crawlab/.github/workflows/docker-crawlab.yml

name: "Docker Image CI: crawlab"

on:
  push:
    branches: [ develop, main ]
  release:
    types: [ published ]
  workflow_dispatch:
  repository_dispatch:
    types: [ docker-crawlab ]

env:
  IMAGE_PATH_CRAWLAB_BACKEND: backend
  IMAGE_PATH_CRAWLAB_FRONTEND: frontend
  GH_PKG_NAME_CRAWLAB_BACKEND: ${{ github.repository_owner }}/crawlab-backend
  GH_PKG_NAME_CRAWLAB_FRONTEND: ${{ github.repository_owner }}/crawlab-frontend
  GH_PKG_NAME_CRAWLAB: ${{ github.repository_owner }}/crawlab
  IMAGE_NAME_CRAWLAB_BACKEND: ghcr.io/${{ github.repository_owner }}/crawlab-backend
  IMAGE_NAME_CRAWLAB_FRONTEND: ghcr.io/${{ github.repository_owner }}/crawlab-frontend
  IMAGE_NAME_CRAWLAB: ghcr.io/${{ github.repository_owner }}/crawlab
  IMAGE_NAME_CRAWLAB_DOCKERHUB: crawlabteam/crawlab
  IMAGE_NAME_CRAWLAB_TENCENT: ccr.ccs.tencentyun.com/crawlab/crawlab
  E2E_TESTS_IMAGE_NAME: ghcr.io/${{ github.repository_owner }}/e2e-tests

jobs:
  setup:
    runs-on: ubuntu-latest
    outputs:
      backend_changed: ${{ steps.check_changed_files.outputs.backend_changed }}
      frontend_changed: ${{ steps.check_changed_files.outputs.frontend_changed }}
      dockerfile_changed: ${{ steps.check_changed_files.outputs.dockerfile_changed }}
      workflow_changed: ${{ steps.check_changed_files.outputs.workflow_changed }}
      version: ${{ steps.version.outputs.version }}

    steps:
      - uses: actions/checkout@v4

      - name: Get changed files
        id: changed_files
        uses: tj-actions/changed-files@v45
        with:
          files_yaml: |
            backend:
              - 'backend/**'
              - 'core/**'
              - 'db/**'
              - 'fs/**'
              - 'grpc/**'
              - 'vcs/**'
              - 'trace/**'
            frontend:
              - '${{ env.IMAGE_PATH_CRAWLAB_FRONTEND }}/**'
            dockerfile:
              - 'Dockerfile'
            workflow:
              - '.github/workflows/docker-crawlab.yml'

      - id: check_changed_files
        name: Check changed files
        run: |
          # set outputs
          if [[ "${{ github.ref }}" == "refs/tags/"* ]]; then
            echo "backend_changed=true" >> $GITHUB_OUTPUT
            echo "frontend_changed=true" >> $GITHUB_OUTPUT
            echo "dockerfile_changed=true" >> $GITHUB_OUTPUT
            echo "workflow_changed=true" >> $GITHUB_OUTPUT
          else
            echo "backend_changed=${{ steps.changed_files.outputs.backend_any_changed }}" >> $GITHUB_OUTPUT
            echo "frontend_changed=${{ steps.changed_files.outputs.frontend_any_changed }}" >> $GITHUB_OUTPUT
            echo "dockerfile_changed=${{ steps.changed_files.outputs.dockerfile_any_changed }}" >> $GITHUB_OUTPUT
            echo "workflow_changed=${{ steps.changed_files.outputs.workflow_any_changed }}" >> $GITHUB_OUTPUT
          fi
          
          # Display change status for each component
          echo "Backend changed: ${{ steps.changed_files.outputs.backend_any_changed }}"
          echo "Frontend changed: ${{ steps.changed_files.outputs.frontend_any_changed }}"
          echo "Dockerfile changed: ${{ steps.changed_files.outputs.dockerfile_any_changed }}"
          echo "Workflow changed: ${{ steps.changed_files.outputs.workflow_any_changed }}"

      - id: version
        name: Get version
        run: |
          # Strip git ref prefix from version
          VERSION=$(echo "${{ github.ref }}" | sed -e 's,.*/\(.*\),\1,')
          
          # Strip "v" prefix from tag name
          [[ "${{ github.ref }}" == "refs/tags/"* ]] && VERSION=$(echo $VERSION | sed -e 's/^v//')
          
          # Use Docker `latest` tag convention
          [ "$VERSION" == "main" ] && VERSION=latest
          
          echo "version=$VERSION" >> $GITHUB_OUTPUT

      - id: hash
        name: Generate unique hash
        run: echo "hash=$(echo ${{ github.sha }} | cut -c1-7)" >> $GITHUB_OUTPUT

  test_backend:
    needs: [ setup ]
    if: needs.setup.outputs.backend_changed == 'true' || needs.setup.outputs.workflow_changed == 'true'
    runs-on: ubuntu-latest
    services:
      mongo:
        image: mongo:5
        ports:
          - 27017:27017
    steps:
      - uses: actions/checkout@v4

      - name: Set up Go
        uses: actions/setup-go@v5
        with:
          go-version-file: 'core/go.mod'
          cache-dependency-path: 'core/go.sum'

      - name: Run tests
        working-directory: core
        run: |
          # Find all directories containing *_test.go files
          test_dirs=$(find . -name "*_test.go" -exec dirname {} \; | sort -u)

          # Run go test on each directory
          for dir in $test_dirs
          do
            echo "Running tests in $dir"
            go test ./$dir
          done

  build_backend:
    needs: [ setup, test_backend ]
    if: needs.test_backend.result == 'success' || needs.setup.outputs.workflow_changed == 'true'
    runs-on: ubuntu-latest
    outputs:
      failed: ${{ steps.set_output.outputs.failed }}
    steps:
      - uses: actions/checkout@v4
      - name: Log in to GitHub Container Registry
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}
      - name: Build and push image
        uses: docker/build-push-action@v5
        with:
          context: ${{ env.IMAGE_PATH_CRAWLAB_BACKEND }}
          push: true
          tags: ${{ env.IMAGE_NAME_CRAWLAB_BACKEND }}:${{ needs.setup.outputs.version }}
      - name: Set output
        id: set_output
        if: failure()
        run: echo "failed=true" >> $GITHUB_OUTPUT

  build_frontend:
    needs: [ setup ]
    if: needs.setup.outputs.frontend_changed == 'true' || needs.setup.outputs.workflow_changed == 'true'
    runs-on: ubuntu-latest
    outputs:
      failed: ${{ steps.set_output.outputs.failed }}
    steps:
      - uses: actions/checkout@v4
        with:
          submodules: 'recursive'
      - name: Log in to GitHub Container Registry
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}
      - name: Build and push image
        uses: docker/build-push-action@v5
        with:
          context: ${{ env.IMAGE_PATH_CRAWLAB_FRONTEND }}
          push: true
          tags: ${{ env.IMAGE_NAME_CRAWLAB_FRONTEND }}:${{ needs.setup.outputs.version }}
      - name: Set output
        id: set_output
        if: failure()
        run: echo "failed=true" >> $GITHUB_OUTPUT

  build_crawlab:
    needs: [setup, build_backend, build_frontend]
    if: |
      always() &&
      (needs.build_backend.outputs.failed != 'true' && needs.build_frontend.outputs.failed != 'true') &&
      (
        needs.setup.outputs.backend_changed == 'true' ||
        needs.setup.outputs.frontend_changed == 'true' ||
        needs.setup.outputs.dockerfile_changed == 'true' ||
        needs.setup.outputs.workflow_changed == 'true'
      )
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Update Dockerfile
        run: |
          for name in crawlab-backend crawlab-frontend; do
            IMAGE_NAME=ghcr.io/${{ github.repository_owner }}/$name
            sed -i "s|crawlabteam/${name}:latest|${IMAGE_NAME}:${{ needs.setup.outputs.version }}|" Dockerfile
          done

      - name: Log in to GitHub Container Registry
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Build and push image
        uses: docker/build-push-action@v5
        with:
          context: .
          file: ./Dockerfile
          push: true
          tags: |
            ${{ env.IMAGE_NAME_CRAWLAB }}:${{ needs.setup.outputs.version }}

  test_crawlab:
    needs: [setup, build_crawlab]
    if: ${{ always() && needs.build_crawlab.result == 'success' }}
    runs-on: ubuntu-latest
    services:
      mongo:
        image: mongo:5
        options: >-
          --health-cmd "mongosh --eval 'db.adminCommand(\"ping\")' || exit 1"
          --health-interval 10s
          --health-timeout 5s
          --health-retries 5
      crawlab:
        image: ghcr.io/${{ github.repository_owner }}/crawlab:${{ needs.setup.outputs.version }}
        env:
          CRAWLAB_NODE_MASTER: Y
          CRAWLAB_MONGO_HOST: mongo 
          CRAWLAB_MONGO_PORT: 27017
        ports:
          - 8080:8080
        options: >-
          --health-cmd "curl -f http://localhost:8080/api/system-info || exit 1"
          --health-interval 30s
          --health-timeout 10s
          --health-retries 5
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4

      - name: Log in to GitHub Container Registry
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Pull and run e2e test image
        run: |
          docker pull ${{ env.E2E_TESTS_IMAGE_NAME }}:latest
          docker run --network host \
            -e BASE_URL=http://localhost:8080 \
            ${{ env.E2E_TESTS_IMAGE_NAME }}:latest

      - name: Copy test results
        if: always()
        run: |
          docker cp $(docker ps -aq -f ancestor=${{ env.E2E_TESTS_IMAGE_NAME }}:latest):/app/playwright-report ./playwright-report

      - name: Deploy to GitHub Pages
        if: always()
        uses: peaceiris/actions-gh-pages@v3
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}
          publish_dir: ./playwright-report
          destination_dir: playwright-report/${{ needs.setup.outputs.version }}

  push_images:
    if: ${{ always() && needs.test_crawlab.result == 'success' }}
    needs: [setup, test_crawlab]
    runs-on: ubuntu-latest
    strategy:
      matrix:
        registry: [dockerhub, tencent]
    steps:
      - name: Login to GitHub Container Registry
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Pull Docker image from GitHub Container Registry
        run: docker pull ${{ env.IMAGE_NAME_CRAWLAB }}:${{ needs.setup.outputs.version }}

      - name: Login to DockerHub or Tencent Registry
        uses: docker/login-action@v3
        with:
          registry: ${{ (matrix.registry == 'dockerhub' && 'https://index.docker.io/v1/') || (matrix.registry == 'tencent' && 'ccr.ccs.tencentyun.com') }}
          username: ${{ (matrix.registry == 'dockerhub' && secrets.DOCKER_USERNAME) || (matrix.registry == 'tencent' && secrets.DOCKER_TENCENT_USERNAME) }}
          password: ${{ (matrix.registry == 'dockerhub' && secrets.DOCKER_PASSWORD) || (matrix.registry == 'tencent' && secrets.DOCKER_TENCENT_PASSWORD) }}

      - name: Tag and push image
        run: |
          docker tag ${{ env.IMAGE_NAME_CRAWLAB }}:${{ needs.setup.outputs.version }} ${{ (matrix.registry == 'dockerhub' && env.IMAGE_NAME_CRAWLAB_DOCKERHUB) || (matrix.registry == 'tencent' && env.IMAGE_NAME_CRAWLAB_TENCENT) }}:${{ needs.setup.outputs.version }}
          docker push ${{ (matrix.registry == 'dockerhub' && env.IMAGE_NAME_CRAWLAB_DOCKERHUB) || (matrix.registry == 'tencent' && env.IMAGE_NAME_CRAWLAB_TENCENT) }}:${{ needs.setup.outputs.version }}