ArchiveBox/.github/workflows/test-parallel.yml

name: Parallel Tests

on:
  pull_request:
    branches: [dev, main, master]
  push:
    branches: [dev]

env:
  PYTHONIOENCODING: utf-8
  PYTHONLEGACYWINDOWSSTDIO: utf-8
  USE_COLOR: False

jobs:
  discover-tests:
    name: Discover test files
    runs-on: ubuntu-22.04
    outputs:
      test-files: ${{ steps.set-matrix.outputs.test-files }}

    steps:
      - uses: actions/checkout@v4
        with:
          submodules: true
          fetch-depth: 1

      - name: Discover test files
        id: set-matrix
        run: |
          # Find all main test files
          main_tests=$(find tests -maxdepth 1 -name "test_*.py" -type f | sort)

          # Find all plugin test files
          plugin_tests=$(find archivebox/plugins -path "*/tests/test_*.py" -type f | sort)

          # Combine and format as JSON array
          all_tests=$(echo "$main_tests $plugin_tests" | tr ' ' '\n' | grep -v '^$')

          # Create JSON array with test file info
          json_array="["
          first=true
          for test_file in $all_tests; do
            if [ "$first" = true ]; then
              first=false
            else
              json_array+=","
            fi

            # Extract a display name for the test
            if [[ $test_file == tests/* ]]; then
              name="main/$(basename $test_file .py | sed 's/^test_//')"
            else
              plugin=$(echo $test_file | sed 's|archivebox/plugins/\([^/]*\)/.*|\1|')
              test_name=$(basename $test_file .py | sed 's/^test_//')
              name="plugin/$plugin/$test_name"
            fi

            json_array+="{\"path\":\"$test_file\",\"name\":\"$name\"}"
          done
          json_array+="]"

          echo "test-files=$json_array" >> $GITHUB_OUTPUT
          echo "Found $(echo $all_tests | wc -w) test files"
          echo "$json_array" | jq '.'

  run-tests:
    name: ${{ matrix.test.name }}
    runs-on: ubuntu-22.04
    needs: discover-tests

    strategy:
      fail-fast: false
      matrix:
        test: ${{ fromJson(needs.discover-tests.outputs.test-files) }}
        python: ["3.13"]

    steps:
      - uses: actions/checkout@v4
        with:
          submodules: true
          fetch-depth: 1

      - name: Set up Python ${{ matrix.python }}
        uses: actions/setup-python@v4
        with:
          python-version: ${{ matrix.python }}
          architecture: x64

      - name: Set up Node JS
        uses: actions/setup-node@v4
        with:
          node-version: 20.10.0

      - name: Get pip cache dir
        id: pip-cache
        run: |
          echo "dir=$(pip cache dir)" >> $GITHUB_OUTPUT

      - name: Cache pip
        uses: actions/cache@v3
        id: cache-pip
        with:
          path: ${{ steps.pip-cache.outputs.dir }}
          key: ${{ runner.os }}-${{ matrix.python }}-venv-${{ hashFiles('setup.py') }}
          restore-keys: |
            ${{ runner.os }}-${{ matrix.python }}-venv-

      - uses: awalsh128/cache-apt-pkgs-action@latest
        with:
          packages: ripgrep build-essential python3-dev python3-setuptools libssl-dev libldap2-dev libsasl2-dev zlib1g-dev libatomic1 python3-minimal gnupg2 curl wget python3-ldap python3-msgpack python3-mutagen python3-regex python3-pycryptodome procps
          version: 1.0

      - name: Install pip dependencies
        run: |
          python -m pip install --upgrade pip setuptools wheel pytest bottle build
          python -m pip install -r requirements.txt
          python -m pip install -e .[sonic,ldap]

      - name: Get npm cache dir
        id: npm-cache
        run: |
          echo "dir=$GITHUB_WORKSPACE/node_modules" >> $GITHUB_OUTPUT

      - name: Cache npm
        uses: actions/cache@v3
        id: cache-npm
        with:
          path: ${{ steps.npm-cache.outputs.dir }}
          key: ${{ runner.os }}-node_modules-${{ hashFiles('package-lock.json') }}
          restore-keys: |
            ${{ runner.os }}-node_modules

      - name: Install npm requirements
        run: |
          npm install
          echo "SINGLEFILE_BINARY=$GITHUB_WORKSPACE/node_modules/.bin/single-file" >> $GITHUB_ENV
          echo "READABILITY_BINARY=$GITHUB_WORKSPACE/node_modules/.bin/readability-extractor" >> $GITHUB_ENV
          echo "MERCURY_BINARY=$GITHUB_WORKSPACE/node_modules/.bin/mercury-parser" >> $GITHUB_ENV

      - name: Run test - ${{ matrix.test.name }}
        run: |
          python -m pytest -xvs "${{ matrix.test.path }}" --basetemp=tests/out --ignore=archivebox/pkgs