951 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -213,11 +213,10 @@ func GetMe(c *gin.Context) {
 
 func PostMe(c *gin.Context) {
 	type ReqBody struct {
-		Email               string `json:"email"`
-		Password            string `json:"password"`
-		NotificationTrigger string `json:"notification_trigger"`
-		DingTalkAppKey      string `json:"ding_talk_app_key"`
-		DingTalkAppSecret   string `json:"ding_talk_app_secret"`
+		Email                string `json:"email"`
+		Password             string `json:"password"`
+		NotificationTrigger  string `json:"notification_trigger"`
+		DingTalkRobotWebhook string `json:"ding_talk_robot_webhook"`
 	}
 	ctx := context.WithGinContext(c)
 	user := ctx.User()
@@ -239,11 +238,8 @@ func PostMe(c *gin.Context) {
 	if reqBody.NotificationTrigger != "" {
 		user.Setting.NotificationTrigger = reqBody.NotificationTrigger
 	}
-	if reqBody.DingTalkAppKey != "" {
-		user.Setting.DingTalkAppKey = reqBody.DingTalkAppKey
-	}
-	if reqBody.DingTalkAppSecret != "" {
-		user.Setting.DingTalkAppSecret = reqBody.DingTalkAppSecret
+	if reqBody.DingTalkRobotWebhook != "" {
+		user.Setting.DingTalkRobotWebhook = reqBody.DingTalkRobotWebhook
 	}
 	if err := user.Save(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
diff --git a/backend/services/notification/ding_talk.go b/backend/services/notification/ding_talk.go
index ae1756aa..963334b8 100644
--- a/backend/services/notification/ding_talk.go
+++ b/backend/services/notification/ding_talk.go
@@ -1,95 +1,58 @@
 package notification
 
 import (
-	"crawlab/model"
 	"errors"
-	"fmt"
 	"github.com/apex/log"
 	"github.com/imroc/req"
-	"github.com/royeo/dingrobot"
 	"runtime/debug"
 )
 
-func SendDingTalkNotification(t model.Task, s model.Spider) error {
-	// 获取用户
-	user, _ := model.GetUser(t.UserId)
-
-	// 如果AppKey或AppSecret未设置，则返回错误
-	if user.Setting.DingTalkAppKey == "" || user.Setting.DingTalkAppSecret == "" {
-		return errors.New("ding_talk_app_key or ding_talk_app_secret is empty")
-	}
-
-	// 获取access_token
-	accessToken, err := GetDingTalkAccessToken(user)
-	if err != nil {
-		return err
-	}
-
-	// 时间戳
-	//timestamp := time.Now().Unix()
-
-	// 计算sign
-	//signRawString := fmt.Sprintf("%d\n%s", timestamp, user.Setting.DingTalkAppSecret)
-	//sign := utils.ComputeHmacSha256(signRawString, user.Setting.DingTalkAppSecret)
-
-	// 请求数据
-	url := fmt.Sprintf("https://oapi.dingtalk.com/robot/send?access_token=%s", accessToken)
-	robot := dingrobot.NewRobot(url)
-
-	text := "it works"
-	if err := robot.SendText(text, []string{}, false); err != nil {
-		log.Errorf(err.Error())
-		debug.PrintStack()
-		return err
-	}
-	//header := req.Header{
-	//	"Content-Type": "application/json; charset=utf-8",
-	//	"timestamp":    fmt.Sprintf("%d000", timestamp),
-	//	"sign":         sign,
-	//}
-	//data := req.Param{
-	//	"msgtype": "text",
-	//	"text": req.Param{
-	//		"text": "it works",
-	//	},
-	//	"at": req.Param{
-	//		"atMobiles": []string{},
-	//		"isAtAll": false,
-	//	},
-	//}
-	//res, err := req.Post(url, header, req.BodyJSON(&data))
-	//if err != nil {
-	//	log.Errorf("dingtalk notification error: " + err.Error())
-	//	debug.PrintStack()
-	//	return err
-	//}
-	//log.Infof(fmt.Sprintf("%+v", res))
-	return nil
-}
-
-func GetDingTalkAccessToken(u model.User) (string, error) {
+func SendDingTalkNotification(webhook string, title string, content string) error {
 	type ResBody struct {
-		ErrCode     int    `json:"errcode"`
-		ErrMsg      string `json:"errmsg"`
-		AccessToken string `json:"access_token"`
+		ErrCode int    `json:"errcode"`
+		ErrMsg  string `json:"errmsg"`
+	}
+
+	// 请求头
+	header := req.Header{
+		"Content-Type": "application/json; charset=utf-8",
 	}
 
 	// 请求数据
-	url := fmt.Sprintf("https://oapi.dingtalk.com/gettoken?appkey=%s&appsecret=%s", u.Setting.DingTalkAppKey, u.Setting.DingTalkAppSecret)
-	res, err := req.Get(url)
-	if err != nil {
-		log.Errorf("get dingtalk access_token error: " + err.Error())
-		debug.PrintStack()
-		return "", err
+	data := req.Param{
+		"msgtype": "markdown",
+		"markdown": req.Param{
+			"title": title,
+			"text":  content,
+		},
+		"at": req.Param{
+			"atMobiles": []string{},
+			"isAtAll":   false,
+		},
 	}
 
-	// 解析相应body
+	// 发起请求
+	res, err := req.Post(webhook, header, req.BodyJSON(&data))
+	if err != nil {
+		log.Errorf("dingtalk notification error: " + err.Error())
+		debug.PrintStack()
+		return err
+	}
+
+	// 解析响应
 	var resBody ResBody
 	if err := res.ToJSON(&resBody); err != nil {
-		log.Errorf("get dingtalk access_token error: " + err.Error())
+		log.Errorf("dingtalk notification error: " + err.Error())
 		debug.PrintStack()
-		return "", err
+		return err
 	}
 
-	return resBody.AccessToken, nil
+	// 判断响应是否报错
+	if resBody.ErrCode != 0 {
+		log.Errorf("dingtalk notification error: " + resBody.ErrMsg)
+		debug.PrintStack()
+		return errors.New(resBody.ErrMsg)
+	}
+
+	return nil
 }
diff --git a/backend/services/task.go b/backend/services/task.go
index 12c3580a..2e4d1e09 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -489,9 +489,14 @@ func ExecuteTask(id int) {
 
 		// 如果发生错误，则发送通知
 		t, _ = model.GetTask(t.Id)
-		if user.Email != "" &&
-			(user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd || user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskError) {
-			SendTaskEmail(user, t, spider)
+		if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd || user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskError {
+			if user.Email != "" {
+				SendTaskEmail(user, t, spider)
+			}
+
+			if user.Setting.DingTalkRobotWebhook != "" {
+				SendTaskDingTalk(user, t, spider)
+			}
 		}
 		return
 	}
@@ -516,12 +521,13 @@ func ExecuteTask(id int) {
 	t.TotalDuration = t.FinishTs.Sub(t.CreateTs).Seconds()  // 总时长
 
 	// 如果是任务结束时发送通知，则发送通知
-	if user.Email != "" &&
-		user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
-		SendTaskEmail(user, t, spider)
-		if err := notification.SendDingTalkNotification(t, spider); err != nil {
-			log.Errorf(err.Error())
-			debug.PrintStack()
+	if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
+		if user.Email != "" {
+			SendTaskEmail(user, t, spider)
+		}
+
+		if user.Setting.DingTalkRobotWebhook != "" {
+			SendTaskDingTalk(user, t, spider)
 		}
 	}
 
@@ -695,7 +701,7 @@ func HandleTaskError(t model.Task, err error) {
 	debug.PrintStack()
 }
 
-func GetTaskEmailContent(t model.Task, s model.Spider) string {
+func GetTaskEmailMarkdownContent(t model.Task, s model.Spider) string {
 	n, _ := model.GetNode(t.NodeId)
 	errMsg := ""
 	statusMsg := fmt.Sprintf(`<span style="color:green">%s</span>`, t.Status)
@@ -743,22 +749,82 @@ Please login to Crawlab to view the details.
 	)
 }
 
+func GetTaskDingTalkMarkdownContent(t model.Task, s model.Spider) string {
+	n, _ := model.GetNode(t.NodeId)
+	errMsg := ""
+	statusMsg := fmt.Sprintf(`<font color=green>%s</font>`, t.Status)
+	if t.Status == constants.StatusError {
+		errMsg = `<font color="red">(有错误)</font>`
+		statusMsg = fmt.Sprintf(`<font color=red>%s</font>`, t.Status)
+	}
+	return fmt.Sprintf(`
+您的任务已完成%s，请查看任务信息如下。
+
+> **任务ID:** %s  
+> **任务状态:** %s  
+> **任务参数:** %s  
+> **爬虫ID:** %s  
+> **爬虫名称:** %s  
+> **节点:** %s  
+> **创建时间:** %s  
+> **开始时间:** %s  
+> **完成时间:** %s  
+> **等待时间:** %.0f秒   
+> **运行时间:** %.0f秒  
+> **总时间:** %.0f秒  
+> **结果数:** %d  
+> **错误:** %s  
+
+请登录Crawlab查看详情。
+`,
+		errMsg,
+		t.Id,
+		statusMsg,
+		t.Param,
+		s.Id.Hex(),
+		s.Name,
+		n.Name,
+		utils.GetLocalTimeString(t.CreateTs),
+		utils.GetLocalTimeString(t.StartTs),
+		utils.GetLocalTimeString(t.FinishTs),
+		t.WaitDuration,
+		t.RuntimeDuration,
+		t.TotalDuration,
+		t.ResultCount,
+		t.Error,
+	)
+}
+
 func SendTaskEmail(u model.User, t model.Task, s model.Spider) {
 	statusMsg := "has finished"
 	if t.Status == constants.StatusError {
 		statusMsg = "has an error"
 	}
+	title := fmt.Sprintf("[Crawlab] Task for \"%s\" %s", s.Name, statusMsg)
 	if err := notification.SendMail(
 		u.Email,
 		u.Username,
-		fmt.Sprintf("[Crawlab] Task for \"%s\" %s", s.Name, statusMsg),
-		GetTaskEmailContent(t, s),
+		title,
+		GetTaskEmailMarkdownContent(t, s),
 	); err != nil {
 		log.Errorf("mail error: " + err.Error())
 		debug.PrintStack()
 	}
 }
 
+func SendTaskDingTalk(u model.User, t model.Task, s model.Spider) {
+	statusMsg := "has finished"
+	if t.Status == constants.StatusError {
+		statusMsg = "has an error"
+	}
+	title := fmt.Sprintf("[Crawlab] Task for \"%s\" %s", s.Name, statusMsg)
+	content := GetTaskDingTalkMarkdownContent(t, s)
+	if err := notification.SendDingTalkNotification(u.Setting.DingTalkRobotWebhook, title, content); err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+	}
+}
+
 func InitTaskExecutor() error {
 	c := cron.New(cron.WithSeconds())
 	Exec = &Executor{
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index 740408a2..a3ea2768 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -353,8 +353,7 @@ export default {
   'On Task End': '当任务结束',
   'On Task Error': '当任务发生错误',
   'Never': '从不',
-  'DingTalk AppKey': '钉钉 AppKey',
-  'DingTalk AppSecret': '钉钉 AppSecret',
+  'DingTalk Robot Webhook': '钉钉机器人 Webhook',
 
   // 其他
   tagsView: {

From 9c241262236ae94df2ab585a716a86376a2892e8 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 13:04:59 +0800
Subject: [PATCH 14/42] =?UTF-8?q?=E5=AE=8C=E6=88=90=E9=92=89=E9=92=89?=
 =?UTF-8?q?=E6=9C=BA=E5=99=A8=E4=BA=BA=E6=B6=88=E6=81=AF=E9=80=9A=E7=9F=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/services/task.go               |  6 ++--
 frontend/src/views/setting/Setting.vue | 42 +++++++++++---------------
 2 files changed, 21 insertions(+), 27 deletions(-)

diff --git a/backend/services/task.go b/backend/services/task.go
index 2e4d1e09..42a54bf4 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -813,11 +813,11 @@ func SendTaskEmail(u model.User, t model.Task, s model.Spider) {
 }
 
 func SendTaskDingTalk(u model.User, t model.Task, s model.Spider) {
-	statusMsg := "has finished"
+	statusMsg := "已完成"
 	if t.Status == constants.StatusError {
-		statusMsg = "has an error"
+		statusMsg = "发生错误"
 	}
-	title := fmt.Sprintf("[Crawlab] Task for \"%s\" %s", s.Name, statusMsg)
+	title := fmt.Sprintf("[Crawlab] \"%s\" 任务%s", s.Name, statusMsg)
 	content := GetTaskDingTalkMarkdownContent(t, s)
 	if err := notification.SendDingTalkNotification(u.Setting.DingTalkRobotWebhook, title, content); err != nil {
 		log.Errorf(err.Error())
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
index 4ca476a5..33a134a6 100644
--- a/frontend/src/views/setting/Setting.vue
+++ b/frontend/src/views/setting/Setting.vue
@@ -1,6 +1,6 @@
 <template>
   <div class="app-container">
-    <el-form :model="userInfo" class="setting-form" ref="setting-form" label-width="150px" :rules="rules"
+    <el-form :model="userInfo" class="setting-form" ref="setting-form" label-width="200px" :rules="rules"
              inline-message>
       <el-form-item prop="username" :label="$t('Username')">
         <el-input v-model="userInfo.username" disabled></el-input>
@@ -24,26 +24,8 @@
           </el-radio>
         </el-radio-group>
       </el-form-item>
-      <el-form-item prop="ding_talk_app_key" :label="$t('DingTalk AppKey')">
-        <el-input v-model="userInfo.setting.ding_talk_app_key" :placeholder="$t('DingTalk AppKey')"></el-input>
-      </el-form-item>
-      <el-form-item prop="ding_talk_app_secret" :label="$t('DingTalk AppSecret')">
-        <template v-if="isShowDingTalkAppSecret">
-          <el-input
-            v-model="userInfo.setting.ding_talk_app_secret"
-            :placeholder="$t('DingTalk AppSecret')"
-          />
-
-          <i class="icon el-icon-view" @click="toggleDingTalkAppSecret"></i>
-        </template>
-        <template v-else>
-          <el-input
-            v-model="userInfo.setting.ding_talk_app_secret"
-            type="password"
-            :placeholder="$t('DingTalk AppSecret')"
-          />
-          <i class="icon el-icon-view" @click="toggleDingTalkAppSecret"></i>
-        </template>
+      <el-form-item prop="setting.ding_talk_robot_webhook" :label="$t('DingTalk Robot Webhook')">
+        <el-input v-model="userInfo.setting.ding_talk_robot_webhook" :placeholder="$t('DingTalk Robot Webhook')"></el-input>
       </el-form-item>
       <el-form-item>
         <div class="buttons">
@@ -74,11 +56,20 @@ export default {
         callback()
       }
     }
+    const validateDingTalkRobotWebhook = (rule, value, callback) => {
+      if (!value) return callback()
+      if (!value.match(/^https:\/\/oapi.dingtalk.com\/robot\/send\?access_token=[a-f0-9]+/i)) {
+        callback(new Error(this.$t('DingTalk Robot Webhook format invalid')))
+      } else {
+        callback()
+      }
+    }
     return {
       userInfo: { setting: {} },
       rules: {
         password: [{ trigger: 'blur', validator: validatePass }],
-        email: [{ trigger: 'blur', validator: validateEmail }]
+        email: [{ trigger: 'blur', validator: validateEmail }],
+        'setting.ding_talk_robot_webhook': [{ trigger: 'blur', validator: validateDingTalkRobotWebhook }]
       },
       isShowDingTalkAppSecret: false
     }
@@ -97,8 +88,7 @@ export default {
           password: this.userInfo.password,
           email: this.userInfo.email,
           notification_trigger: this.userInfo.setting.notification_trigger,
-          ding_talk_app_key: this.userInfo.setting.ding_talk_app_key,
-          ding_talk_app_secret: this.userInfo.setting.ding_talk_app_secret
+          ding_talk_robot_webhook: this.userInfo.setting.ding_talk_robot_webhook
         })
         if (!res || res.error) {
           this.$message.error(res.error)
@@ -133,4 +123,8 @@ export default {
     position: absolute;
     color: #DCDFE6;
   }
+
+  .setting-form >>> .el-form-item__label {
+    height: 40px;
+  }
 </style>

From 00c9312f83b1bbcdec3a683da9065d6631a2ea04 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 13:14:56 +0800
Subject: [PATCH 15/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 3 +++
 CHANGELOG.md    | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index b9b8cc09..51062ab5 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -1,11 +1,14 @@
 # 0.4.4 (未知)
 
 ### 功能 / 优化
+- **邮件通知**. 允许用户发送邮件消息通知.
+- **钉钉机器人通知**. 允许用户发送钉钉机器人消息通知.
 - **API 地址优化**. 在前端加入相对路径，因此用户不需要特别注明 `CRAWLAB_API_ADDRESS`.
 - **SDK 兼容**. 允许用户通过 Crawlab SDK 与 Scrapy 或通用爬虫集成.
 
 ### Bug 修复
 - **`nil retuened` 错误**.
+- **使用 HTTPS 出现的报错**.
 
 # 0.4.3 (2020-01-07)
 
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 8e0a5ef1..01ccb3e8 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,10 +1,13 @@
 # 0.4.4 (unknown)
 ### Features / Enhancement
+- **Email Notification**. Allow users to send email notifications.
+- **DingTalk Robot Notification**. Allow users to send DingTalk Robot notifications.
 - **API Address Optimization**. Added relative URL path in frontend so that users don't have to specify `CRAWLAB_API_ADDRESS` explicitly.
 - **SDK Compatiblity**. Allow users to integrate Scrapy or general spiders with Crawlab SDK.
 
 ### Bug Fixes
 - **`nil retuened` error**.
+- **Error when using HTTPS**.
 
 # 0.4.3 (2020-01-07)
 

From 1fde42b932c1cf01b3c622b7bb3692ee40a3341a Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 13:51:12 +0800
Subject: [PATCH 16/42] =?UTF-8?q?=E5=B0=86=E5=8F=91=E9=80=81=E6=B6=88?=
 =?UTF-8?q?=E6=81=AF=E9=80=9A=E7=9F=A5=E8=B0=83=E6=95=B4=E4=B8=BA=E5=BC=82?=
 =?UTF-8?q?=E6=AD=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/model/user.go                         |  1 +
 backend/routes/user.go                        | 19 +++---
 .../notification/{ding_talk.go => mobile.go}  |  7 ++-
 backend/services/task.go                      | 58 +++++++++++++++----
 4 files changed, 59 insertions(+), 26 deletions(-)
 rename backend/services/notification/{ding_talk.go => mobile.go} (87%)

diff --git a/backend/model/user.go b/backend/model/user.go
index 096f0ba3..e37783c9 100644
--- a/backend/model/user.go
+++ b/backend/model/user.go
@@ -26,6 +26,7 @@ type User struct {
 type UserSetting struct {
 	NotificationTrigger  string `json:"notification_trigger" bson:"notification_trigger"`
 	DingTalkRobotWebhook string `json:"ding_talk_robot_webhook" bson:"ding_talk_robot_webhook"`
+	WechatRobotWebhook   string `json:"wechat_robot_webhook" bson:"wechat_robot_webhook"`
 }
 
 func (user *User) Save() error {
diff --git a/backend/routes/user.go b/backend/routes/user.go
index 95fd2951..9408024a 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -212,19 +212,13 @@ func GetMe(c *gin.Context) {
 }
 
 func PostMe(c *gin.Context) {
-	type ReqBody struct {
-		Email                string `json:"email"`
-		Password             string `json:"password"`
-		NotificationTrigger  string `json:"notification_trigger"`
-		DingTalkRobotWebhook string `json:"ding_talk_robot_webhook"`
-	}
 	ctx := context.WithGinContext(c)
 	user := ctx.User()
 	if user == nil {
 		ctx.FailedWithError(constants.ErrorUserNotFound, http.StatusUnauthorized)
 		return
 	}
-	var reqBody ReqBody
+	var reqBody model.User
 	if err := c.ShouldBindJSON(&reqBody); err != nil {
 		HandleErrorF(http.StatusBadRequest, c, "invalid request")
 		return
@@ -235,11 +229,14 @@ func PostMe(c *gin.Context) {
 	if reqBody.Password != "" {
 		user.Password = utils.EncryptPassword(reqBody.Password)
 	}
-	if reqBody.NotificationTrigger != "" {
-		user.Setting.NotificationTrigger = reqBody.NotificationTrigger
+	if reqBody.Setting.NotificationTrigger != "" {
+		user.Setting.NotificationTrigger = reqBody.Setting.NotificationTrigger
 	}
-	if reqBody.DingTalkRobotWebhook != "" {
-		user.Setting.DingTalkRobotWebhook = reqBody.DingTalkRobotWebhook
+	if reqBody.Setting.DingTalkRobotWebhook != "" {
+		user.Setting.DingTalkRobotWebhook = reqBody.Setting.DingTalkRobotWebhook
+	}
+	if reqBody.Setting.WechatRobotWebhook != "" {
+		user.Setting.WechatRobotWebhook = reqBody.Setting.WechatRobotWebhook
 	}
 	if err := user.Save(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
diff --git a/backend/services/notification/ding_talk.go b/backend/services/notification/mobile.go
similarity index 87%
rename from backend/services/notification/ding_talk.go
rename to backend/services/notification/mobile.go
index 963334b8..e140ecc5 100644
--- a/backend/services/notification/ding_talk.go
+++ b/backend/services/notification/mobile.go
@@ -7,7 +7,7 @@ import (
 	"runtime/debug"
 )
 
-func SendDingTalkNotification(webhook string, title string, content string) error {
+func SendMobileNotification(webhook string, title string, content string) error {
 	type ResBody struct {
 		ErrCode int    `json:"errcode"`
 		ErrMsg  string `json:"errmsg"`
@@ -22,8 +22,9 @@ func SendDingTalkNotification(webhook string, title string, content string) erro
 	data := req.Param{
 		"msgtype": "markdown",
 		"markdown": req.Param{
-			"title": title,
-			"text":  content,
+			"title":   title,
+			"text":    content,
+			"content": content,
 		},
 		"at": req.Param{
 			"atMobiles": []string{},
diff --git a/backend/services/task.go b/backend/services/task.go
index 42a54bf4..507fb8e6 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -491,11 +491,21 @@ func ExecuteTask(id int) {
 		t, _ = model.GetTask(t.Id)
 		if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd || user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskError {
 			if user.Email != "" {
-				SendTaskEmail(user, t, spider)
+				go func() {
+					SendTaskEmail(user, t, spider)
+				}()
 			}
 
 			if user.Setting.DingTalkRobotWebhook != "" {
-				SendTaskDingTalk(user, t, spider)
+				go func() {
+					SendTaskDingTalk(user, t, spider)
+				}()
+			}
+
+			if user.Setting.WechatRobotWebhook != "" {
+				go func() {
+					SendTaskWechat(user, t, spider)
+				}()
 			}
 		}
 		return
@@ -523,11 +533,25 @@ func ExecuteTask(id int) {
 	// 如果是任务结束时发送通知，则发送通知
 	if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
 		if user.Email != "" {
-			SendTaskEmail(user, t, spider)
+			go func() {
+				SendTaskEmail(user, t, spider)
+			}()
 		}
 
-		if user.Setting.DingTalkRobotWebhook != "" {
-			SendTaskDingTalk(user, t, spider)
+		if user.Email != "" {
+			go func() {
+				if user.Setting.DingTalkRobotWebhook != "" {
+					SendTaskDingTalk(user, t, spider)
+				}
+			}()
+		}
+
+		if user.Email != "" {
+			go func() {
+				if user.Setting.WechatRobotWebhook != "" {
+					SendTaskWechat(user, t, spider)
+				}
+			}()
 		}
 	}
 
@@ -749,13 +773,15 @@ Please login to Crawlab to view the details.
 	)
 }
 
-func GetTaskDingTalkMarkdownContent(t model.Task, s model.Spider) string {
+func GetTaskMarkdownContent(t model.Task, s model.Spider) string {
 	n, _ := model.GetNode(t.NodeId)
 	errMsg := ""
-	statusMsg := fmt.Sprintf(`<font color=green>%s</font>`, t.Status)
+	errLog := "-"
+	statusMsg := fmt.Sprintf(`<font color="#00FF00">%s</font>`, t.Status)
 	if t.Status == constants.StatusError {
-		errMsg = `<font color="red">(有错误)</font>`
-		statusMsg = fmt.Sprintf(`<font color=red>%s</font>`, t.Status)
+		errMsg = `<font color="#FF0000">（有错误）</font>`
+		errLog = fmt.Sprintf(`<font color="#FF0000">%s</font>`, t.Error)
+		statusMsg = fmt.Sprintf(`<font color="#FF0000">%s</font>`, t.Status)
 	}
 	return fmt.Sprintf(`
 您的任务已完成%s，请查看任务信息如下。
@@ -791,7 +817,7 @@ func GetTaskDingTalkMarkdownContent(t model.Task, s model.Spider) string {
 		t.RuntimeDuration,
 		t.TotalDuration,
 		t.ResultCount,
-		t.Error,
+		errLog,
 	)
 }
 
@@ -818,8 +844,16 @@ func SendTaskDingTalk(u model.User, t model.Task, s model.Spider) {
 		statusMsg = "发生错误"
 	}
 	title := fmt.Sprintf("[Crawlab] \"%s\" 任务%s", s.Name, statusMsg)
-	content := GetTaskDingTalkMarkdownContent(t, s)
-	if err := notification.SendDingTalkNotification(u.Setting.DingTalkRobotWebhook, title, content); err != nil {
+	content := GetTaskMarkdownContent(t, s)
+	if err := notification.SendMobileNotification(u.Setting.DingTalkRobotWebhook, title, content); err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+	}
+}
+
+func SendTaskWechat(u model.User, t model.Task, s model.Spider) {
+	content := GetTaskMarkdownContent(t, s)
+	if err := notification.SendMobileNotification(u.Setting.WechatRobotWebhook, "", content); err != nil {
 		log.Errorf(err.Error())
 		debug.PrintStack()
 	}

From b70949cc87b4b01e47b320f29a7765f081bf5a57 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 14:39:51 +0800
Subject: [PATCH 17/42] =?UTF-8?q?=E5=8A=A0=E5=85=A5=E6=B6=88=E6=81=AF?=
 =?UTF-8?q?=E9=80=9A=E7=9F=A5=E7=B1=BB=E5=9E=8B=E9=80=89=E6=8B=A9?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/constants/notification.go      |  6 +++
 backend/model/user.go                  |  7 +--
 backend/routes/user.go                 | 12 +----
 backend/services/task.go               | 62 ++++++++++----------------
 backend/services/user.go               | 33 ++++++++++----
 backend/utils/array.go                 | 10 +++++
 frontend/src/i18n/zh.js                |  3 +-
 frontend/src/views/setting/Setting.vue | 43 ++++++++++++------
 8 files changed, 101 insertions(+), 75 deletions(-)
 create mode 100644 backend/utils/array.go

diff --git a/backend/constants/notification.go b/backend/constants/notification.go
index cad3b19b..cf3da062 100644
--- a/backend/constants/notification.go
+++ b/backend/constants/notification.go
@@ -5,3 +5,9 @@ const (
 	NotificationTriggerOnTaskError = "notification_trigger_on_task_error"
 	NotificationTriggerNever       = "notification_trigger_never"
 )
+
+const (
+	NotificationTypeMail     = "notification_type_mail"
+	NotificationTypeDingTalk = "notification_type_ding_talk"
+	NotificationTypeWechat   = "notification_type_wechat"
+)
diff --git a/backend/model/user.go b/backend/model/user.go
index e37783c9..9dadec0f 100644
--- a/backend/model/user.go
+++ b/backend/model/user.go
@@ -24,9 +24,10 @@ type User struct {
 }
 
 type UserSetting struct {
-	NotificationTrigger  string `json:"notification_trigger" bson:"notification_trigger"`
-	DingTalkRobotWebhook string `json:"ding_talk_robot_webhook" bson:"ding_talk_robot_webhook"`
-	WechatRobotWebhook   string `json:"wechat_robot_webhook" bson:"wechat_robot_webhook"`
+	NotificationTrigger  string   `json:"notification_trigger" bson:"notification_trigger"`
+	DingTalkRobotWebhook string   `json:"ding_talk_robot_webhook" bson:"ding_talk_robot_webhook"`
+	WechatRobotWebhook   string   `json:"wechat_robot_webhook" bson:"wechat_robot_webhook"`
+	EnabledNotifications []string `json:"enabled_notifications" bson:"enabled_notifications"`
 }
 
 func (user *User) Save() error {
diff --git a/backend/routes/user.go b/backend/routes/user.go
index 9408024a..fcca967e 100644
--- a/backend/routes/user.go
+++ b/backend/routes/user.go
@@ -96,16 +96,7 @@ func PutUser(c *gin.Context) {
 	}
 
 	// 添加用户
-	user := model.User{
-		Username: strings.ToLower(reqData.Username),
-		Password: utils.EncryptPassword(reqData.Password),
-		Role:     reqData.Role,
-		Email:    reqData.Email,
-		Setting: model.UserSetting{
-			NotificationTrigger: constants.NotificationTriggerNever,
-		},
-	}
-	if err := user.Add(); err != nil {
+	if err := services.CreateNewUser(reqData.Username, reqData.Password, reqData.Role, reqData.Email); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
 		return
 	}
@@ -238,6 +229,7 @@ func PostMe(c *gin.Context) {
 	if reqBody.Setting.WechatRobotWebhook != "" {
 		user.Setting.WechatRobotWebhook = reqBody.Setting.WechatRobotWebhook
 	}
+	user.Setting.EnabledNotifications = reqBody.Setting.EnabledNotifications
 	if err := user.Save(); err != nil {
 		HandleError(http.StatusInternalServerError, c, err)
 		return
diff --git a/backend/services/task.go b/backend/services/task.go
index 507fb8e6..da4eabc8 100644
--- a/backend/services/task.go
+++ b/backend/services/task.go
@@ -490,23 +490,7 @@ func ExecuteTask(id int) {
 		// 如果发生错误，则发送通知
 		t, _ = model.GetTask(t.Id)
 		if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd || user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskError {
-			if user.Email != "" {
-				go func() {
-					SendTaskEmail(user, t, spider)
-				}()
-			}
-
-			if user.Setting.DingTalkRobotWebhook != "" {
-				go func() {
-					SendTaskDingTalk(user, t, spider)
-				}()
-			}
-
-			if user.Setting.WechatRobotWebhook != "" {
-				go func() {
-					SendTaskWechat(user, t, spider)
-				}()
-			}
+			SendNotifications(user, t, spider)
 		}
 		return
 	}
@@ -532,27 +516,7 @@ func ExecuteTask(id int) {
 
 	// 如果是任务结束时发送通知，则发送通知
 	if user.Setting.NotificationTrigger == constants.NotificationTriggerOnTaskEnd {
-		if user.Email != "" {
-			go func() {
-				SendTaskEmail(user, t, spider)
-			}()
-		}
-
-		if user.Email != "" {
-			go func() {
-				if user.Setting.DingTalkRobotWebhook != "" {
-					SendTaskDingTalk(user, t, spider)
-				}
-			}()
-		}
-
-		if user.Email != "" {
-			go func() {
-				if user.Setting.WechatRobotWebhook != "" {
-					SendTaskWechat(user, t, spider)
-				}
-			}()
-		}
+		SendNotifications(user, t, spider)
 	}
 
 	// 保存任务
@@ -779,7 +743,7 @@ func GetTaskMarkdownContent(t model.Task, s model.Spider) string {
 	errLog := "-"
 	statusMsg := fmt.Sprintf(`<font color="#00FF00">%s</font>`, t.Status)
 	if t.Status == constants.StatusError {
-		errMsg = `<font color="#FF0000">（有错误）</font>`
+		errMsg = `（有错误）`
 		errLog = fmt.Sprintf(`<font color="#FF0000">%s</font>`, t.Error)
 		statusMsg = fmt.Sprintf(`<font color="#FF0000">%s</font>`, t.Status)
 	}
@@ -859,6 +823,26 @@ func SendTaskWechat(u model.User, t model.Task, s model.Spider) {
 	}
 }
 
+func SendNotifications(u model.User, t model.Task, s model.Spider) {
+	if u.Email != "" && utils.StringArrayContains(u.Setting.EnabledNotifications, constants.NotificationTypeMail) {
+		go func() {
+			SendTaskEmail(u, t, s)
+		}()
+	}
+
+	if u.Setting.DingTalkRobotWebhook != "" && utils.StringArrayContains(u.Setting.EnabledNotifications, constants.NotificationTypeDingTalk) {
+		go func() {
+			SendTaskDingTalk(u, t, s)
+		}()
+	}
+
+	if u.Setting.WechatRobotWebhook != "" && utils.StringArrayContains(u.Setting.EnabledNotifications, constants.NotificationTypeWechat) {
+		go func() {
+			SendTaskWechat(u, t, s)
+		}()
+	}
+}
+
 func InitTaskExecutor() error {
 	c := cron.New(cron.WithSeconds())
 	Exec = &Executor{
diff --git a/backend/services/user.go b/backend/services/user.go
index 483a7408..a01e721b 100644
--- a/backend/services/user.go
+++ b/backend/services/user.go
@@ -9,21 +9,15 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/globalsign/mgo/bson"
 	"github.com/spf13/viper"
+	"strings"
 	"time"
 )
 
 func InitUserService() error {
-	adminUser := model.User{
-		Username: "admin",
-		Password: utils.EncryptPassword("admin"),
-		Role:     constants.RoleAdmin,
-		Setting: model.UserSetting{
-			NotificationTrigger: constants.NotificationTriggerNever,
-		},
-	}
-	_ = adminUser.Add()
+	_ = CreateNewUser("admin", "admin", constants.RoleAdmin, "")
 	return nil
 }
+
 func MakeToken(user *model.User) (tokenStr string, err error) {
 	token := jwt.NewWithClaims(jwt.SigningMethodHS256, jwt.MapClaims{
 		"id":       user.Id,
@@ -96,6 +90,27 @@ func CheckToken(tokenStr string) (user model.User, err error) {
 	return
 }
 
+func CreateNewUser(username string, password string, role string, email string) error {
+	user := model.User{
+		Username: strings.ToLower(username),
+		Password: utils.EncryptPassword(password),
+		Role:     role,
+		Email:    email,
+		Setting: model.UserSetting{
+			NotificationTrigger: constants.NotificationTriggerNever,
+			EnabledNotifications: []string{
+				constants.NotificationTypeMail,
+				constants.NotificationTypeDingTalk,
+				constants.NotificationTypeWechat,
+			},
+		},
+	}
+	if err := user.Add(); err != nil {
+		return err
+	}
+	return nil
+}
+
 func GetCurrentUser(c *gin.Context) *model.User {
 	data, _ := c.Get("currentUser")
 	return data.(*model.User)
diff --git a/backend/utils/array.go b/backend/utils/array.go
new file mode 100644
index 00000000..889430ed
--- /dev/null
+++ b/backend/utils/array.go
@@ -0,0 +1,10 @@
+package utils
+
+func StringArrayContains(arr []string, str string) bool {
+	for _, s := range arr {
+		if s == str {
+			return true
+		}
+	}
+	return false
+}
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index a3ea2768..136dbf3f 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -349,11 +349,12 @@ export default {
   'Optional': '可选',
 
   // 设置
-  'Notification Trigger': '通知触发',
+  'Notification Trigger Timing': '消息通知触发时机',
   'On Task End': '当任务结束',
   'On Task Error': '当任务发生错误',
   'Never': '从不',
   'DingTalk Robot Webhook': '钉钉机器人 Webhook',
+  'Wechat Robot Webhook': '微信机器人 Webhook',
 
   // 其他
   tagsView: {
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
index 33a134a6..75dc856f 100644
--- a/frontend/src/views/setting/Setting.vue
+++ b/frontend/src/views/setting/Setting.vue
@@ -8,10 +8,8 @@
       <el-form-item prop="password" :label="$t('Password')">
         <el-input v-model="userInfo.password" type="password" :placeholder="$t('Password')"></el-input>
       </el-form-item>
-      <el-form-item prop="email" :label="$t('Email')">
-        <el-input v-model="userInfo.email" :placeholder="$t('Email')"></el-input>
-      </el-form-item>
-      <el-form-item :label="$t('Notification Trigger')">
+      <div style="border-bottom: 1px solid #DCDFE6"></div>
+      <el-form-item :label="$t('Notification Trigger Timing')">
         <el-radio-group v-model="userInfo.setting.notification_trigger">
           <el-radio label="notification_trigger_on_task_end">
             {{$t('On Task End')}}
@@ -24,8 +22,22 @@
           </el-radio>
         </el-radio-group>
       </el-form-item>
+      <el-form-item prop="enabledNotifications" :label="$t('消息通知方式')">
+        <el-checkbox-group v-model="userInfo.setting.enabled_notifications">
+          <el-checkbox label="notification_type_mail">{{$t('邮件')}}</el-checkbox>
+          <el-checkbox label="notification_type_ding_talk">{{$t('钉钉')}}</el-checkbox>
+          <el-checkbox label="notification_type_wechat">{{$t('企业微信')}}</el-checkbox>
+        </el-checkbox-group>
+      </el-form-item>
+      <el-form-item prop="email" :label="$t('Email')">
+        <el-input v-model="userInfo.email" :placeholder="$t('Email')"></el-input>
+      </el-form-item>
       <el-form-item prop="setting.ding_talk_robot_webhook" :label="$t('DingTalk Robot Webhook')">
-        <el-input v-model="userInfo.setting.ding_talk_robot_webhook" :placeholder="$t('DingTalk Robot Webhook')"></el-input>
+        <el-input v-model="userInfo.setting.ding_talk_robot_webhook"
+                  :placeholder="$t('DingTalk Robot Webhook')"></el-input>
+      </el-form-item>
+      <el-form-item prop="setting.wechat_robot_webhook" :label="$t('Wechat Robot Webhook')">
+        <el-input v-model="userInfo.setting.wechat_robot_webhook" :placeholder="$t('Wechat Robot Webhook')"></el-input>
       </el-form-item>
       <el-form-item>
         <div class="buttons">
@@ -64,12 +76,21 @@ export default {
         callback()
       }
     }
+    const validateWechatRobotWebhook = (rule, value, callback) => {
+      if (!value) return callback()
+      if (!value.match(/^https:\/\/qyapi.weixin.qq.com\/cgi-bin\/webhook\/send\?key=.+/i)) {
+        callback(new Error(this.$t('DingTalk Robot Webhook format invalid')))
+      } else {
+        callback()
+      }
+    }
     return {
-      userInfo: { setting: {} },
+      userInfo: { setting: { enabled_notifications: [] } },
       rules: {
         password: [{ trigger: 'blur', validator: validatePass }],
         email: [{ trigger: 'blur', validator: validateEmail }],
-        'setting.ding_talk_robot_webhook': [{ trigger: 'blur', validator: validateDingTalkRobotWebhook }]
+        'setting.ding_talk_robot_webhook': [{ trigger: 'blur', validator: validateDingTalkRobotWebhook }],
+        'setting.wechat_robot_webhook': [{ trigger: 'blur', validator: validateWechatRobotWebhook }]
       },
       isShowDingTalkAppSecret: false
     }
@@ -80,16 +101,12 @@ export default {
       if (!data) return {}
       this.userInfo = JSON.parse(data)
       if (!this.userInfo.setting) this.userInfo.setting = {}
+      if (!this.userInfo.setting.enabled_notifications) this.userInfo.setting.enabled_notifications = []
     },
     saveUserInfo () {
       this.$refs['setting-form'].validate(async valid => {
         if (!valid) return
-        const res = await this.$store.dispatch('user/postInfo', {
-          password: this.userInfo.password,
-          email: this.userInfo.email,
-          notification_trigger: this.userInfo.setting.notification_trigger,
-          ding_talk_robot_webhook: this.userInfo.setting.ding_talk_robot_webhook
-        })
+        const res = await this.$store.dispatch('user/postInfo', this.userInfo)
         if (!res || res.error) {
           this.$message.error(res.error)
         } else {

From 10bbdef1a2a953758f5de60be9b0d61dbed93f7b Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 14:40:05 +0800
Subject: [PATCH 18/42] =?UTF-8?q?=E5=8A=A0=E5=85=A5=E6=B6=88=E6=81=AF?=
 =?UTF-8?q?=E9=80=9A=E7=9F=A5=E7=B1=BB=E5=88=AB=E9=80=89=E6=8B=A9?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/router/index.js | 1 +
 1 file changed, 1 insertion(+)

diff --git a/frontend/src/router/index.js b/frontend/src/router/index.js
index c019836a..666792f1 100644
--- a/frontend/src/router/index.js
+++ b/frontend/src/router/index.js
@@ -220,6 +220,7 @@ export const constantRouterMap = [
     children: [
       {
         path: '',
+        name: 'Setting',
         component: () => import('../views/setting/Setting'),
         meta: {
           title: 'Setting',

From 7a93a5cb99b28af60e6863bfd4f0738f2e8c1d94 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 14:43:08 +0800
Subject: [PATCH 19/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 1 +
 CHANGELOG.md    | 1 +
 2 files changed, 2 insertions(+)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index 51062ab5..5fa8dc0e 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -3,6 +3,7 @@
 ### 功能 / 优化
 - **邮件通知**. 允许用户发送邮件消息通知.
 - **钉钉机器人通知**. 允许用户发送钉钉机器人消息通知.
+- **企业微信机器人通知**. 允许用户发送企业微信机器人消息通知.
 - **API 地址优化**. 在前端加入相对路径，因此用户不需要特别注明 `CRAWLAB_API_ADDRESS`.
 - **SDK 兼容**. 允许用户通过 Crawlab SDK 与 Scrapy 或通用爬虫集成.
 
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 01ccb3e8..ba842b9a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,7 @@
 ### Features / Enhancement
 - **Email Notification**. Allow users to send email notifications.
 - **DingTalk Robot Notification**. Allow users to send DingTalk Robot notifications.
+- **Wechat Robot Notification**. Allow users to send Wechat Robot notifications.
 - **API Address Optimization**. Added relative URL path in frontend so that users don't have to specify `CRAWLAB_API_ADDRESS` explicitly.
 - **SDK Compatiblity**. Allow users to integrate Scrapy or general spiders with Crawlab SDK.
 

From ef2a4b35318953b759c6ffbf957e09b819634383 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 17:44:11 +0800
Subject: [PATCH 20/42] =?UTF-8?q?=E5=8A=A0=E5=85=A5=E6=A0=91=E5=BD=A2?=
 =?UTF-8?q?=E7=BB=93=E6=9E=84=E6=96=87=E4=BB=B6=E7=9B=AE=E5=BD=95?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/main.go                               |   3 +
 backend/model/file.go                         |  10 +-
 backend/routes/spider.go                      |  34 +-
 backend/services/file.go                      |  60 +++
 frontend/src/components/File/FileList.vue     | 357 ++++++++++++------
 frontend/src/store/modules/file.js            |   3 +-
 frontend/src/store/modules/spider.js          |  13 +-
 .../src/views/layout/components/AppMain.vue   |   2 +-
 8 files changed, 362 insertions(+), 120 deletions(-)
 create mode 100644 backend/services/file.go

diff --git a/backend/main.go b/backend/main.go
index 2b052171..20d70b22 100644
--- a/backend/main.go
+++ b/backend/main.go
@@ -160,6 +160,7 @@ func main() {
 			authGroup.POST("/spiders/:id/upload", routes.UploadSpiderFromId)    // 上传爬虫（ID）
 			authGroup.DELETE("/spiders/:id", routes.DeleteSpider)               // 删除爬虫
 			authGroup.GET("/spiders/:id/tasks", routes.GetSpiderTasks)          // 爬虫任务列表
+			authGroup.GET("/spiders/:id/file/tree", routes.GetSpiderFileTree)   // 爬虫文件目录树读取
 			authGroup.GET("/spiders/:id/file", routes.GetSpiderFile)            // 爬虫文件读取
 			authGroup.POST("/spiders/:id/file", routes.PostSpiderFile)          // 爬虫文件更改
 			authGroup.PUT("/spiders/:id/file", routes.PutSpiderFile)            // 爬虫文件创建
@@ -210,6 +211,8 @@ func main() {
 			// 系统
 			authGroup.GET("/system/deps/:lang", routes.GetAllDepList)             // 节点所有第三方依赖列表
 			authGroup.GET("/system/deps/:lang/:dep_name/json", routes.GetDepJson) // 节点第三方依赖JSON
+			// 文件
+			authGroup.GET("/file", routes.GetFile) // 获取文件
 		}
 
 	}
diff --git a/backend/model/file.go b/backend/model/file.go
index fe3ece0e..1af3b851 100644
--- a/backend/model/file.go
+++ b/backend/model/file.go
@@ -20,10 +20,12 @@ type GridFs struct {
 }
 
 type File struct {
-	Name  string `json:"name"`
-	Path  string `json:"path"`
-	IsDir bool   `json:"is_dir"`
-	Size  int64  `json:"size"`
+	Name     string `json:"name"`
+	Path     string `json:"path"`
+	IsDir    bool   `json:"is_dir"`
+	Size     int64  `json:"size"`
+	Children []File `json:"children"`
+	Label    string `json:"label"`
 }
 
 func (f *GridFs) Remove() {
diff --git a/backend/routes/spider.go b/backend/routes/spider.go
index 91bab47a..2534a7da 100644
--- a/backend/routes/spider.go
+++ b/backend/routes/spider.go
@@ -440,6 +440,8 @@ func GetSpiderTasks(c *gin.Context) {
 	})
 }
 
+// 爬虫文件管理
+
 func GetSpiderDir(c *gin.Context) {
 	// 爬虫ID
 	id := c.Param("id")
@@ -481,8 +483,6 @@ func GetSpiderDir(c *gin.Context) {
 	})
 }
 
-// 爬虫文件管理
-
 type SpiderFileReqBody struct {
 	Path    string `json:"path"`
 	Content string `json:"content"`
@@ -517,6 +517,36 @@ func GetSpiderFile(c *gin.Context) {
 	})
 }
 
+func GetSpiderFileTree(c *gin.Context) {
+	// 爬虫ID
+	id := c.Param("id")
+
+	// 获取爬虫
+	spider, err := model.GetSpider(bson.ObjectIdHex(id))
+	if err != nil {
+		HandleError(http.StatusInternalServerError, c, err)
+		return
+	}
+
+	// 获取目录下文件列表
+	spiderPath := viper.GetString("spider.path")
+	spiderFilePath := filepath.Join(spiderPath, spider.Name)
+
+	// 获取文件目录树
+	fileNodeTree, err := services.GetFileNodeTree(spiderFilePath, 0)
+	if err != nil {
+		HandleError(http.StatusInternalServerError, c, err)
+		return
+	}
+
+	// 返回结果
+	c.JSON(http.StatusOK, Response{
+		Status:  "ok",
+		Message: "success",
+		Data:    fileNodeTree,
+	})
+}
+
 func PostSpiderFile(c *gin.Context) {
 	// 爬虫ID
 	id := c.Param("id")
diff --git a/backend/services/file.go b/backend/services/file.go
new file mode 100644
index 00000000..0b810209
--- /dev/null
+++ b/backend/services/file.go
@@ -0,0 +1,60 @@
+package services
+
+import (
+	"crawlab/model"
+	"github.com/apex/log"
+	"os"
+	"path"
+	"runtime/debug"
+)
+
+func GetFileNodeTree(dstPath string, level int) (f model.File, err error) {
+	dstF, err := os.Open(dstPath)
+	if err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+		return f, err
+	}
+	defer dstF.Close()
+	fileInfo, err := dstF.Stat()
+	if err != nil {
+		log.Errorf(err.Error())
+		debug.PrintStack()
+		return f, nil
+	}
+	if !fileInfo.IsDir() { //如果dstF是文件
+		return model.File{
+			Label:     fileInfo.Name(),
+			Name:     fileInfo.Name(),
+			Path:     dstPath,
+			IsDir:    false,
+			Size:     fileInfo.Size(),
+			Children: nil,
+		}, nil
+	} else { //如果dstF是文件夹
+		dir, err := dstF.Readdir(0) //获取文件夹下各个文件或文件夹的fileInfo
+		if err != nil {
+			log.Errorf(err.Error())
+			debug.PrintStack()
+			return f, nil
+		}
+		f = model.File{
+			Label:    path.Base(dstPath),
+			Name:     path.Base(dstPath),
+			Path:     dstPath,
+			IsDir:    true,
+			Size:     0,
+			Children: nil,
+		}
+		for _, subFileInfo := range dir {
+			subFileNode, err := GetFileNodeTree(path.Join(dstPath, subFileInfo.Name()), level+1)
+			if err != nil {
+				log.Errorf(err.Error())
+				debug.PrintStack()
+				return f, err
+			}
+			f.Children = append(f.Children, subFileNode)
+		}
+		return f, nil
+	}
+}
diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index 50b7d306..f7ec8a21 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -1,101 +1,128 @@
 <template>
   <div class="file-list-container">
-    <div class="top-part">
-      <!--back-->
-      <div class="action-container">
-        <el-button v-if="showFile" type="primary" size="small" style="margin-right: 10px;" @click="onBackFile">
-          <font-awesome-icon :icon="['fa', 'arrow-left']"/>
-          {{$t('Back')}}
-        </el-button>
-        <el-popover v-model="isShowDelete">
-          <el-button size="small" type="default" @click="() => this.isShowDelete = false">
-            {{$t('Cancel')}}
-          </el-button>
-          <el-button size="small" type="danger" @click="onFileDelete">
-            {{$t('Confirm')}}
-          </el-button>
-          <template slot="reference">
-            <el-button v-if="currentPath !== ''" type="danger" size="small" style="margin-right: 10px;"
-                       @click="() => this.isShowDelete = true">
-              <font-awesome-icon :icon="['fa', 'trash']"/>
-              {{$t('Remove')}}
+    <div class="file-tree-wrapper">
+      <el-tree
+        :data="computedFileTree"
+        node-key="path"
+        :highlight-current="true"
+        @node-contextmenu="onFileRightClick"
+        @node-click="onFileClick"
+      >
+        <span class="custom-tree-node" slot-scope="{ node, data }">
+          <el-popover v-model="isShowCreatePopoverDict[data.path]" type="manual" placement="right"
+                      popper-class="create-item-popover">
+            <div class="create-item-title">{{$t('Create')}}</div>
+            <ul class="create-item-list">
+              <li class="create-item">
+                <font-awesome-icon :icon="['fa', 'folder']" color="rgba(3,47,98,.5)"/>
+                <span class="create-item-text">{{$t('Directory')}}</span>
+              </li>
+              <li class="create-item">
+                <font-awesome-icon icon="file-alt" color="rgba(3,47,98,.5)"/>
+                <span class="create-item-text">{{$t('File')}}</span>
+              </li>
+            </ul>
+            <template slot="reference">
+              <div>
+                <span class="item-icon">
+                  <font-awesome-icon v-if="data.is_dir" :icon="['fa', 'folder']" color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else-if="data.path.match(/\.py$/)" :icon="['fab','python']"
+                                     color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else-if="data.path.match(/\.js$/)" :icon="['fab','node-js']"
+                                     color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else-if="data.path.match(/\.(java|jar|class)$/)" :icon="['fab','java']"
+                                     color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else-if="data.path.match(/\.go$/)" :icon="['fab','go']"
+                                     color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else-if="data.path.match(/\.zip$/)" :icon="['fa','file-archive']"
+                                     color="rgba(3,47,98,.5)"/>
+                  <font-awesome-icon v-else icon="file-alt" color="rgba(3,47,98,.5)"/>
+                </span>
+                <span class="item-name" :class="isActiveFile(data) ? 'active' : ''">
+                  {{data.name}}
+                </span>
+              </div>
+            </template>
+          </el-popover>
+        </span>
+      </el-tree>
+    </div>
+    <div class="main-content">
+      <div class="top-part">
+        <!--back-->
+        <div class="action-container">
+          <!--          <el-button v-if="showFile" type="primary" size="small" style="margin-right: 10px;" @click="onBackFile">-->
+          <!--            <font-awesome-icon :icon="['fa', 'arrow-left']"/>-->
+          <!--            {{$t('Back')}}-->
+          <!--          </el-button>-->
+          <el-popover v-model="isShowDelete">
+            <el-button size="small" type="default" @click="() => this.isShowDelete = false">
+              {{$t('Cancel')}}
             </el-button>
-          </template>
-        </el-popover>
-        <el-popover v-model="isShowRename">
-          <el-input v-model="name" :placeholder="$t('Name')" style="margin-bottom: 10px"/>
-          <div style="text-align: right">
-            <el-button size="small" type="warning" @click="onRenameFile">
+            <el-button size="small" type="danger" @click="onFileDelete">
               {{$t('Confirm')}}
             </el-button>
-          </div>
-          <template slot="reference">
-            <el-button v-if="showFile" type="warning" size="small" style="margin-right: 10px;"
-                       @click="onOpenRename">
-              <font-awesome-icon :icon="['fa', 'redo']"/>
-              {{$t('Rename')}}
-            </el-button>
-          </template>
-        </el-popover>
-        <el-button v-if="showFile" type="success" size="small" style="margin-right: 10px;" @click="onFileSave">
-          <font-awesome-icon :icon="['fa', 'save']"/>
-          {{$t('Save')}}
-        </el-button>
-        <el-popover v-if="!showFile" v-model="isShowAdd" @hide="onHideAdd">
-          <el-input v-model="name" :placeholder="$t('Name')"/>
-          <div class="add-type-list">
-            <el-button size="small" type="success" icon="el-icon-document-add" @click="onAddFile">
-              {{$t('File')}}
-            </el-button>
-            <el-button size="small" type="primary" icon="el-icon-folder-add" @click="onAddDir">
-              {{$t('Directory')}}
-            </el-button>
-          </div>
-          <template slot="reference">
-            <el-button type="primary" size="small" style="margin-right: 10px;">
-              <font-awesome-icon :icon="['fa', 'plus']"/>
-              {{$t('Create')}}
-            </el-button>
-          </template>
-        </el-popover>
-      </div>
-      <!--./back-->
+            <template slot="reference">
+              <el-button v-if="currentPath !== ''" type="danger" size="small" style="margin-right: 10px;"
+                         @click="() => this.isShowDelete = true">
+                <font-awesome-icon :icon="['fa', 'trash']"/>
+                {{$t('Remove')}}
+              </el-button>
+            </template>
+          </el-popover>
+          <el-popover v-model="isShowRename">
+            <el-input v-model="name" :placeholder="$t('Name')" style="margin-bottom: 10px"/>
+            <div style="text-align: right">
+              <el-button size="small" type="warning" @click="onRenameFile">
+                {{$t('Confirm')}}
+              </el-button>
+            </div>
+            <template slot="reference">
+              <el-button v-if="showFile" type="warning" size="small" style="margin-right: 10px;"
+                         @click="onOpenRename">
+                <font-awesome-icon :icon="['fa', 'redo']"/>
+                {{$t('Rename')}}
+              </el-button>
+            </template>
+          </el-popover>
+          <el-button v-if="showFile" type="success" size="small" style="margin-right: 10px;" @click="onFileSave">
+            <font-awesome-icon :icon="['fa', 'save']"/>
+            {{$t('Save')}}
+          </el-button>
+          <el-popover v-model="isShowAdd" @hide="onHideAdd">
+            <el-input v-model="name" :placeholder="$t('Name')"/>
+            <div class="add-type-list">
+              <el-button size="small" type="success" icon="el-icon-document-add" @click="onAddFile">
+                {{$t('File')}}
+              </el-button>
+              <el-button size="small" type="primary" icon="el-icon-folder-add" @click="onAddDir">
+                {{$t('Directory')}}
+              </el-button>
+            </div>
+            <template slot="reference">
+              <el-button type="primary" size="small" style="margin-right: 10px;">
+                <font-awesome-icon :icon="['fa', 'plus']"/>
+                {{$t('Create')}}
+              </el-button>
+            </template>
+          </el-popover>
+        </div>
+        <!--./back-->
 
-      <!--file path-->
-      <div class="file-path-container">
-        <div class="file-path">. / {{currentPath}}</div>
+        <!--file path-->
+        <div class="file-path-container">
+          <div class="file-path">{{currentPath}}</div>
+        </div>
+        <!--./file path-->
       </div>
-      <!--./file path-->
+
+      <!--file list-->
+      <div v-if="!showFile" class="file-list">
+        {{$t('Please select a file on the left')}}
+      </div>
+
+      <file-detail v-else/>
     </div>
-
-    <!--file list-->
-    <ul v-if="!showFile" class="file-list">
-      <li v-if="currentPath" class="item" @click="onBack">
-        <span class="item-icon"></span>
-        <span class="item-name">..</span>
-      </li>
-      <li v-for="(item, index) in fileList" :key="index" class="item" @click="onItemClick(item)">
-            <span class="item-icon">
-              <font-awesome-icon v-if="item.is_dir" :icon="['fa', 'folder']" color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else-if="item.path.match(/\.py$/)" :icon="['fab','python']"
-                                 color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else-if="item.path.match(/\.js$/)" :icon="['fab','node-js']"
-                                 color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else-if="item.path.match(/\.(java|jar|class)$/)" :icon="['fab','java']"
-                                 color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else-if="item.path.match(/\.go$/)" :icon="['fab','go']"
-                                 color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else-if="item.path.match(/\.zip$/)" :icon="['fa','file-archive']"
-                                 color="rgba(3,47,98,.5)"/>
-              <font-awesome-icon v-else icon="file-alt" color="rgba(3,47,98,.5)"/>
-            </span>
-        <span class="item-name">
-              {{item.name}}
-            </span>
-      </li>
-    </ul>
-
-    <file-detail v-else/>
   </div>
 </template>
 
@@ -116,10 +143,20 @@ export default {
       name: '',
       isShowAdd: false,
       isShowDelete: false,
-      isShowRename: false
+      isShowRename: false,
+      isShowCreatePopoverDict: {},
+      currentFilePath: '.',
+      ignoreFileRegexList: [
+        '__pycache__',
+        'md5.txt',
+        '.pyc'
+      ]
     }
   },
   computed: {
+    ...mapState('spider', [
+      'fileTree'
+    ]),
     ...mapState('file', [
       'fileList'
     ]),
@@ -130,6 +167,12 @@ export default {
       get () {
         return this.$store.state.file.currentPath
       }
+    },
+    computedFileTree () {
+      if (!this.fileTree || !this.fileTree.children) return []
+      let nodes = this.sortFiles(this.fileTree.children)
+      nodes = this.filterFiles(nodes)
+      return nodes
     }
   },
   methods: {
@@ -221,9 +264,58 @@ export default {
       this.$message.success(this.$t('Renamed successfully'))
       this.isShowRename = false
       this.$st.sendEv('爬虫详情', '文件', '重命名')
+    },
+    async getFileTree () {
+      const arr = this.$route.path.split('/')
+      const id = arr[arr.length - 1]
+      await this.$store.dispatch('spider/getFileTree', { id })
+    },
+    async onFileClick (data) {
+      if (data.is_dir) {
+        return
+      }
+      this.currentFilePath = data.path
+      this.onItemClick(data)
+    },
+    sortFiles (nodes) {
+      nodes.forEach(node => {
+        if (node.is_dir) {
+          if (!node.children) node.children = []
+          node.children = this.sortFiles(node.children)
+        }
+      })
+      return nodes.sort((a, b) => {
+        if ((a.is_dir && b.is_dir) || (!a.is_dir && !b.is_dir)) {
+          return a.name > b.name ? 1 : -1
+        } else {
+          return a.is_dir ? -1 : 1
+        }
+      })
+    },
+    filterFiles (nodes) {
+      return nodes.filter(node => {
+        if (node.is_dir) {
+          node.children = this.filterFiles(node.children)
+        }
+        for (let i = 0; i < this.ignoreFileRegexList.length; i++) {
+          const regex = this.ignoreFileRegexList[i]
+          if (node.name.match(regex)) {
+            return false
+          }
+        }
+        return true
+      })
+    },
+    isActiveFile (node) {
+      return node.path === this.currentFilePath
+    },
+    onFileRightClick (ev, data) {
+      console.log(data.path)
+      this.$set(this.isShowCreatePopoverDict, data.path, true)
     }
   },
-  created () {
+  async created () {
+    await this.getFileTree()
   }
 }
 </script>
@@ -282,29 +374,17 @@ export default {
     }
 
     .file-list {
-      padding: 0;
-      margin: 0;
+      padding: 10px;
       list-style: none;
-      height: 450px;
+      height: 100%;
       overflow-y: auto;
       min-height: 100%;
       border-radius: 5px;
       border: 1px solid #eaecef;
-
-      .item {
-        padding: 10px 20px;
-        cursor: pointer;
-        color: #303133;
-
-        .item-icon {
-          .fa-folder {
-          }
-        }
-      }
-
-      .item:hover {
-        background-color: rgba(48, 65, 86, 0.1);
-      }
+      display: flex;
+      align-items: center;
+      justify-content: center;
+      font-size: 24px;
     }
   }
 </style>
@@ -348,4 +428,61 @@ export default {
     cursor: pointer;
     font-weight: bolder;
   }
+
+  .file-tree-wrapper {
+    float: left;
+    width: 240px;
+  }
+
+  .file-tree-wrapper >>> .el-tree-node__content {
+    height: 30px;
+  }
+
+  .file-tree-wrapper >>> .el-tree-node__content .item-name.active {
+    font-weight: bolder;
+  }
+
+  .main-content {
+    float: left;
+    width: calc(100% - 240px);
+    height: calc(100vh - 200px);
+  }
+
+</style>
+
+<style>
+  .create-item-popover {
+    padding: 0;
+    margin: 0;
+  }
+
+  .create-item-list {
+    list-style: none;
+    padding: 0;
+    margin: 0
+  }
+
+  .create-item-title {
+    padding-top: 10px;
+    padding-left: 15px;
+    padding-bottom: 10px;
+    border-bottom: 1px solid #eaecef;
+  }
+
+  .create-item-list .create-item {
+    display: flex;
+    align-items: center;
+    height: 35px;
+    margin: 0 0 0 15px;
+    padding: 0;
+    cursor: pointer;
+  }
+
+  .create-item-list .create-item:hover {
+    background: #F5F7FA;
+  }
+
+  .create-item-list .create-item .create-item-text {
+    margin-left: 5px;
+  }
 </style>
diff --git a/frontend/src/store/modules/file.js b/frontend/src/store/modules/file.js
index abdf5638..33f089fe 100644
--- a/frontend/src/store/modules/file.js
+++ b/frontend/src/store/modules/file.js
@@ -38,8 +38,7 @@ const actions = {
   },
   getFileContent ({ commit, rootState }, payload) {
     const { path } = payload
-    const spiderId = rootState.spider.spiderForm._id
-    return request.get(`/spiders/${spiderId}/file`, { path })
+    return request.get(`/file`, { path })
       .then(response => {
         commit('SET_FILE_CONTENT', response.data.data)
       })
diff --git a/frontend/src/store/modules/spider.js b/frontend/src/store/modules/spider.js
index 4632483a..ea913287 100644
--- a/frontend/src/store/modules/spider.js
+++ b/frontend/src/store/modules/spider.js
@@ -38,7 +38,10 @@ const state = {
   previewCrawlData: [],
 
   // template list
-  templateList: []
+  templateList: [],
+
+  // spider file tree
+  fileTree: {}
 }
 
 const getters = {}
@@ -86,6 +89,9 @@ const mutations = {
   },
   SET_TEMPLATE_LIST (state, value) {
     state.templateList = value
+  },
+  SET_FILE_TREE (state, value) {
+    state.fileTree = value
   }
 }
 
@@ -185,6 +191,11 @@ const actions = {
     const { id } = payload
     const res = await request.get(`/spiders/${id}/schedules`)
     commit('schedule/SET_SCHEDULE_LIST', res.data.data, { root: true })
+  },
+  async getFileTree ({ state, commit }, payload) {
+    const { id } = payload
+    const res = await request.get(`/spiders/${id}/file/tree`)
+    commit('SET_FILE_TREE', res.data.data)
   }
 }
 
diff --git a/frontend/src/views/layout/components/AppMain.vue b/frontend/src/views/layout/components/AppMain.vue
index fae53f18..25dfabd9 100644
--- a/frontend/src/views/layout/components/AppMain.vue
+++ b/frontend/src/views/layout/components/AppMain.vue
@@ -22,7 +22,7 @@ export default {
 <style scoped>
   .app-main {
     /*50 = navbar  */
-    min-height: calc(100vh - 50px);
+    min-height: calc(100vh - 100px);
     position: relative;
     overflow: hidden;
   }

From 39c6df11c1d3585e832e4fe340921be0016e2af8 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:10:43 +0800
Subject: [PATCH 21/42] =?UTF-8?q?=E8=B0=83=E6=95=B4=E5=9C=A8=E7=BA=BF?=
 =?UTF-8?q?=E7=BC=96=E8=BE=91=E6=96=87=E4=BB=B6?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 backend/model/file.go    | 13 +++++++------
 backend/routes/spider.go |  2 +-
 backend/services/file.go | 13 +++++++++----
 3 files changed, 17 insertions(+), 11 deletions(-)

diff --git a/backend/model/file.go b/backend/model/file.go
index 1af3b851..a2ad34eb 100644
--- a/backend/model/file.go
+++ b/backend/model/file.go
@@ -20,12 +20,13 @@ type GridFs struct {
 }
 
 type File struct {
-	Name     string `json:"name"`
-	Path     string `json:"path"`
-	IsDir    bool   `json:"is_dir"`
-	Size     int64  `json:"size"`
-	Children []File `json:"children"`
-	Label    string `json:"label"`
+	Name         string `json:"name"`
+	Path         string `json:"path"`
+	RelativePath string `json:"relative_path"`
+	IsDir        bool   `json:"is_dir"`
+	Size         int64  `json:"size"`
+	Children     []File `json:"children"`
+	Label        string `json:"label"`
 }
 
 func (f *GridFs) Remove() {
diff --git a/backend/routes/spider.go b/backend/routes/spider.go
index 2534a7da..4adfb707 100644
--- a/backend/routes/spider.go
+++ b/backend/routes/spider.go
@@ -708,7 +708,7 @@ func RenameSpiderFile(c *gin.Context) {
 
 	// 原文件路径
 	filePath := path.Join(spider.Src, reqBody.Path)
-	newFilePath := path.Join(spider.Src, reqBody.NewPath)
+	newFilePath := path.Join(path.Join(path.Dir(filePath), reqBody.NewPath))
 
 	// 如果新文件已存在，则报错
 	if utils.Exists(newFilePath) {
diff --git a/backend/services/file.go b/backend/services/file.go
index 0b810209..d126fcab 100644
--- a/backend/services/file.go
+++ b/backend/services/file.go
@@ -6,9 +6,14 @@ import (
 	"os"
 	"path"
 	"runtime/debug"
+	"strings"
 )
 
 func GetFileNodeTree(dstPath string, level int) (f model.File, err error) {
+	return getFileNodeTree(dstPath, level, dstPath)
+}
+
+func getFileNodeTree(dstPath string, level int, rootPath string) (f model.File, err error) {
 	dstF, err := os.Open(dstPath)
 	if err != nil {
 		log.Errorf(err.Error())
@@ -24,9 +29,9 @@ func GetFileNodeTree(dstPath string, level int) (f model.File, err error) {
 	}
 	if !fileInfo.IsDir() { //如果dstF是文件
 		return model.File{
-			Label:     fileInfo.Name(),
+			Label:    fileInfo.Name(),
 			Name:     fileInfo.Name(),
-			Path:     dstPath,
+			Path:     strings.Replace(dstPath, rootPath, "", -1),
 			IsDir:    false,
 			Size:     fileInfo.Size(),
 			Children: nil,
@@ -41,13 +46,13 @@ func GetFileNodeTree(dstPath string, level int) (f model.File, err error) {
 		f = model.File{
 			Label:    path.Base(dstPath),
 			Name:     path.Base(dstPath),
-			Path:     dstPath,
+			Path:     strings.Replace(dstPath, rootPath, "", -1),
 			IsDir:    true,
 			Size:     0,
 			Children: nil,
 		}
 		for _, subFileInfo := range dir {
-			subFileNode, err := GetFileNodeTree(path.Join(dstPath, subFileInfo.Name()), level+1)
+			subFileNode, err := getFileNodeTree(path.Join(dstPath, subFileInfo.Name()), level+1, rootPath)
 			if err != nil {
 				log.Errorf(err.Error())
 				debug.PrintStack()

From ce82377dad7a62fde03944e81a36184e973302c7 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:10:51 +0800
Subject: [PATCH 22/42] =?UTF-8?q?=E8=B0=83=E6=95=B4=E5=9C=A8=E7=BA=BF?=
 =?UTF-8?q?=E7=BC=96=E8=BE=91=E6=96=87=E4=BB=B6?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue | 191 ++++++++++------------
 frontend/src/i18n/zh.js                   |   1 +
 frontend/src/store/modules/file.js        |   3 +-
 frontend/src/store/modules/spider.js      |   2 +-
 4 files changed, 91 insertions(+), 106 deletions(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index f7ec8a21..eae2b4a0 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -9,9 +9,14 @@
         @node-click="onFileClick"
       >
         <span class="custom-tree-node" slot-scope="{ node, data }">
-          <el-popover v-model="isShowCreatePopoverDict[data.path]" type="manual" placement="right"
-                      popper-class="create-item-popover">
-            <div class="create-item-title">{{$t('Create')}}</div>
+          <el-popover v-model="isShowCreatePopoverDict[data.path]" trigger="manual" placement="right"
+                      popper-class="create-item-popover" :visible-arrow="false" @hide="onHideCreate(data)">
+            <div class="create-item-title">
+              <span class="item-icon">
+                <font-awesome-icon icon="plus" color="rgba(3,47,98,.5)"/>
+              </span>
+              <span class="create-item-text">{{$t('Create')}}</span>
+            </div>
             <ul class="create-item-list">
               <li class="create-item">
                 <font-awesome-icon :icon="['fa', 'folder']" color="rgba(3,47,98,.5)"/>
@@ -47,81 +52,60 @@
         </span>
       </el-tree>
     </div>
+
     <div class="main-content">
-      <div class="top-part">
-        <!--back-->
-        <div class="action-container">
-          <!--          <el-button v-if="showFile" type="primary" size="small" style="margin-right: 10px;" @click="onBackFile">-->
-          <!--            <font-awesome-icon :icon="['fa', 'arrow-left']"/>-->
-          <!--            {{$t('Back')}}-->
-          <!--          </el-button>-->
-          <el-popover v-model="isShowDelete">
-            <el-button size="small" type="default" @click="() => this.isShowDelete = false">
-              {{$t('Cancel')}}
-            </el-button>
-            <el-button size="small" type="danger" @click="onFileDelete">
-              {{$t('Confirm')}}
-            </el-button>
-            <template slot="reference">
-              <el-button v-if="currentPath !== ''" type="danger" size="small" style="margin-right: 10px;"
-                         @click="() => this.isShowDelete = true">
-                <font-awesome-icon :icon="['fa', 'trash']"/>
-                {{$t('Remove')}}
+      <div v-if="!showFile" class="file-list">
+        {{$t('Please select a file on the left.')}}
+      </div>
+      <template v-else>
+        <div class="top-part">
+          <!--back-->
+          <div class="action-container">
+            <el-popover v-model="isShowDelete" trigger="click">
+              <el-button size="small" type="default" @click="() => this.isShowDelete = false">
+                {{$t('Cancel')}}
               </el-button>
-            </template>
-          </el-popover>
-          <el-popover v-model="isShowRename">
-            <el-input v-model="name" :placeholder="$t('Name')" style="margin-bottom: 10px"/>
-            <div style="text-align: right">
-              <el-button size="small" type="warning" @click="onRenameFile">
+              <el-button size="small" type="danger" @click="onFileDelete">
                 {{$t('Confirm')}}
               </el-button>
-            </div>
-            <template slot="reference">
-              <el-button v-if="showFile" type="warning" size="small" style="margin-right: 10px;"
-                         @click="onOpenRename">
-                <font-awesome-icon :icon="['fa', 'redo']"/>
-                {{$t('Rename')}}
-              </el-button>
-            </template>
-          </el-popover>
-          <el-button v-if="showFile" type="success" size="small" style="margin-right: 10px;" @click="onFileSave">
-            <font-awesome-icon :icon="['fa', 'save']"/>
-            {{$t('Save')}}
-          </el-button>
-          <el-popover v-model="isShowAdd" @hide="onHideAdd">
-            <el-input v-model="name" :placeholder="$t('Name')"/>
-            <div class="add-type-list">
-              <el-button size="small" type="success" icon="el-icon-document-add" @click="onAddFile">
-                {{$t('File')}}
-              </el-button>
-              <el-button size="small" type="primary" icon="el-icon-folder-add" @click="onAddDir">
-                {{$t('Directory')}}
-              </el-button>
-            </div>
-            <template slot="reference">
-              <el-button type="primary" size="small" style="margin-right: 10px;">
-                <font-awesome-icon :icon="['fa', 'plus']"/>
-                {{$t('Create')}}
-              </el-button>
-            </template>
-          </el-popover>
+              <template slot="reference">
+                <el-button type="danger" size="small" style="margin-right: 10px;">
+                  <font-awesome-icon :icon="['fa', 'trash']"/>
+                  {{$t('Remove')}}
+                </el-button>
+              </template>
+            </el-popover>
+            <el-popover v-model="isShowRename" trigger="click">
+              <el-input v-model="name" :placeholder="$t('Name')" style="margin-bottom: 10px"/>
+              <div style="text-align: right">
+                <el-button size="small" type="warning" @click="onRenameFile">
+                  {{$t('Confirm')}}
+                </el-button>
+              </div>
+              <template slot="reference">
+                <div>
+                  <el-button type="warning" size="small" style="margin-right: 10px;" @click="onOpenRename">
+                    <font-awesome-icon :icon="['fa', 'redo']"/>
+                    {{$t('Rename')}}
+                  </el-button>
+                </div>
+              </template>
+            </el-popover>
+            <el-button type="success" size="small" style="margin-right: 10px;" @click="onFileSave">
+              <font-awesome-icon :icon="['fa', 'save']"/>
+              {{$t('Save')}}
+            </el-button>
+          </div>
+          <!--./back-->
+
+          <!--file path-->
+          <div class="file-path-container">
+            <div class="file-path">{{currentPath}}</div>
+          </div>
+          <!--./file path-->
         </div>
-        <!--./back-->
-
-        <!--file path-->
-        <div class="file-path-container">
-          <div class="file-path">{{currentPath}}</div>
-        </div>
-        <!--./file path-->
-      </div>
-
-      <!--file list-->
-      <div v-if="!showFile" class="file-list">
-        {{$t('Please select a file on the left')}}
-      </div>
-
-      <file-detail v-else/>
+        <file-detail/>
+      </template>
     </div>
   </div>
 </template>
@@ -192,24 +176,11 @@ export default {
       }
       this.$st.sendEv('爬虫详情', '文件', '点击')
     },
-    onBack () {
-      const sep = '/'
-      let arr = this.currentPath.split(sep)
-      arr.splice(arr.length - 1, 1)
-      const path = arr.join(sep)
-      this.$store.commit('file/SET_CURRENT_PATH', path)
-      this.$store.dispatch('file/getFileList', { path: this.currentPath })
-      this.$st.sendEv('爬虫详情', '文件', '回退')
-    },
     async onFileSave () {
       await this.$store.dispatch('file/saveFileContent', { path: this.currentPath })
       this.$message.success(this.$t('Saved file successfully'))
       this.$st.sendEv('爬虫详情', '文件', '保存')
     },
-    onBackFile () {
-      this.showFile = false
-      this.onBack()
-    },
     onHideAdd () {
       this.name = ''
     },
@@ -240,27 +211,24 @@ export default {
       this.$st.sendEv('爬虫详情', '文件', '添加')
     },
     async onFileDelete () {
-      await this.$store.dispatch('file/deleteFile', { path: this.currentPath })
+      await this.$store.dispatch('file/deleteFile', { path: this.currentFilePath })
+      await this.$store.dispatch('spider/getFileTree')
       this.$message.success(this.$t('Deleted successfully'))
       this.isShowDelete = false
-      this.onBackFile()
+      this.showFile = false
       this.$st.sendEv('爬虫详情', '文件', '删除')
     },
     onOpenRename () {
-      this.isShowRename = true
-      const arr = this.currentPath.split('/')
+      const arr = this.currentFilePath.split('/')
       this.name = arr[arr.length - 1]
     },
     async onRenameFile () {
-      let newPath
-      if (this.currentPath.split('/').length === 1) {
-        newPath = this.name
-      } else {
-        const arr = this.currentPath.split('/')
-        newPath = arr[0] + '/' + this.name
-      }
-      await this.$store.dispatch('file/renameFile', { path: this.currentPath, newPath })
-      this.$store.commit('file/SET_CURRENT_PATH', newPath)
+      await this.$store.dispatch('file/renameFile', { path: this.currentFilePath, newPath: this.name })
+      await this.$store.dispatch('spider/getFileTree')
+      const arr = this.currentFilePath.split('/')
+      arr[arr.length - 1] = this.name
+      this.currentFilePath = arr.join('/')
+      this.$store.commit('file/SET_CURRENT_PATH', this.currentFilePath)
       this.$message.success(this.$t('Renamed successfully'))
       this.isShowRename = false
       this.$st.sendEv('爬虫详情', '文件', '重命名')
@@ -310,18 +278,30 @@ export default {
       return node.path === this.currentFilePath
     },
     onFileRightClick (ev, data) {
-      console.log(data.path)
+      this.isShowCreatePopoverDict = {}
       this.$set(this.isShowCreatePopoverDict, data.path, true)
+    },
+    onHideCreate (data) {
+      this.$set(this.isShowCreatePopoverDict, data.path, false)
     }
   },
   async created () {
     await this.getFileTree()
+  },
+  mounted () {
+    this.listener = document.querySelector('body').addEventListener('click', ev => {
+      this.isShowCreatePopoverDict = {}
+    })
+  },
+  destroyed () {
+    document.querySelector('body').removeEventListener('click', this.listener)
   }
 }
 </script>
 
 <style scoped lang="scss">
   .file-list-container {
+    //margin-left: -15px;
     height: 100%;
     min-height: 100%;
 
@@ -379,8 +359,8 @@ export default {
       height: 100%;
       overflow-y: auto;
       min-height: 100%;
-      border-radius: 5px;
-      border: 1px solid #eaecef;
+      /*border-radius: 5px;*/
+      /*border: 1px solid #eaecef;*/
       display: flex;
       align-items: center;
       justify-content: center;
@@ -432,6 +412,7 @@ export default {
   .file-tree-wrapper {
     float: left;
     width: 240px;
+    height: calc(100vh - 200px);
   }
 
   .file-tree-wrapper >>> .el-tree-node__content {
@@ -446,6 +427,8 @@ export default {
     float: left;
     width: calc(100% - 240px);
     height: calc(100vh - 200px);
+    border-left: 1px solid #eaecef;
+    padding-left: 20px;
   }
 
 </style>
@@ -473,8 +456,8 @@ export default {
     display: flex;
     align-items: center;
     height: 35px;
-    margin: 0 0 0 15px;
-    padding: 0;
+    padding: 0 0 0 15px;
+    margin: 0;
     cursor: pointer;
   }
 
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index 136dbf3f..27f32d61 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -325,6 +325,7 @@ export default {
   'The schedule has been added': '已添加定时任务',
   'The schedule has been saved': '已保存定时任务',
   'Email format invalid': '邮箱地址格式不正确',
+  'Please select a file on the left.': '请在左侧选择一个文件.',
 
   // 登录
   'Sign in': '登录',
diff --git a/frontend/src/store/modules/file.js b/frontend/src/store/modules/file.js
index 33f089fe..abdf5638 100644
--- a/frontend/src/store/modules/file.js
+++ b/frontend/src/store/modules/file.js
@@ -38,7 +38,8 @@ const actions = {
   },
   getFileContent ({ commit, rootState }, payload) {
     const { path } = payload
-    return request.get(`/file`, { path })
+    const spiderId = rootState.spider.spiderForm._id
+    return request.get(`/spiders/${spiderId}/file`, { path })
       .then(response => {
         commit('SET_FILE_CONTENT', response.data.data)
       })
diff --git a/frontend/src/store/modules/spider.js b/frontend/src/store/modules/spider.js
index ea913287..ac1fc81b 100644
--- a/frontend/src/store/modules/spider.js
+++ b/frontend/src/store/modules/spider.js
@@ -193,7 +193,7 @@ const actions = {
     commit('schedule/SET_SCHEDULE_LIST', res.data.data, { root: true })
   },
   async getFileTree ({ state, commit }, payload) {
-    const { id } = payload
+    const id = payload ? payload.id : state.spiderForm._id
     const res = await request.get(`/spiders/${id}/file/tree`)
     commit('SET_FILE_TREE', res.data.data)
   }

From 5ec89bddaba2a2f64b9ad0f680fb1539372a799e Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:34:44 +0800
Subject: [PATCH 23/42] =?UTF-8?q?=E5=8A=A0=E5=85=A5=E6=B7=BB=E5=8A=A0?=
 =?UTF-8?q?=E6=96=87=E4=BB=B6=E7=9B=AE=E5=BD=95?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue | 69 +++++++++++++++++++----
 frontend/src/i18n/zh.js                   |  5 ++
 2 files changed, 63 insertions(+), 11 deletions(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index eae2b4a0..5723418b 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -1,5 +1,35 @@
 <template>
   <div class="file-list-container">
+    <el-dialog
+      :title="$t('New Directory')"
+      :visible.sync="dirDialogVisible"
+      width="30%">
+      <el-form>
+        <el-form-item :label="$t('Enter new directory name')">
+          <el-input v-model="name" :placeholder="$t('New directory name')"></el-input>
+        </el-form-item>
+      </el-form>
+      <span slot="footer" class="dialog-footer">
+        <el-button @click="dirDialogVisible = false">{{$t('Cancel')}}</el-button>
+        <el-button type="primary" @click="onAddDir">{{$t('Confirm')}}</el-button>
+      </span>
+    </el-dialog>
+
+    <el-dialog
+      :title="$t('New File')"
+      :visible.sync="fileDialogVisible"
+      width="30%">
+      <el-form>
+        <el-form-item :label="$t('Enter new file name')">
+          <el-input v-model="name" :placeholder="$t('New file name')"></el-input>
+        </el-form-item>
+      </el-form>
+      <span slot="footer" class="dialog-footer">
+        <el-button @click="fileDialogVisible = false">{{$t('Cancel')}}</el-button>
+        <el-button type="primary" @click="onAddFile">{{$t('Confirm')}}</el-button>
+      </span>
+    </el-dialog>
+
     <div class="file-tree-wrapper">
       <el-tree
         :data="computedFileTree"
@@ -18,11 +48,11 @@
               <span class="create-item-text">{{$t('Create')}}</span>
             </div>
             <ul class="create-item-list">
-              <li class="create-item">
+              <li class="create-item" @click="dirDialogVisible = true">
                 <font-awesome-icon :icon="['fa', 'folder']" color="rgba(3,47,98,.5)"/>
                 <span class="create-item-text">{{$t('Directory')}}</span>
               </li>
-              <li class="create-item">
+              <li class="create-item" @click="fileDialogVisible = true">
                 <font-awesome-icon icon="file-alt" color="rgba(3,47,98,.5)"/>
                 <span class="create-item-text">{{$t('File')}}</span>
               </li>
@@ -134,7 +164,10 @@ export default {
         '__pycache__',
         'md5.txt',
         '.pyc'
-      ]
+      ],
+      activeFileNode: {},
+      dirDialogVisible: false,
+      fileDialogVisible: false
     }
   },
   computed: {
@@ -181,19 +214,22 @@ export default {
       this.$message.success(this.$t('Saved file successfully'))
       this.$st.sendEv('爬虫详情', '文件', '保存')
     },
-    onHideAdd () {
-      this.name = ''
-    },
     async onAddFile () {
       if (!this.name) {
         this.$message.error(this.$t('Name cannot be empty'))
         return
       }
-      const path = this.currentPath + '/' + this.name
+      const arr = this.activeFileNode.path.split('/')
+      if (this.activeFileNode.is_dir) {
+        arr.push(this.name)
+      } else {
+        arr[arr.length - 1] = this.name
+      }
+      const path = arr.join('/')
       await this.$store.dispatch('file/addFile', { path })
-      await this.$store.dispatch('file/getFileList', { path: this.currentPath })
+      await this.$store.dispatch('spider/getFileTree')
       this.isShowAdd = false
-
+      this.fileDialogVisible = false
       this.showFile = true
       this.$store.commit('file/SET_FILE_CONTENT', '')
       this.$store.commit('file/SET_CURRENT_PATH', path)
@@ -205,9 +241,17 @@ export default {
         this.$message.error(this.$t('Name cannot be empty'))
         return
       }
-      await this.$store.dispatch('file/addDir', { path: this.currentPath + '/' + this.name })
-      await this.$store.dispatch('file/getFileList', { path: this.currentPath })
+      const arr = this.activeFileNode.path.split('/')
+      if (this.activeFileNode.is_dir) {
+        arr.push(this.name)
+      } else {
+        arr[arr.length - 1] = this.name
+      }
+      const path = arr.join('/')
+      await this.$store.dispatch('file/addDir', { path })
+      await this.$store.dispatch('spider/getFileTree')
       this.isShowAdd = false
+      this.dirDialogVisible = false
       this.$st.sendEv('爬虫详情', '文件', '添加')
     },
     async onFileDelete () {
@@ -280,9 +324,11 @@ export default {
     onFileRightClick (ev, data) {
       this.isShowCreatePopoverDict = {}
       this.$set(this.isShowCreatePopoverDict, data.path, true)
+      this.activeFileNode = data
     },
     onHideCreate (data) {
       this.$set(this.isShowCreatePopoverDict, data.path, false)
+      this.name = ''
     }
   },
   async created () {
@@ -413,6 +459,7 @@ export default {
     float: left;
     width: 240px;
     height: calc(100vh - 200px);
+    overflow: auto;
   }
 
   .file-tree-wrapper >>> .el-tree-node__content {
diff --git a/frontend/src/i18n/zh.js b/frontend/src/i18n/zh.js
index 27f32d61..a5dddfcc 100644
--- a/frontend/src/i18n/zh.js
+++ b/frontend/src/i18n/zh.js
@@ -326,6 +326,11 @@ export default {
   'The schedule has been saved': '已保存定时任务',
   'Email format invalid': '邮箱地址格式不正确',
   'Please select a file on the left.': '请在左侧选择一个文件.',
+  'New Directory': '新建目录',
+  'Enter new directory name': '输入新目录名称',
+  'New directory name': '新目录名称',
+  'Enter new file name': '输入新文件名称',
+  'New file name': '新文件名称',
 
   // 登录
   'Sign in': '登录',

From c76d88733e4891418d443c6b49689c2e4dcaa90b Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:36:16 +0800
Subject: [PATCH 24/42] =?UTF-8?q?=E6=94=AF=E6=8C=81Spiderfile=E9=AB=98?=
 =?UTF-8?q?=E4=BA=AE?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileDetail.vue | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/frontend/src/components/File/FileDetail.vue b/frontend/src/components/File/FileDetail.vue
index f5f8a4cc..4e11e019 100644
--- a/frontend/src/components/File/FileDetail.vue
+++ b/frontend/src/components/File/FileDetail.vue
@@ -62,7 +62,7 @@ export default {
         return 'text/x-php'
       } else if (fileName.match(/\.md$/)) {
         return 'text/x-markdown'
-      } else if (fileName === 'Spiderfile') {
+      } else if (fileName.match('Spiderfile')) {
         return 'text/x-yaml'
       } else {
         return 'text'

From 5ad3b07627d05deecc988b2cdff7b03b8dd027f4 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:40:02 +0800
Subject: [PATCH 25/42] updated version

---
 backend/template/spiderfile/Spiderfile.163_news       | 2 +-
 backend/template/spiderfile/Spiderfile.baidu          | 2 +-
 backend/template/spiderfile/Spiderfile.toscrapy_books | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/backend/template/spiderfile/Spiderfile.163_news b/backend/template/spiderfile/Spiderfile.163_news
index c2a73be7..bc224395 100644
--- a/backend/template/spiderfile/Spiderfile.163_news
+++ b/backend/template/spiderfile/Spiderfile.163_news
@@ -1,4 +1,4 @@
-version: "0.4.0"
+version: "0.4.4"
 name: "toscrapy_books"
 start_url: "http://news.163.com/special/0001386F/rank_news.html"
 start_stage: "list"
diff --git a/backend/template/spiderfile/Spiderfile.baidu b/backend/template/spiderfile/Spiderfile.baidu
index 5643c980..4c8db4a3 100644
--- a/backend/template/spiderfile/Spiderfile.baidu
+++ b/backend/template/spiderfile/Spiderfile.baidu
@@ -1,4 +1,4 @@
-version: 0.4.0
+version: 0.4.4
 name: toscrapy_books
 start_url: http://www.baidu.com/s?wd=crawlab
 start_stage: list
diff --git a/backend/template/spiderfile/Spiderfile.toscrapy_books b/backend/template/spiderfile/Spiderfile.toscrapy_books
index 247b4f40..6d1542f7 100644
--- a/backend/template/spiderfile/Spiderfile.toscrapy_books
+++ b/backend/template/spiderfile/Spiderfile.toscrapy_books
@@ -1,4 +1,4 @@
-version: "0.4.0"
+version: "0.4.4"
 name: "toscrapy_books"
 start_url: "http://books.toscrape.com"
 start_stage: "list"

From 4416c01e887d43976615349b32857bab179a11ac Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:44:00 +0800
Subject: [PATCH 26/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 1 +
 CHANGELOG.md    | 1 +
 2 files changed, 2 insertions(+)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index 5fa8dc0e..0b877c5b 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -6,6 +6,7 @@
 - **企业微信机器人通知**. 允许用户发送企业微信机器人消息通知.
 - **API 地址优化**. 在前端加入相对路径，因此用户不需要特别注明 `CRAWLAB_API_ADDRESS`.
 - **SDK 兼容**. 允许用户通过 Crawlab SDK 与 Scrapy 或通用爬虫集成.
+- **优化文件管理**. 加入树状文件侧边栏，让用户更方便的编辑文件.
 
 ### Bug 修复
 - **`nil retuened` 错误**.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index ba842b9a..19dad6da 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,7 @@
 - **Wechat Robot Notification**. Allow users to send Wechat Robot notifications.
 - **API Address Optimization**. Added relative URL path in frontend so that users don't have to specify `CRAWLAB_API_ADDRESS` explicitly.
 - **SDK Compatiblity**. Allow users to integrate Scrapy or general spiders with Crawlab SDK.
+- **Enhanced File Management**. Added tree-like file sidebar to allow users to edit files much more easier.
 
 ### Bug Fixes
 - **`nil retuened` error**.

From 4b3af25258a6c82103d61ead35904ed06b27c261 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Wed, 15 Jan 2020 20:48:49 +0800
Subject: [PATCH 27/42] =?UTF-8?q?=E6=A0=B7=E5=BC=8F=E8=B0=83=E6=95=B4?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index 5723418b..26ed393a 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -347,7 +347,7 @@ export default {
 
 <style scoped lang="scss">
   .file-list-container {
-    //margin-left: -15px;
+    margin-left: -15px;
     height: 100%;
     min-height: 100%;
 

From 0e4bf36f7c00a1f85be570197906df323b5b3b95 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 14:45:48 +0800
Subject: [PATCH 28/42] =?UTF-8?q?=E5=AE=9A=E6=97=B6=E4=BB=BB=E5=8A=A1?=
 =?UTF-8?q?=E7=94=9F=E6=88=90=E5=89=8D=E7=AB=AF=E7=95=8C=E9=9D=A2?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/.eslintignore                        |   1 +
 frontend/package.json                         |   1 +
 frontend/src/components/Cron/index.vue        | 466 ++++++++++++++++++
 frontend/src/components/Cron/language/cn.js   |  61 +++
 frontend/src/components/Cron/language/en.js   |  62 +++
 .../src/components/Cron/language/index.js     |   7 +
 frontend/src/views/schedule/ScheduleList.vue  |  40 +-
 frontend/yarn.lock                            |  14 +
 8 files changed, 643 insertions(+), 9 deletions(-)
 create mode 100644 frontend/.eslintignore
 create mode 100644 frontend/src/components/Cron/index.vue
 create mode 100644 frontend/src/components/Cron/language/cn.js
 create mode 100644 frontend/src/components/Cron/language/en.js
 create mode 100644 frontend/src/components/Cron/language/index.js

diff --git a/frontend/.eslintignore b/frontend/.eslintignore
new file mode 100644
index 00000000..c3af8579
--- /dev/null
+++ b/frontend/.eslintignore
@@ -0,0 +1 @@
+lib/
diff --git a/frontend/package.json b/frontend/package.json
index af39a971..36a6abdb 100644
--- a/frontend/package.json
+++ b/frontend/package.json
@@ -20,6 +20,7 @@
     "@tinymce/tinymce-vue": "^2.0.0",
     "ansi-to-html": "^0.6.13",
     "axios": "0.18.0",
+    "babel-polyfill": "^6.26.0",
     "cross-env": "^5.2.0",
     "dayjs": "^1.8.6",
     "echarts": "^4.1.0",
diff --git a/frontend/src/components/Cron/index.vue b/frontend/src/components/Cron/index.vue
new file mode 100644
index 00000000..6959d1f5
--- /dev/null
+++ b/frontend/src/components/Cron/index.vue
@@ -0,0 +1,466 @@
+<style lang="scss" scoped>
+  #changeContab {
+    .language {
+      position: absolute;
+      right: 25px;
+      z-index: 1;
+    }
+
+    .el-tabs {
+      box-shadow: none;
+    }
+
+    .tabBody {
+      .el-row {
+        margin: 10px 0;
+
+        .long {
+          .el-select {
+            width: 350px;
+          }
+        }
+
+        .el-input-number {
+          width: 110px;
+        }
+      }
+    }
+
+    .bottom {
+      width: 100%;
+      text-align: center;
+      margin-top: 5px;
+      position: relative;
+
+      .value {
+        font-size: 18px;
+        vertical-align: middle;
+      }
+    }
+  }
+</style>
+<template>
+  <div id="changeContab">
+    <!--        <el-button class="language" type="text" @click="i18n=(i18n==='en'?'cn':'en')">{{i18n}}</el-button>-->
+    <el-tabs type="border-card">
+      <el-tab-pane>
+        <span slot="label"><i class="el-icon-date"></i> {{text.Minutes.name}}</span>
+        <div class="tabBody">
+          <el-row>
+            <el-radio v-model="minute.cronEvery" label="1">{{text.Minutes.every}}</el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="minute.cronEvery" label="2">{{text.Minutes.interval[0]}}
+              <el-input-number size="small" v-model="minute.incrementIncrement" :min="1" :max="60"></el-input-number>
+              {{text.Minutes.interval[1]}}
+              <el-input-number size="small" v-model="minute.incrementStart" :min="0" :max="59"></el-input-number>
+              {{text.Minutes.interval[2]||''}}
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="minute.cronEvery" label="3">{{text.Minutes.specific}}
+              <el-select size="small" multiple v-model="minute.specificSpecific">
+                <el-option v-for="val in 60" :key="val" :value="val-1">{{val-1}}</el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="minute.cronEvery" label="4">{{text.Minutes.cycle[0]}}
+              <el-input-number size="small" v-model="minute.rangeStart" :min="0" :max="60"></el-input-number>
+              {{text.Minutes.cycle[1]}}
+              <el-input-number size="small" v-model="minute.rangeEnd" :min="0" :max="60"></el-input-number>
+              {{text.Minutes.cycle[2]}}
+            </el-radio>
+          </el-row>
+        </div>
+      </el-tab-pane>
+      <el-tab-pane>
+        <span slot="label"><i class="el-icon-date"></i> {{text.Hours.name}}</span>
+        <div class="tabBody">
+          <el-row>
+            <el-radio v-model="hour.cronEvery" label="1">{{text.Hours.every}}</el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="hour.cronEvery" label="2">{{text.Hours.interval[0]}}
+              <el-input-number size="small" v-model="hour.incrementIncrement" :min="0" :max="23"></el-input-number>
+              {{text.Hours.interval[1]}}
+              <el-input-number size="small" v-model="hour.incrementStart" :min="0" :max="23"></el-input-number>
+              {{text.Hours.interval[2]}}
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="hour.cronEvery" label="3">{{text.Hours.specific}}
+              <el-select size="small" multiple v-model="hour.specificSpecific">
+                <el-option v-for="val in 24" :key="val" :value="val-1">{{val-1}}</el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="hour.cronEvery" label="4">{{text.Hours.cycle[0]}}
+              <el-input-number size="small" v-model="hour.rangeStart" :min="0" :max="23"></el-input-number>
+              {{text.Hours.cycle[1]}}
+              <el-input-number size="small" v-model="hour.rangeEnd" :min="0" :max="23"></el-input-number>
+              {{text.Hours.cycle[2]}}
+            </el-radio>
+          </el-row>
+        </div>
+      </el-tab-pane>
+      <el-tab-pane>
+        <span slot="label"><i class="el-icon-date"></i> {{text.Day.name}}</span>
+        <div class="tabBody">
+          <el-row>
+            <el-radio v-model="day.cronEvery" label="1">{{text.Day.every}}</el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="day.cronEvery" label="2">{{text.Day.intervalDay[0]}}
+              <el-input-number size="small" v-model="day.incrementIncrement" :min="1" :max="31"></el-input-number>
+              {{text.Day.intervalDay[1]}}
+              <el-input-number size="small" v-model="day.incrementStart" :min="1" :max="31"></el-input-number>
+              {{text.Day.intervalDay[2]}}
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="day.cronEvery" label="3">{{text.Day.specificDay}}
+              <el-select size="small" multiple v-model="day.specificSpecific">
+                <el-option v-for="val in 31" :key="val" :value="val">{{val}}</el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="day.cronEvery" label="4">{{text.Day.cycle[0]}}
+              <el-input-number size="small" v-model="day.rangeStart" :min="1" :max="12"></el-input-number>
+              {{text.Day.cycle[1]}}
+              <el-input-number size="small" v-model="day.rangeEnd" :min="1" :max="12"></el-input-number>
+            </el-radio>
+          </el-row>
+        </div>
+      </el-tab-pane>
+      <el-tab-pane>
+        <span slot="label"><i class="el-icon-date"></i> {{text.Month.name}}</span>
+        <div class="tabBody">
+          <el-row>
+            <el-radio v-model="month.cronEvery" label="1">{{text.Month.every}}</el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="month.cronEvery" label="2">{{text.Month.interval[0]}}
+              <el-input-number size="small" v-model="month.incrementIncrement" :min="0" :max="12"></el-input-number>
+              {{text.Month.interval[1]}}
+              <el-input-number size="small" v-model="month.incrementStart" :min="0" :max="12"></el-input-number>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="month.cronEvery" label="3">{{text.Month.specific}}
+              <el-select size="small" multiple v-model="month.specificSpecific">
+                <el-option v-for="val in 12" :key="val" :label="val" :value="val"></el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="month.cronEvery" label="4">{{text.Month.cycle[0]}}
+              <el-input-number size="small" v-model="month.rangeStart" :min="1" :max="31"></el-input-number>
+              {{text.Month.cycle[1]}}
+              <el-input-number size="small" v-model="month.rangeEnd" :min="1" :max="31"></el-input-number>
+              {{text.Month.cycle[2]}}
+            </el-radio>
+          </el-row>
+        </div>
+      </el-tab-pane>
+      <el-tab-pane>
+        <span slot="label"><i class="el-icon-date"></i> {{text.Week.name}}</span>
+        <div class="tabBody">
+          <el-row>
+            <el-radio v-model="week.cronEvery" label="1">{{text.Week.every}}</el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="week.cronEvery" label="2">{{text.Week.specific}}
+              <el-select size="small" multiple v-model="week.specificSpecific">
+                <el-option v-for="(val, $index) in text.Week.list" :key="val" :label="val"
+                           :value="($index + 1)"></el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio class="long" v-model="week.cronEvery" label="3">{{text.Week.specific}}
+              <el-select size="small" multiple v-model="week.specificSpecific">
+                <el-option v-for="val in 12" :key="val" :label="val" :value="val"></el-option>
+              </el-select>
+            </el-radio>
+          </el-row>
+          <el-row>
+            <el-radio v-model="week.cronEvery" label="4">{{text.Week.cycle[0]}}
+              <el-input-number size="small" v-model="week.rangeStart" :min="1" :max="7"></el-input-number>
+              {{text.Week.cycle[1]}}
+              <el-input-number size="small" v-model="week.rangeEnd" :min="1" :max="7"></el-input-number>
+            </el-radio>
+          </el-row>
+        </div>
+      </el-tab-pane>
+    </el-tabs>
+  </div>
+</template>
+<script>
+import Language from './language/index'
+
+export default {
+  name: 'VueCronLinux',
+  props: ['data', 'i18n'],
+  data () {
+    return {
+      second: {
+        cronEvery: '',
+        incrementStart: '3',
+        incrementIncrement: '5',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: []
+      },
+      minute: {
+        cronEvery: '',
+        incrementStart: '3',
+        incrementIncrement: '5',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: []
+      },
+      hour: {
+        cronEvery: '',
+        incrementStart: '3',
+        incrementIncrement: '5',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: []
+      },
+      day: {
+        cronEvery: '',
+        incrementStart: '1',
+        incrementIncrement: '1',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: [],
+        cronLastSpecificDomDay: 1,
+        cronDaysBeforeEomMinus: '',
+        cronDaysNearestWeekday: ''
+      },
+      week: {
+        cronEvery: '',
+        incrementStart: '1',
+        incrementIncrement: '1',
+        specificSpecific: [],
+        cronNthDayDay: 1,
+        cronNthDayNth: '1',
+        rangeStart: '',
+        rangeEnd: ''
+      },
+      month: {
+        cronEvery: '',
+        incrementStart: '3',
+        incrementIncrement: '5',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: []
+      },
+      year: {
+        cronEvery: '',
+        incrementStart: '2017',
+        incrementIncrement: '1',
+        rangeStart: '',
+        rangeEnd: '',
+        specificSpecific: []
+      },
+      output: {
+        second: '',
+        minute: '',
+        hour: '',
+        day: '',
+        month: '',
+        Week: '',
+        year: ''
+      }
+    }
+  },
+  watch: {
+    data () {
+      if (!this.data) {
+        this.data = '* * * * *'
+        return
+      }
+      this.updateCron()
+    },
+    cron () {
+      this.$emit('change', this.cron)
+    }
+  },
+  computed: {
+    text () {
+      return Language[this.i18n || 'cn']
+    },
+    minutesText () {
+      let minutes = ''
+      let cronEvery = this.minute.cronEvery
+      switch (cronEvery.toString()) {
+        case '1':
+          minutes = '*'
+          break
+        case '2':
+          minutes = this.minute.incrementStart + '/' + this.minute.incrementIncrement
+          break
+        case '3':
+          this.minute.specificSpecific.map(val => {
+            minutes += val + ','
+          })
+          minutes = minutes.slice(0, -1)
+          break
+        case '4':
+          minutes = this.minute.rangeStart + '-' + this.minute.rangeEnd
+          break
+      }
+      return minutes
+    },
+    hoursText () {
+      let hours = ''
+      let cronEvery = this.hour.cronEvery
+      switch (cronEvery.toString()) {
+        case '1':
+          hours = '*'
+          break
+        case '2':
+          hours = this.hour.incrementStart + '/' + this.hour.incrementIncrement
+          break
+        case '3':
+          this.hour.specificSpecific.map(val => {
+            hours += val + ','
+          })
+          hours = hours.slice(0, -1)
+          break
+        case '4':
+          hours = this.hour.rangeStart + '-' + this.hour.rangeEnd
+          break
+      }
+      return hours
+    },
+    daysText () {
+      let days = ''
+      let cronEvery = this.day.cronEvery
+      switch (cronEvery.toString()) {
+        case '1':
+          break
+        case '2':
+        case '4':
+        case '11':
+          days = '*'
+          break
+        case '3':
+          days = this.day.incrementStart + '/' + this.day.incrementIncrement
+          break
+        case '5':
+          this.day.specificSpecific.map(val => {
+            days += val + ','
+          })
+          days = days.slice(0, -1)
+          break
+        case '6':
+          days = 'L'
+          break
+        case '7':
+          days = 'LW'
+          break
+        case '8':
+          days = this.day.cronLastSpecificDomDay + 'L'
+          break
+        case '9':
+          days = 'L-' + this.day.cronDaysBeforeEomMinus
+          break
+        case '10':
+          days = this.day.cronDaysNearestWeekday + 'W'
+          break
+      }
+      return days
+    },
+    weeksText () {
+      let weeks = ''
+      let cronEvery = this.week.cronEvery
+      switch (cronEvery.toString()) {
+        case '1':
+          weeks = '*'
+          break
+        case '2':
+          this.week.specificSpecific.map(val => {
+            weeks += val + ','
+          })
+          weeks = weeks.slice(0, -1)
+          break
+      }
+      return weeks
+    },
+    monthsText () {
+      let months = ''
+      let cronEvery = this.month.cronEvery
+      switch (cronEvery.toString()) {
+        case '1':
+          months = '*'
+          break
+        case '2':
+          months = this.month.incrementStart + '/' + this.month.incrementIncrement
+          break
+        case '3':
+          this.month.specificSpecific.map(val => {
+            months += val + ','
+          })
+          months = months.slice(0, -1)
+          break
+        case '4':
+          months = this.month.rangeStart + '-' + this.month.rangeEnd
+          break
+      }
+      return months
+    },
+    cron () {
+      return `${this.minutesText || '*'} ${this.hoursText || '*'} ${this.daysText || '*'} ${this.monthsText || '*'} ${this.weeksText || '*'}`
+    }
+  },
+  methods: {
+    getValue () {
+      return this.cron
+    },
+    change () {
+      this.$emit('change', this.cron)
+      this.close()
+    },
+    close () {
+      this.$emit('close')
+    },
+    updateCronItem (key, value) {
+      if (value.match(/\//)) {
+        this[key].cronEvery = '2'
+        this[key].incrementStart = value.split('/')[0]
+        this[key].incrementIncrement = value.split('/')[1]
+      } else if (value.match(/,|^\d+$/)) {
+        this[key].cronEvery = '3'
+        this[key].specificSpecific = value.split(',')
+      } else if (value.match(/-/)) {
+        this[key].cronEvery = '4'
+        this[key].rangeStart = value.split('-')[0]
+        this[key].rangeEnd = value.split('-')[1]
+      } else {
+        this[key].cronEvery = '1'
+      }
+    },
+    updateCron () {
+      const arr = this.data.split(' ')
+      const minute = arr[0]
+      const hour = arr[1]
+      const day = arr[2]
+      const month = arr[3]
+      const week = arr[4]
+
+      this.updateCronItem('minute', minute)
+      this.updateCronItem('hour', hour)
+      this.updateCronItem('day', day)
+      this.updateCronItem('month', month)
+      this.updateCronItem('week', week)
+    }
+  },
+  mounted () {
+    this.updateCron()
+  }
+}</script>
diff --git a/frontend/src/components/Cron/language/cn.js b/frontend/src/components/Cron/language/cn.js
new file mode 100644
index 00000000..7c6df9f7
--- /dev/null
+++ b/frontend/src/components/Cron/language/cn.js
@@ -0,0 +1,61 @@
+export default {
+  Seconds: {
+    name: '秒',
+    every: '每一秒钟',
+    interval: ['每隔', '秒执行 从', '秒开始'],
+    specific: '具体秒数(可多选)',
+    cycle: ['周期从', '到', '秒']
+  },
+  Minutes: {
+    name: '分',
+    every: '每一分钟',
+    interval: ['每隔', '分执行 从', '分开始'],
+    specific: '具体分钟数(可多选)',
+    cycle: ['周期从', '到', '分']
+  },
+  Hours: {
+    name: '时',
+    every: '每一小时',
+    interval: ['每隔', '小时执行 从', '小时开始'],
+    specific: '具体小时数(可多选)',
+    cycle: ['周期从', '到', '小时']
+  },
+  Day: {
+    name: '天',
+    every: '每一天',
+    intervalWeek: ['每隔', '周执行 从', '开始'],
+    intervalDay: ['每隔', '天执行 从', '天开始'],
+    specificWeek: '具体星期几(可多选)',
+    specificDay: '具体天数(可多选)',
+    lastDay: '在这个月的最后一天',
+    lastWeekday: '在这个月的最后一个工作日',
+    lastWeek: ['在这个月的最后一个'],
+    beforeEndMonth: ['在本月底前', '天'],
+    nearestWeekday: ['最近的工作日（周一至周五）至本月', '日'],
+    someWeekday: ['在这个月的第', '个'],
+    cycle: ['从', '到']
+  },
+  Week: {
+    name: '周',
+    every: '每天',
+    specific: '具体天数(可多选)',
+    list: ['一', '二', '三', '四', '五', '六', '天'].map(val => '星期' + val),
+    cycle: ['从', '到']
+  },
+  Month: {
+    name: '月',
+    every: '每一月',
+    interval: ['每隔', '月执行 从', '月开始'],
+    specific: '具体月数(可多选)',
+    cycle: ['从', '到', '月之间的每个月']
+  },
+  Year: {
+    name: '年',
+    every: '每一年',
+    interval: ['每隔', '年执行 从', '年开始'],
+    specific: '具体年份(可多选)',
+    cycle: ['从', '到', '年之间的每一年']
+  },
+  Save: '保存',
+  Close: '关闭'
+}
diff --git a/frontend/src/components/Cron/language/en.js b/frontend/src/components/Cron/language/en.js
new file mode 100644
index 00000000..c645de47
--- /dev/null
+++ b/frontend/src/components/Cron/language/en.js
@@ -0,0 +1,62 @@
+export default {
+  Seconds: {
+    name: 'Seconds',
+    every: 'Every second',
+    interval: ['Every', 'second(s) starting at second'],
+    specific: 'Specific second (choose one or many)',
+    cycle: ['Every second between second', 'and second']
+  },
+  Minutes: {
+    name: 'Minutes',
+    every: 'Every minute',
+    interval: ['Every', 'minute(s) starting at minute'],
+    specific: 'Specific minute (choose one or many)',
+    cycle: ['Every minute between minute', 'and minute']
+  },
+  Hours: {
+    name: 'Hours',
+    every: 'Every hour',
+    interval: ['Every', 'hour(s) starting at hour'],
+    specific: 'Specific hour (choose one or many)',
+    cycle: ['Every hour between hour', 'and hour']
+  },
+  Day: {
+    name: 'Day',
+    every: 'Every day',
+    intervalWeek: ['Every', 'day(s) starting on'],
+    intervalDay: ['Every', 'day(s) starting at the', 'of the month'],
+    specificWeek: 'Specific day of week (choose one or many)',
+    specificDay: 'Specific day of month (choose one or many)',
+    lastDay: 'On the last day of the month',
+    lastWeekday: 'On the last weekday of the month',
+    lastWeek: ['On the last', ' of the month'],
+    beforeEndMonth: ['day(s) before the end of the month'],
+    nearestWeekday: ['Nearest weekday (Monday to Friday) to the', 'of the month'],
+    someWeekday: ['On the', 'of the month'],
+    cycle: ['From', 'to']
+  },
+  Week: {
+    name: 'Week',
+    every: 'Every day',
+    specific: 'Specific weekday (choose on or many)',
+    list: ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday'],
+    cycle: ['From', 'to']
+  },
+  // Week:['Sunday','Monday','Tuesday','Wednesday','Thursday','Friday','Saturday'],
+  Month: {
+    name: 'Month',
+    every: 'Every month',
+    interval: ['Every', 'month(s) starting in'],
+    specific: 'Specific month (choose one or many)',
+    cycle: ['Every month between', 'and']
+  },
+  Year: {
+    name: 'Year',
+    every: 'Any year',
+    interval: ['Every', 'year(s) starting in'],
+    specific: 'Specific year (choose one or many)',
+    cycle: ['Every year between', 'and']
+  },
+  Save: 'Save',
+  Close: 'Close'
+}
diff --git a/frontend/src/components/Cron/language/index.js b/frontend/src/components/Cron/language/index.js
new file mode 100644
index 00000000..1e592c13
--- /dev/null
+++ b/frontend/src/components/Cron/language/index.js
@@ -0,0 +1,7 @@
+import en from './en'
+import cn from './cn'
+
+export default {
+  en,
+  cn
+}
diff --git a/frontend/src/views/schedule/ScheduleList.vue b/frontend/src/views/schedule/ScheduleList.vue
index 69843a85..cd66231e 100644
--- a/frontend/src/views/schedule/ScheduleList.vue
+++ b/frontend/src/views/schedule/ScheduleList.vue
@@ -67,11 +67,18 @@
           </el-select>
         </el-form-item>
         <el-form-item :label="$t('Cron')" prop="cron" required>
-          <el-input
-            v-model="scheduleForm.cron"
-            :placeholder="`${$t('[minute] [hour] [day] [month] [day of week]')}`"
-          >
-          </el-input>
+          <el-popover>
+            <template>
+              <vue-cron-linux :data="scheduleForm.cron" :i18n="lang" @change="onCronChange"/>
+            </template>
+            <template slot="reference">
+              <el-input
+                v-model="scheduleForm.cron"
+                :placeholder="`${$t('[minute] [hour] [day] [month] [day of week]')}`"
+              >
+              </el-input>
+            </template>
+          </el-popover>
           <!--<el-button size="small" style="width:100px" type="primary" @click="onShowCronDialog">{{$t('schedules.add_cron')}}</el-button>-->
         </el-form-item>
         <el-form-item :label="$t('Execute Command')" prop="params">
@@ -139,14 +146,16 @@
               </el-tag>
             </template>
           </el-table-column>
-          <el-table-column v-else-if="col.name === 'run_type'" :key="col.name" :label="$t(col.label)" :width="col.width">
+          <el-table-column v-else-if="col.name === 'run_type'" :key="col.name" :label="$t(col.label)"
+                           :width="col.width">
             <template slot-scope="scope">
               <template v-if="scope.row.run_type === 'all-nodes'">{{$t('All Nodes')}}</template>
               <template v-else-if="scope.row.run_type === 'selected-nodes'">{{$t('Selected Nodes')}}</template>
               <template v-else-if="scope.row.run_type === 'random'">{{$t('Random')}}</template>
             </template>
           </el-table-column>
-          <el-table-column v-else-if="col.name === 'node_names'" :key="col.name" :label="$t(col.label)" :width="col.width">
+          <el-table-column v-else-if="col.name === 'node_names'" :key="col.name" :label="$t(col.label)"
+                           :width="col.width">
             <template slot-scope="scope">
               {{scope.row.nodes.map(d => d.name).join(', ')}}
             </template>
@@ -183,7 +192,7 @@
               <el-button type="danger" icon="el-icon-delete" size="mini" @click="onRemove(scope.row)"></el-button>
             </el-tooltip>
             <!--<el-tooltip :content="$t(getStatusTooltip(scope.row))" placement="top">-->
-              <!--<el-button type="success" icon="fa fa-bug" size="mini" @click="onCrawl(scope.row)"></el-button>-->
+            <!--<el-button type="success" icon="fa fa-bug" size="mini" @click="onCrawl(scope.row)"></el-button>-->
             <!--</el-tooltip>-->
           </template>
         </el-table-column>
@@ -194,14 +203,17 @@
 </template>
 
 <script>
-// import vcrontab from 'vcrontab'
 import request from '../../api/request'
+import VueCronLinux from '../../components/Cron'
 import {
   mapState
 } from 'vuex'
 
 export default {
   name: 'ScheduleList',
+  components: {
+    VueCronLinux
+  },
   data () {
     return {
       columns: [
@@ -229,6 +241,12 @@ export default {
       'scheduleList',
       'scheduleForm'
     ]),
+    lang () {
+      const lang = this.$store.state.lang.lang || window.localStorage.getItem('lang')
+      if (!lang) return 'cn'
+      if (lang === 'zh') return 'cn'
+      return 'en'
+    },
     filteredTableData () {
       return this.scheduleList
     },
@@ -340,6 +358,10 @@ export default {
       } else {
         this.$message.success(this.$t(`${row.enabled ? 'Enabling' : 'Disabling'} the schedule successful`))
       }
+    },
+    onCronChange (value) {
+      console.log(value)
+      this.$set(this.scheduleForm, 'cron', value)
     }
   },
   created () {
diff --git a/frontend/yarn.lock b/frontend/yarn.lock
index b885732c..aac06724 100644
--- a/frontend/yarn.lock
+++ b/frontend/yarn.lock
@@ -1508,6 +1508,15 @@ babel-plugin-transform-strict-mode@^6.24.1:
     babel-runtime "^6.22.0"
     babel-types "^6.24.1"
 
+babel-polyfill@^6.26.0:
+  version "6.26.0"
+  resolved "https://registry.npm.taobao.org/babel-polyfill/download/babel-polyfill-6.26.0.tgz#379937abc67d7895970adc621f284cd966cf2153"
+  integrity sha1-N5k3q8Z9eJWXCtxiHyhM2WbPIVM=
+  dependencies:
+    babel-runtime "^6.26.0"
+    core-js "^2.5.0"
+    regenerator-runtime "^0.10.5"
+
 babel-preset-jest@^23.2.0:
   version "23.2.0"
   resolved "http://registry.npm.taobao.org/babel-preset-jest/download/babel-preset-jest-23.2.0.tgz#8ec7a03a138f001a1a8fb1e8113652bf1a55da46"
@@ -7139,6 +7148,11 @@ regenerate@^1.2.1, regenerate@^1.4.0:
   version "1.4.0"
   resolved "http://registry.npm.taobao.org/regenerate/download/regenerate-1.4.0.tgz#4a856ec4b56e4077c557589cae85e7a4c8869a11"
 
+regenerator-runtime@^0.10.5:
+  version "0.10.5"
+  resolved "https://registry.npm.taobao.org/regenerator-runtime/download/regenerator-runtime-0.10.5.tgz#336c3efc1220adcedda2c9fab67b5a7955a33658"
+  integrity sha1-M2w+/BIgrc7dosn6tntaeVWjNlg=
+
 regenerator-runtime@^0.11.0:
   version "0.11.1"
   resolved "http://registry.npm.taobao.org/regenerator-runtime/download/regenerator-runtime-0.11.1.tgz#be05ad7f9bf7d22e056f9726cee5017fbf19e2e9"

From 637ece7cfe744aa9c9eb6ca8530b20eb71a2d939 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 14:49:14 +0800
Subject: [PATCH 29/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 1 +
 CHANGELOG.md    | 1 +
 2 files changed, 2 insertions(+)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index 0b877c5b..5b761863 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -7,6 +7,7 @@
 - **API 地址优化**. 在前端加入相对路径，因此用户不需要特别注明 `CRAWLAB_API_ADDRESS`.
 - **SDK 兼容**. 允许用户通过 Crawlab SDK 与 Scrapy 或通用爬虫集成.
 - **优化文件管理**. 加入树状文件侧边栏，让用户更方便的编辑文件.
+- **高级定时任务 Cron**. 允许用户通过 Cron 可视化编辑器编辑定时任务.
 
 ### Bug 修复
 - **`nil retuened` 错误**.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 19dad6da..584a6fae 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,7 @@
 - **API Address Optimization**. Added relative URL path in frontend so that users don't have to specify `CRAWLAB_API_ADDRESS` explicitly.
 - **SDK Compatiblity**. Allow users to integrate Scrapy or general spiders with Crawlab SDK.
 - **Enhanced File Management**. Added tree-like file sidebar to allow users to edit files much more easier.
+- **Advanced Schedule Cron**. Allow users to edit schedule cron with visualized cron editor.
 
 ### Bug Fixes
 - **`nil retuened` error**.

From b2aa88a72e586eb2c22e093e6c922fd98925f4a6 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 14:51:21 +0800
Subject: [PATCH 30/42] updated version

---
 backend/conf/config.yml | 2 +-
 frontend/package.json   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/backend/conf/config.yml b/backend/conf/config.yml
index 6e3e0611..90fa8fe1 100644
--- a/backend/conf/config.yml
+++ b/backend/conf/config.yml
@@ -35,7 +35,7 @@ task:
   workers: 4
 other:
   tmppath: "/tmp"
-version: 0.4.3
+version: 0.4.4
 setting:
   allowRegister: "N"
 notification:
diff --git a/frontend/package.json b/frontend/package.json
index 36a6abdb..64d80294 100644
--- a/frontend/package.json
+++ b/frontend/package.json
@@ -1,6 +1,6 @@
 {
   "name": "crawlab",
-  "version": "0.4.3",
+  "version": "0.4.4",
   "private": true,
   "scripts": {
     "serve": "vue-cli-service serve --ip=0.0.0.0 --mode=development",

From 7d27f014cfad3ceb401100e6a00772e09af6cb2b Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 15:32:48 +0800
Subject: [PATCH 31/42] updated README

---
 README-zh.md | 4 ++--
 README.md    | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/README-zh.md b/README-zh.md
index 57f4af47..85cfe053 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -138,9 +138,9 @@ Docker部署的详情，请见[相关文档](https://tikazyq.github.io/crawlab-d
 
 ![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/spider-analytics.png)
 
-#### 爬虫文件
+#### 爬虫文件编辑
 
-![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/spider-file.png)
+![](http://static-docs.crawlab.cn/file-edit.png)
 
 #### 任务详情 - 抓取结果
 
diff --git a/README.md b/README.md
index 3f6224a8..190e6eb1 100644
--- a/README.md
+++ b/README.md
@@ -136,9 +136,9 @@ For Docker Deployment details, please refer to [relevant documentation](https://
 
 ![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/spider-analytics.png)
 
-#### Spider Files
+#### Spider File Edit
 
-![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/spider-file.png)
+![](http://static-docs.crawlab.cn/file-edit.png)
 
 #### Task Results
 

From 356b5c056ba5bd0f1224be9e0bc4201e565d8ccf Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 16:03:04 +0800
Subject: [PATCH 32/42] =?UTF-8?q?=E6=9B=B4=E6=96=B0=E7=BB=9F=E8=AE=A1?=
 =?UTF-8?q?=E5=9F=8B=E7=82=B9?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue    |  1 +
 frontend/src/views/schedule/ScheduleList.vue | 12 ++----------
 frontend/src/views/setting/Setting.vue       |  4 +---
 3 files changed, 4 insertions(+), 13 deletions(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index 26ed393a..c3f47aa1 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -325,6 +325,7 @@ export default {
       this.isShowCreatePopoverDict = {}
       this.$set(this.isShowCreatePopoverDict, data.path, true)
       this.activeFileNode = data
+      this.$st.sendEv('爬虫详情', '文件', '右键点击导航栏')
     },
     onHideCreate (data) {
       this.$set(this.isShowCreatePopoverDict, data.path, false)
diff --git a/frontend/src/views/schedule/ScheduleList.vue b/frontend/src/views/schedule/ScheduleList.vue
index cd66231e..c1955dd5 100644
--- a/frontend/src/views/schedule/ScheduleList.vue
+++ b/frontend/src/views/schedule/ScheduleList.vue
@@ -337,15 +337,6 @@ export default {
         return false
       }
     },
-    getStatusTooltip (row) {
-      if (row.status === 'stop') {
-        return 'Start'
-      } else if (row.status === 'running') {
-        return 'Stop'
-      } else if (row.status === 'error') {
-        return 'Start'
-      }
-    },
     async onEnabledChange (row) {
       let res
       if (row.enabled) {
@@ -358,10 +349,11 @@ export default {
       } else {
         this.$message.success(this.$t(`${row.enabled ? 'Enabling' : 'Disabling'} the schedule successful`))
       }
+      this.$st.sendEv('定时任务', '启用/禁用')
     },
     onCronChange (value) {
-      console.log(value)
       this.$set(this.scheduleForm, 'cron', value)
+      this.$st.sendEv('定时任务', '配置Cron')
     }
   },
   created () {
diff --git a/frontend/src/views/setting/Setting.vue b/frontend/src/views/setting/Setting.vue
index 75dc856f..e82d47b6 100644
--- a/frontend/src/views/setting/Setting.vue
+++ b/frontend/src/views/setting/Setting.vue
@@ -113,9 +113,7 @@ export default {
           this.$message.success(this.$t('Saved successfully'))
         }
       })
-    },
-    toggleDingTalkAppSecret () {
-      this.isShowDingTalkAppSecret = !this.isShowDingTalkAppSecret
+      this.$st.sendEv('设置', '保存')
     }
   },
   async created () {

From 1d0b9006cce27c6d918b7fff14d7b667294f627e Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 17:02:44 +0800
Subject: [PATCH 33/42] updated README

---
 README-zh.md | 4 ++++
 README.md    | 4 ++++
 2 files changed, 8 insertions(+)

diff --git a/README-zh.md b/README-zh.md
index 85cfe053..61d586be 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -154,6 +154,10 @@ Docker部署的详情，请见[相关文档](https://tikazyq.github.io/crawlab-d
 
 ![](http://static-docs.crawlab.cn/node-install-dependencies.png)
 
+#### 消息通知
+
+![](http://static-docs.crawlab.cn/notification-mail.jpeg)
+
 ## 架构
 
 Crawlab的架构包括了一个主节点（Master Node）和多个工作节点（Worker Node），以及负责通信和数据储存的Redis和MongoDB数据库。
diff --git a/README.md b/README.md
index 190e6eb1..7cbc876e 100644
--- a/README.md
+++ b/README.md
@@ -148,6 +148,10 @@ For Docker Deployment details, please refer to [relevant documentation](https://
 
 ![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/schedule.png)
 
+#### Notifications
+
+![](http://static-docs.crawlab.cn/notification-mail.jpeg)
+
 #### Dependency Installation
 
 ![](http://static-docs.crawlab.cn/node-install-dependencies.png)

From 65353f9d81bda45339421be95ea225ce6e67fcc6 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 17:08:37 +0800
Subject: [PATCH 34/42] updated README

---
 README-zh.md | 2 +-
 README.md    | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/README-zh.md b/README-zh.md
index 61d586be..89f7df5e 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -156,7 +156,7 @@ Docker部署的详情，请见[相关文档](https://tikazyq.github.io/crawlab-d
 
 #### 消息通知
 
-![](http://static-docs.crawlab.cn/notification-mail.jpeg)
+<img src="http://static-docs.crawlab.cn/notification-mobile.jpeg" height="480px">
 
 ## 架构
 
diff --git a/README.md b/README.md
index 7cbc876e..b40a7dab 100644
--- a/README.md
+++ b/README.md
@@ -148,14 +148,14 @@ For Docker Deployment details, please refer to [relevant documentation](https://
 
 ![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/schedule.png)
 
-#### Notifications
-
-![](http://static-docs.crawlab.cn/notification-mail.jpeg)
-
 #### Dependency Installation
 
 ![](http://static-docs.crawlab.cn/node-install-dependencies.png)
 
+#### Notifications
+
+<img src="http://static-docs.crawlab.cn/notification-mobile.jpeg" height="480px">
+
 ## Architecture
 
 The architecture of Crawlab is consisted of the Master Node and multiple Worker Nodes, and Redis and MongoDB databases which are mainly for nodes communication and data storage.

From d6f96f1a2495135f1eb6e077ac50f0ad3df69834 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 17:10:47 +0800
Subject: [PATCH 35/42] updated README

---
 README-zh.md | 2 +-
 README.md    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/README-zh.md b/README-zh.md
index 89f7df5e..e2f1e412 100644
--- a/README-zh.md
+++ b/README-zh.md
@@ -148,7 +148,7 @@ Docker部署的详情，请见[相关文档](https://tikazyq.github.io/crawlab-d
 
 #### 定时任务
 
-![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/schedule.png)
+![](http://static-docs.crawlab.cn/schedule-v0.4.4.png)
 
 #### 依赖安装
 
diff --git a/README.md b/README.md
index b40a7dab..81ba33d5 100644
--- a/README.md
+++ b/README.md
@@ -146,7 +146,7 @@ For Docker Deployment details, please refer to [relevant documentation](https://
 
 #### Cron Job
 
-![](https://raw.githubusercontent.com/tikazyq/crawlab-docs/master/images/schedule.png)
+![](http://static-docs.crawlab.cn/schedule-v0.4.4.png)
 
 #### Dependency Installation
 

From 21ade6dca271560cbacea416cab9fad8dddd7fc7 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Thu, 16 Jan 2020 17:15:06 +0800
Subject: [PATCH 36/42] updated CHANGELOG

---
 CHANGELOG-zh.md | 2 +-
 CHANGELOG.md    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index 5b761863..d68a833f 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -1,4 +1,4 @@
-# 0.4.4 (未知)
+# 0.4.4 (2020-01-17)
 
 ### 功能 / 优化
 - **邮件通知**. 允许用户发送邮件消息通知.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 584a6fae..2b6277e2 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,4 +1,4 @@
-# 0.4.4 (unknown)
+# 0.4.4 (2020-01-17)
 ### Features / Enhancement
 - **Email Notification**. Allow users to send email notifications.
 - **DingTalk Robot Notification**. Allow users to send DingTalk Robot notifications.

From 7c09793ab62d704b8137d47941864b8be4d3fa36 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 10:15:55 +0800
Subject: [PATCH 37/42] =?UTF-8?q?=E4=BF=AE=E5=A4=8D=E6=96=87=E4=BB=B6?=
 =?UTF-8?q?=E5=AF=BC=E8=88=AA=E6=A0=8F=E8=87=AA=E5=8A=A8=E6=90=9C=E7=B4=A2?=
 =?UTF-8?q?bug?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue | 36 ++++++++++++++++++++++-
 1 file changed, 35 insertions(+), 1 deletion(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index c3f47aa1..d5c0b6ae 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -33,10 +33,14 @@
     <div class="file-tree-wrapper">
       <el-tree
         :data="computedFileTree"
+        ref="tree"
         node-key="path"
         :highlight-current="true"
+        :default-expanded-keys="expandedPaths"
         @node-contextmenu="onFileRightClick"
         @node-click="onFileClick"
+        @node-expand="onDirClick"
+        @node-collapse="onDirClick"
       >
         <span class="custom-tree-node" slot-scope="{ node, data }">
           <el-popover v-model="isShowCreatePopoverDict[data.path]" trigger="manual" placement="right"
@@ -167,7 +171,8 @@ export default {
       ],
       activeFileNode: {},
       dirDialogVisible: false,
-      fileDialogVisible: false
+      fileDialogVisible: false,
+      nodeExpandedDict: {}
     }
   },
   computed: {
@@ -189,7 +194,19 @@ export default {
       if (!this.fileTree || !this.fileTree.children) return []
       let nodes = this.sortFiles(this.fileTree.children)
       nodes = this.filterFiles(nodes)
+      nodes = this.computeExpanded(nodes)
       return nodes
+    },
+    expandedPaths () {
+      return Object.keys(this.nodeExpandedDict)
+        .map(path => {
+          return {
+            path,
+            expanded: this.nodeExpandedDict[path]
+          }
+        })
+        .filter(d => d.expanded)
+        .map(d => d.path)
     }
   },
   methods: {
@@ -289,6 +306,13 @@ export default {
       this.currentFilePath = data.path
       this.onItemClick(data)
     },
+    onDirClick (data, node) {
+      const vm = this
+      setTimeout(() => {
+        console.log(data.path, node.expanded)
+        vm.$set(vm.nodeExpandedDict, data.path, node.expanded)
+      }, 0)
+    },
     sortFiles (nodes) {
       nodes.forEach(node => {
         if (node.is_dir) {
@@ -318,6 +342,16 @@ export default {
         return true
       })
     },
+    computeExpanded (nodes) {
+      return nodes.map(node => {
+        if (node.is_dir) {
+          node.children = this.computeExpanded(node.children)
+        } else {
+          node.expanded = !!this.nodeExpandedDict[node.path]
+        }
+        return node
+      })
+    },
     isActiveFile (node) {
       return node.path === this.currentFilePath
     },

From bf06fa9bc5ee02aa7d356265df8cb51bb6b6edb5 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 10:16:56 +0800
Subject: [PATCH 38/42] =?UTF-8?q?=E4=BF=AE=E5=A4=8D=E6=96=87=E4=BB=B6?=
 =?UTF-8?q?=E5=AF=BC=E8=88=AA=E6=A0=8F=E8=87=AA=E5=8A=A8=E6=94=B6=E7=BC=A9?=
 =?UTF-8?q?bug?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 frontend/src/components/File/FileList.vue | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/frontend/src/components/File/FileList.vue b/frontend/src/components/File/FileList.vue
index d5c0b6ae..4bf76395 100644
--- a/frontend/src/components/File/FileList.vue
+++ b/frontend/src/components/File/FileList.vue
@@ -194,7 +194,6 @@ export default {
       if (!this.fileTree || !this.fileTree.children) return []
       let nodes = this.sortFiles(this.fileTree.children)
       nodes = this.filterFiles(nodes)
-      nodes = this.computeExpanded(nodes)
       return nodes
     },
     expandedPaths () {
@@ -309,7 +308,6 @@ export default {
     onDirClick (data, node) {
       const vm = this
       setTimeout(() => {
-        console.log(data.path, node.expanded)
         vm.$set(vm.nodeExpandedDict, data.path, node.expanded)
       }, 0)
     },
@@ -342,16 +340,6 @@ export default {
         return true
       })
     },
-    computeExpanded (nodes) {
-      return nodes.map(node => {
-        if (node.is_dir) {
-          node.children = this.computeExpanded(node.children)
-        } else {
-          node.expanded = !!this.nodeExpandedDict[node.path]
-        }
-        return node
-      })
-    },
     isActiveFile (node) {
       return node.path === this.currentFilePath
     },

From c17a05fe2f8ad3fa4e76c101328b9a66b2db5fa9 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 13:43:29 +0800
Subject: [PATCH 39/42] updated readme

---
 docker-compose.yml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/docker-compose.yml b/docker-compose.yml
index c8d44753..c6c92339 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -24,6 +24,12 @@ services:
       # CRAWLAB_TASK_WORKERS: 4  # number of task executors 任务执行器个数（并行执行任务数）
       # CRAWLAB_SERVER_LANG_NODE: "Y"  # whether to pre-install Node.js 预安装 Node.js 语言环境
       # CRAWLAB_SETTING_ALLOWREGISTER: "N"  # whether to allow user registration 是否允许用户注册
+      # CRAWLAB_NOTIFICATION_MAIL_SERVER: smtp.exmaple.com  # STMP server address STMP 服务器地址
+      # CRAWLAB_NOTIFICATION_MAIL_PORT: 465  # STMP server port STMP 服务器端口
+      # CRAWLAB_NOTIFICATION_MAIL_SENDEREMAIL: admin@exmaple.com  # sender email 发送者邮箱
+      # CRAWLAB_NOTIFICATION_MAIL_SENDEREIDENTITY: admin@exmaple.com  # sender ID 发送者 ID
+      # CRAWLAB_NOTIFICATION_MAIL_SMTP_USER: username  # SMTP username SMTP 用户名
+      # CRAWLAB_NOTIFICATION_MAIL_SMTP_PASSWORD: password  # SMTP password SMTP 密码
     ports:    
       - "8080:8080" # frontend port mapping 前端端口映射
     depends_on:

From 128df1c4a2f26e1f93915908c7cd98610841bcd0 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 13:52:52 +0800
Subject: [PATCH 40/42] fixed unable to click crawl button on spider page for
 configurable spiders

---
 frontend/src/views/spider/SpiderList.vue |  7 ++-----
 test/README.md                           | 10 ++++++++++
 2 files changed, 12 insertions(+), 5 deletions(-)
 create mode 100644 test/README.md

diff --git a/frontend/src/views/spider/SpiderList.vue b/frontend/src/views/spider/SpiderList.vue
index 7275e010..9668d8df 100644
--- a/frontend/src/views/spider/SpiderList.vue
+++ b/frontend/src/views/spider/SpiderList.vue
@@ -508,11 +508,8 @@ export default {
       this.dialogVisible = true
     },
     isShowRun (row) {
-      if (row.cmd) {
-        return true
-      } else {
-        return false
-      }
+      if (!this.isCustomized(row)) return true
+      return !!row.cmd
     },
     isCustomized (row) {
       return row.type === 'customized'
diff --git a/test/README.md b/test/README.md
new file mode 100644
index 00000000..392ebe76
--- /dev/null
+++ b/test/README.md
@@ -0,0 +1,10 @@
+# Test for Crawlab
+
+## Test Cases
+
+Test Case | Steps | Expected Results
+--- | --- | ---
+Login | |
+Signup | |
+Upload Spider | |
+Edit Spider File | |

From 4a3d48dde60448a9ec9cc29065e2df409781f91a Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 14:12:53 +0800
Subject: [PATCH 41/42] fixed missing form validation before uploading spider
 file error

---
 frontend/src/views/spider/SpiderList.vue | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/frontend/src/views/spider/SpiderList.vue b/frontend/src/views/spider/SpiderList.vue
index 9668d8df..def443b7 100644
--- a/frontend/src/views/spider/SpiderList.vue
+++ b/frontend/src/views/spider/SpiderList.vue
@@ -55,7 +55,9 @@
                 :data="uploadForm"
                 :headers="{Authorization:token}"
                 :on-success="onUploadSuccess"
-                :file-list="fileList">
+                :file-list="fileList"
+                :before-upload="beforeUpload"
+              >
                 <el-button size="small" type="primary" icon="el-icon-upload">
                   {{$t('Upload')}}
                 </el-button>
@@ -557,6 +559,17 @@ export default {
       // navigate to spider detail
       this.$router.push(`/spiders/${res.data._id}`)
     },
+    beforeUpload (file) {
+      return new Promise((resolve, reject) => {
+        this.$refs['addCustomizedForm'].validate(res => {
+          if (res) {
+            resolve()
+          } else {
+            reject(new Error('form validation error'))
+          }
+        })
+      })
+    },
     getTime (str) {
       if (!str || str.match('^0001')) return 'NA'
       return dayjs(str).format('YYYY-MM-DD HH:mm:ss')

From ca2e77970874cb445eb11c69a179bc08f9ad6a62 Mon Sep 17 00:00:00 2001
From: marvzhang <tikazyq@163.com>
Date: Fri, 17 Jan 2020 14:14:30 +0800
Subject: [PATCH 42/42] updated CHANGELOG

---
 .gitignore      | 3 ++-
 CHANGELOG-zh.md | 2 ++
 CHANGELOG.md    | 2 ++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/.gitignore b/.gitignore
index 2c38a7c2..0b6328c9 100644
--- a/.gitignore
+++ b/.gitignore
@@ -121,4 +121,5 @@ _book/
 .idea
 *.lock
 
-backend/spiders
\ No newline at end of file
+backend/spiders
+spiders/*.zip
diff --git a/CHANGELOG-zh.md b/CHANGELOG-zh.md
index d68a833f..d0f85f2a 100644
--- a/CHANGELOG-zh.md
+++ b/CHANGELOG-zh.md
@@ -12,6 +12,8 @@
 ### Bug 修复
 - **`nil retuened` 错误**.
 - **使用 HTTPS 出现的报错**.
+- **无法在爬虫列表页运行可配置爬虫**.
+- **上传爬虫文件缺少表单验证**.
 
 # 0.4.3 (2020-01-07)
 
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2b6277e2..c7d5f6e5 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -11,6 +11,8 @@
 ### Bug Fixes
 - **`nil retuened` error**.
 - **Error when using HTTPS**.
+- **Unable to run Configurable Spiders on Spider List**.
+- **Missing form validation before uploading spider files**.
 
 # 0.4.3 (2020-01-07)