release-build-publish.yml

name: Release Build Publish

on:
    release:
      types: [ published ]

jobs:
  extract_project_name:
    runs-on: ubuntu-latest
    outputs:
      cleanlab_package_name: ${{ steps.extract_name.outputs.value }}
    steps:
      - name: Checkout code
        uses: actions/checkout@v4

      - name: TOML Reader
        uses: SebRollen/toml-action@v1.2.0
        id: extract_name
        with:
          file: "pyproject.toml"
          field: "project.name"

  store_published_version:
    runs-on: ubuntu-latest
    outputs:
      version: ${{ steps.compare.outputs.version }}
    steps:

      - name: Checkout code
        uses: actions/checkout@v4

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11

      - name: Install package
        run: pip install .

      - name: Compare the published version with the package version
        id: compare
        run: |
          TAG_NAME="${{ github.ref_name }}"
          # TAG_NAME must be in the form "v1.2.3[...]", error if it doesn't match
            if [[ ! $TAG_NAME =~ ^v[0-9]+\.[0-9]+\.[0-9]+ ]]; then
                echo "Tag name '$TAG_NAME' does not match the expected pattern 'vX.Y.Z'"
                exit 1
            fi
          # Strip the 'v' prefix from the tag, and store the version in an output variable
          VERSION=${TAG_NAME:1} # Strip the 'v' prefix from the tag
          echo "version=$VERSION" >> $GITHUB_OUTPUT
          # Check that the package version matches the tag version
          PACKAGE_VERSION=$(python -c "import cleanlab; print(cleanlab.__version__)")
          if [ "$PACKAGE_VERSION" != "$VERSION" ]; then
            echo "Package version $PACKAGE_VERSION does not match tag $VERSION"
            exit 1
          fi

  build:
    needs: [store_published_version, extract_project_name]
    runs-on: ubuntu-latest

    steps:
      - name: Checkout code
        uses: actions/checkout@v4

      - name: Remove tests directory
        run: |
          # Remove the tests directory from the package
          TEST_DIR=$(find . -type d -name "tests")
          if [ -n "$TEST_DIR" ]; then
            rm -rf $TEST_DIR
          else
            echo "No tests directory found! This is unexpected."
            exit 1
          fi

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11

      - name: Install dependencies
        run: pip install --upgrade setuptools wheel twine

      - name: Perform some tests on the metadata of the package
        run: |
          python setup.py check --metadata --strict

      - name: Build package
        run: |
          python setup.py sdist bdist_wheel

      - name: Test package
        run: |
          twine check dist/*

      - name: Verify contents of the tar.gz
        env:
          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
        run: |
          # Extract the tar.gz to a temporary directory
          TMPDIR=$(mktemp -d)
          tar -xzf dist/*.tar.gz -C $TMPDIR
          PACKAGE_DIR=$TMPDIR/$(ls $TMPDIR) # Assuming there's only one directory extracted

          # Define expected files and directories
          EXPECTED_FILES="LICENSE PKG-INFO MANIFEST.in DEVELOPMENT.md CONTRIBUTING.md CODE_OF_CONDUCT.md README.md pyproject.toml setup.cfg setup.py"
          EXPECTED_DIRS="cleanlab $CLEANLAB_PACKAGE_NAME.egg-info"

          # Collect actual top-level files and directories
          ACTUAL_CONTENTS=$(ls $PACKAGE_DIR)
          ACTUAL_FILES=$(find $PACKAGE_DIR -maxdepth 1 -type f -exec basename {} \;)
          ACTUAL_DIRS=$(find $PACKAGE_DIR -maxdepth 1 -type d -exec basename {} \; | sed "1d") # Remove the package directory itself from the list

          # Function to check for unexpected or missing files and directories
          check_contents() {
            local missing=0
            # Check for expected files
            for expected_file in $EXPECTED_FILES; do
              echo $ACTUAL_FILES | grep -qw $expected_file || { echo "Missing expected file: $expected_file"; missing=1; }
            done
            
            # Check for expected directories
            for expected_dir in $EXPECTED_DIRS; do
              echo $ACTUAL_DIRS | grep -qw $expected_dir || { echo "Missing expected directory: $expected_dir"; missing=1; }
            done

            # Check for unexpected files and directories
            for actual in $ACTUAL_CONTENTS; do
              echo $EXPECTED_FILES $EXPECTED_DIRS | grep -qw $actual || { echo "Unexpected item in package: $actual"; missing=1; }
            done

            # Exit if anything unexpected or missing
            if [ $missing -ne 0 ]; then
              echo "Package content verification failed."
              exit 1
            fi
          }

          # Execute checks
          check_contents

      - name: Store build artifacts
        uses: actions/upload-artifact@v4
        with:
          name: package
          path: dist/*

  publish-testpypi:
    needs: build
    runs-on: ubuntu-latest
    permissions:
        id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing

    steps:
      - name: Fetch build artifacts
        uses: actions/download-artifact@v4
        with:
          name: package
          path: dist/

      - name: Publish to TestPyPI
        uses: pypa/gh-action-pypi-publish@e53eb8b103ffcb59469888563dc324e3c8ba6f06
        with:
          repository-url: https://test.pypi.org/legacy/
          verify-metadata: true

  verify-version:
    needs: [publish-testpypi, store_published_version, extract_project_name]
    runs-on: ubuntu-latest
    environment: testpypi

    steps:
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11

      - name: Install package from TestPyPI
        env:
          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME

      - name: Verify the published version
        env:
          VERSION: ${{ needs.store_published_version.outputs.version }}
        run: |
          TEST_IMPORT_VERSION=$(python -c "import cleanlab; print(cleanlab.__version__)")
          if [ "$TEST_IMPORT_VERSION" != "$VERSION" ]; then
            echo "Imported version $TEST_IMPORT_VERSION does not match tag $VERSION"
            exit 1
          fi

  test-basic-import:
    needs: [verify-version, extract_project_name]
    runs-on: ubuntu-latest

    steps:
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11

      - name: Install package from TestPyPI
        env:
          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME

      - name: Test importing package
        run: |
          python -c "import cleanlab
          from cleanlab.outlier import OutOfDistribution
          ood = OutOfDistribution()
          print(ood)"

  test-extras-import:
    needs: [verify-version, extract_project_name]
    runs-on: ubuntu-latest
 
    steps:
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: 3.11

      - name: Install package with extras
        env:
          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME[all]

      - name: Test importing package with extras
        run: |
          python -c "import cleanlab
          from cleanlab import Datalab
          lab = Datalab({'a': [1, 2, 3, 4, 5], 'b': ['a', 'b', 'c', 'b', 'a']})
          print(lab)"

  publish-pypi:
    needs: [test-basic-import, test-extras-import]
    runs-on: ubuntu-latest
    environment: pypi
    permissions:
        id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing

    steps:
      - name: Fetch build artifacts
        uses: actions/download-artifact@v4
        with:
          name: package
          path: dist/

      - name: Publish to PyPI
        uses: pypa/gh-action-pypi-publish@e53eb8b103ffcb59469888563dc324e3c8ba6f06
        with:
          verify-metadata: true