git-sync/crawlab
1
0
Fork 0
mirror of https://github.com/crawlab-team/crawlab.git synced 2026-01-26 17:49:15 +01:00
Code Issues Packages Projects Releases Wiki Activity

优化可配置爬虫前端界面

Browse Source
This commit is contained in:
marvzhang
2019-11-28 21:36:17 +08:00
parent bc81462a80
commit ba4f5fd20c
11 changed files with 186 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
backend/model/config_spider/scrapy.go
View File

@@ -216,7 +216,7 @@ func (g ScrapyGenerator) GetExtractStringFromField(f entity.Field) string {
// 如果为CSS
if f.Attr == "" {
// 文本
return fmt.Sprintf(`css('%s::text()')`, f.Css)
return fmt.Sprintf(`css('%s::text')`, f.Css)
} else {
// 属性
return fmt.Sprintf(`css('%s::attr("%s")')`, f.Css, f.Attr)
@@ -242,9 +242,9 @@ func (g ScrapyGenerator) GetExtractStringFromStage(stage entity.Stage) string {
if stage.PageCss != "" {
// 如果为CSS
return fmt.Sprintf(`css(%s::attr("%s"))`, stage.PageCss, pageAttr)
return fmt.Sprintf(`css('%s::attr("%s")')`, stage.PageCss, pageAttr)
} else {
// 如果为XPath
return fmt.Sprintf(`xpath(%s/@%s)`, stage.PageXpath, pageAttr)
return fmt.Sprintf(`xpath('%s/@%s')`, stage.PageXpath, pageAttr)
}
}