ci_html_build.yml

name: Build HTML on merge

on:
  pull_request_target:
    types: [closed]
    paths:
      - '**/.ipynb'
      - '**/requirements.txt'

env:
  CASJOBS_PW: ${{ secrets.CASJOBS_PW }}
  CASJOBS_USERID: ${{ secrets.CASJOBS_USERID }}
jobs:
  prepare_matrix:
    if: github.event.pull_request.merged == true
    runs-on: ubuntu-latest
    outputs:
      matrix: ${{ steps.set-matrix.outputs.matrix }}
    steps:
      - uses: actions/checkout@v3
        with:
          fetch-depth: 0 
          ref: ${{ github.event.pull_request.head.sha }}

      - id: set-matrix
        shell: bash
        run: |
          # Convert the comma-separated list to an array.
          IFS=',' read -r -a files <<< "${{ steps.get-changed-files.outputs.all_changed_files }}"
          echo "Changed files: ${files[*]}"
          
          # Array to hold the notebooks that will be executed.
          notebooks=()
          notebook_changed=false
          
          # First, if any changed file is a notebook, add it.
          for file in "${files[@]}"; do
            if [[ "$file" == *.ipynb ]]; then
              notebook_changed=true
              notebooks+=("$file")
            fi
          done
          
          # If no notebook file changed, look for requirements.txt changes.
          if [ "$notebook_changed" = false ]; then
            for file in "${files[@]}"; do
              if [[ "$file" == *requirements.txt ]]; then
                dir=$(dirname "$file")
                echo "Found requirements file in folder '$dir'; gathering notebooks from that folder."
                # Find all notebooks in the folder (only that folder, not recursing).
                while IFS= read -r nb; do
                  notebooks+=("$nb")
                done < <(find "$dir" -maxdepth 1 -type f -name '*.ipynb')
              fi
            done
          fi
          
          # Remove duplicate entries.
          unique_notebooks=($(printf "%s\n" "${notebooks[@]}" | sort -u))
          
          if [ ${#unique_notebooks[@]} -eq 0 ]; then
            echo "No relevant changes found; skipping notebook execution."
            echo "has_notebooks=false" >> "$GITHUB_OUTPUT"
            echo "matrix=[]" >> "$GITHUB_OUTPUT"
          else
            echo "Notebooks selected for testing: ${unique_notebooks[*]}"
            echo "has_notebooks=true" >> "$GITHUB_OUTPUT"
            # Build a JSON array for the matrix.
            matrix_json=$(printf '%s\n' "${unique_notebooks[@]}" | jq -R . | jq -s .)
            # Write the JSON to GITHUB_OUTPUT using the heredoc syntax.
            printf "matrix<<EOF\n%s\nEOF\n" "$matrix_json" >> "$GITHUB_OUTPUT"
          fi

  execute-notebooks:
    if: github.event.pull_request.merged == true
    needs: prepare_matrix
    runs-on: ubuntu-latest
    strategy:
      fail-fast: false
      matrix:
        notebook: ${{fromJson(needs.prepare_matrix.outputs.matrix)}}

    steps:
      - uses: actions/checkout@v3

      - name: Setup Python
        uses: actions/setup-python@v3
        with:
          python-version:  ${{ vars.PYTHON_VERSION }}

      - name: Install Dependencies (if requirements.txt exists) and Execute Notebook
        run: |
          notebook="${{ matrix.notebook }}"
          dir=$(dirname "$notebook")
          if [ -f "$dir/requirements.txt" ]; then
            pip install -r "$dir/requirements.txt"
          fi
          pip install notebook
          jupyter nbconvert --to notebook --execute --inplace "$notebook" 

      - name: Commit modified file on current branch
        run: |
          git config user.name 'CI Bot'
          git config user.email 'action@github.com'
          git add "${{ matrix.notebook }}"
          git commit -m "Storing executed notebook ${{ matrix.notebook }}"

      - name: Checkout only the file to the target branch
        run: |
          git fetch
          git pull
          git checkout -f gh-storage
          git checkout @{-1} "${{ matrix.notebook }}"

      - name: Commit and push modifications to target branch
        run: |
          git commit -m "Storing executed notebook ${{ matrix.notebook }}"

          MAX_RETRIES=5
          RETRY_DELAY=10s
          for i in $(seq 1 $MAX_RETRIES); do
            git push origin gh-storage --force && break || {
              echo "Push $i failed... waiting $RETRY_DELAY"
              sleep $RETRY_DELAY
            }
          done
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}


  generate_html:
    if: github.event.pull_request.merged == true
    needs: execute-notebooks
    runs-on: ubuntu-latest
    permissions:
      contents: write
    steps:
    - name: Checkout repository
      uses: actions/checkout@v4

    - name: Set up Python  ${{ vars.PYTHON_VERSION }}
      uses: actions/setup-python@v5
      with:
        python-version:  ${{ vars.PYTHON_VERSION }}
        
    - name: Install dependencies
      run: |
        pip install ghp-import
        pip install jupyter-book==v0.15.1
        pip install myst-nb
        pip install astroid
        pip install nbval
        #pip install bs4
        #pip install lxml
        ## test to bypass the jupyter-book lower version
        pip install jsonschema==4.6.0
        PATH="${PATH}:${HOME}/.local/bin"

    - name: Build HTML
      run: |
        git fetch
        git checkout origin/gh-storage -- notebooks/
        jupyter-book build .
        
        # Push the book's HTML to github-pages
    - name: GitHub Pages action
      uses: peaceiris/actions-gh-pages@v3
      with:
        github_token: ${{ secrets.GITHUB_TOKEN }}
        publish_dir: ./_build/html