check-broken-link.yml

name: check_broken_links

on:
  schedule:
    - cron: 30 17 * * *
  push:
    paths:
      - '**/*.ipynb'
  pull_request:
    paths:
      - '**/*.ipynb'
  # Trigger the workflow on manual dispatch
  workflow_dispatch:

permissions:
  contents: read
  pages: write
  id-token: write

defaults:
  run:
    shell: bash

jobs:
  get-notebooks:
    runs-on: ubuntu-22.04
    outputs:
      notebook_list: ${{ steps.list_all_notebooks.outputs.notebook_list }}
    steps:
    - uses: actions/checkout@v4
    - name: Get all notebooks
      working-directory: ./books
      id: list_all_notebooks
      run: |
        all_notebooks=($(find . -type f -name "*.ipynb"))
        notebook_list='['
        for NOTEBOOK in "${all_notebooks[@]}"; do
          echo "NOTEBOOK=${NOTEBOOK}"
          notebook_list+="\"${NOTEBOOK}\","
        done
        notebook_list=$(sed '$s/,$//' <<< $notebook_list)
        notebook_list+=']'
        echo "notebook_list=${notebook_list}"
        echo "notebook_list=${notebook_list}" >> $GITHUB_OUTPUT

  check-links:
    needs: [ get-notebooks ]
    if: ${{ needs.get-notebooks.outputs.notebook_list != '[]' }}
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@master
      - name: Set up Python
        uses: actions/setup-python@v4
        with:
          python-version: '3.9'

      - name: Install Python dependencies
        run: |
          python -m pip install pytest-check-links

      - name: Check broken links
        working-directory: ./books
        run: |
          notebook_list='${{ needs.get-notebooks.outputs.notebook_list }}'
          echo "notebook_list=${notebook_list}"
          for notebook in $(echo $notebook_list | jq -r '.[]'); do
            echo "Checking links in $notebook"
            pytest --check-links --check-links-ignore "https://doi.org/.*/" "https://www.sciencedirect.com/.*" $notebook
          done