release.yml

name: Release

on:
  schedule:
    # Monday-Wednesday at 2 AM UTC: alpha builds from dev
    - cron: "0 2 * * 1-3"
    # Thursday at 2 AM UTC: beta builds from dev
    - cron: "0 2 * * 4"
  workflow_dispatch:
    inputs:
      release_type:
        description: "Release type"
        required: true
        type: choice
        options:
          - alpha
          - beta
          - stable
      dry_run:
        description: "Dry run (skip PyPI publish)"
        required: false
        default: false
        type: boolean
      force_build:
        description: "Force build even if no changes"
        required: false
        default: false
        type: boolean
      version_override:
        description: "Override version (e.g. 4.4.0) — stable only"
        required: false
        type: string

concurrency:
  group: release-${{ github.ref }}
  cancel-in-progress: false

jobs:
  # ── 1. Determine release type and check for changes ──────────────────
  determine-release:
    runs-on: ubuntu-latest
    outputs:
      release_type: ${{ steps.resolve.outputs.release_type }}
      has_changes: ${{ steps.changes.outputs.has_changes }}
      target_branch: ${{ steps.resolve.outputs.target_branch }}
    steps:
      - name: Resolve release type
        id: resolve
        run: |
          if [ "${{ github.event_name }}" = "workflow_dispatch" ]; then
            TYPE="${{ inputs.release_type }}"
          elif [ "${{ github.event.schedule }}" = "0 2 * * 4" ]; then
            TYPE="beta"
          else
            TYPE="alpha"
          fi

          if [ "$TYPE" = "stable" ]; then
            BRANCH="main"
          else
            BRANCH="dev"
          fi

          echo "release_type=$TYPE" >> $GITHUB_OUTPUT
          echo "target_branch=$BRANCH" >> $GITHUB_OUTPUT
          echo "Release type: $TYPE from $BRANCH"

      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
          ref: ${{ steps.resolve.outputs.target_branch }}

      - name: Check for changes
        id: changes
        run: |
          TYPE="${{ steps.resolve.outputs.release_type }}"

          if [ "$TYPE" = "alpha" ]; then
            LAST_TAG=$(git tag -l "*a*" --sort=-version:refname | head -n1)
          elif [ "$TYPE" = "beta" ]; then
            LAST_TAG=$(git tag -l "*b*" --sort=-version:refname | head -n1)
          else
            LAST_TAG=$(git describe --tags --abbrev=0 --match "v[0-9]*.[0-9]*.[0-9]*" 2>/dev/null || echo "")
          fi

          if [ -z "$LAST_TAG" ]; then
            COMMIT_COUNT=$(git rev-list --count --since="7 days ago" HEAD)
          else
            COMMIT_COUNT=$(git rev-list --count ${LAST_TAG}..HEAD)
          fi

          echo "Commits since ${LAST_TAG:-'(none)'}: $COMMIT_COUNT"

          if [ "$COMMIT_COUNT" -gt 0 ] || [ "${{ inputs.force_build }}" = "true" ]; then
            echo "has_changes=true" >> $GITHUB_OUTPUT
          else
            echo "has_changes=false" >> $GITHUB_OUTPUT
            echo "No changes detected, skipping release"
          fi

  # ── 2. Test gate ──────────────────────────────────────────────────────
  test:
    needs: determine-release
    if: needs.determine-release.outputs.has_changes == 'true'
    runs-on: ubuntu-latest
    strategy:
      matrix:
        python-version: ["3.10", "3.11", "3.12"]
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
          ref: ${{ needs.determine-release.outputs.target_branch }}

      - name: Set up Python ${{ matrix.python-version }}
        uses: actions/setup-python@v5
        with:
          python-version: ${{ matrix.python-version }}
          cache: "pip"

      - name: Install Tesseract OCR
        run: |
          sudo apt-get update
          sudo apt-get install -y tesseract-ocr libtesseract-dev

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -e ".[all,dev]"
          pip install -r requirements-dev.txt
          pip install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.7.1/en_core_web_sm-3.7.1.tar.gz

      - name: Run tests with segfault protection
        run: |
          python run_tests.py tests/ --ignore=tests/test_gliner_annotator.py --cov-report=xml --cov-config=.coveragerc

      - name: Run performance validation
        run: |
          OMP_NUM_THREADS=4 MKL_NUM_THREADS=4 OPENBLAS_NUM_THREADS=4 python tests/simple_performance_test.py

  # ── 3. Build & Publish ────────────────────────────────────────────────
  publish:
    needs: [determine-release, test]
    runs-on: ubuntu-latest
    outputs:
      version: ${{ steps.version.outputs.version }}
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
          ref: ${{ needs.determine-release.outputs.target_branch }}
          token: ${{ secrets.GH_PAT }}

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

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install build twine bump2version

      - name: Configure git
        run: |
          git config --local user.email "action@github.com"
          git config --local user.name "GitHub Action"

      - name: Generate version
        id: version
        run: |
          set -e
          git fetch --tags

          TYPE="${{ needs.determine-release.outputs.release_type }}"
          CURRENT=$(sed -n 's/^__version__ = "\([^"]*\)"/\1/p' datafog/__about__.py)
          if [ -z "$CURRENT" ]; then
            echo "Failed to parse current version from datafog/__about__.py"
            exit 1
          fi
          echo "Current version: $CURRENT"

          # Strip any pre-release suffix to get base version
          BASE=$(echo "$CURRENT" | sed -E 's/(a|b)[0-9]+([.][0-9A-Za-z]+)?$//')
          echo "Base version: $BASE"

          if [ "$TYPE" = "alpha" ]; then
            ALPHA_NUM=1
            while git tag -l "v${BASE}a${ALPHA_NUM}" | grep -q .; do
              ALPHA_NUM=$((ALPHA_NUM + 1))
            done
            VERSION="${BASE}a${ALPHA_NUM}"

          elif [ "$TYPE" = "beta" ]; then
            BETA_NUM=1
            while git tag -l "v${BASE}b${BETA_NUM}" | grep -q .; do
              BETA_NUM=$((BETA_NUM + 1))
            done
            VERSION="${BASE}b${BETA_NUM}"

          else
            # Stable: use override or base version
            if [ -n "${{ inputs.version_override }}" ]; then
              VERSION="${{ inputs.version_override }}"
            else
              VERSION="$BASE"
            fi
          fi

          echo "version=$VERSION" >> $GITHUB_OUTPUT
          echo "Publishing version: $VERSION"

          sed -i "s/__version__ = \".*\"/__version__ = \"$VERSION\"/" datafog/__about__.py
          if grep -q 'version="' setup.py 2>/dev/null; then
            sed -i "s/version=\".*\"/version=\"$VERSION\"/" setup.py
          fi

      - name: Generate changelog
        run: |
          TYPE="${{ needs.determine-release.outputs.release_type }}"
          if [ "$TYPE" = "alpha" ]; then
            python scripts/generate_changelog.py --alpha --output RELEASE_CHANGELOG.md
          elif [ "$TYPE" = "beta" ]; then
            python scripts/generate_changelog.py --beta --output RELEASE_CHANGELOG.md
          else
            python scripts/generate_changelog.py --output RELEASE_CHANGELOG.md
          fi

      - name: Build package
        run: |
          python -m build
          python scripts/check_wheel_size.py

      - name: Publish to PyPI
        if: inputs.dry_run != true
        env:
          TWINE_USERNAME: __token__
          TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
        run: |
          python -m twine upload dist/* --verbose

      - name: Commit version bump & create release
        if: inputs.dry_run != true
        env:
          GITHUB_TOKEN: ${{ secrets.GH_PAT }}
        run: |
          VERSION="${{ steps.version.outputs.version }}"
          TYPE="${{ needs.determine-release.outputs.release_type }}"
          BRANCH="${{ needs.determine-release.outputs.target_branch }}"

          git add datafog/__about__.py setup.py
          git commit -m "chore: bump version to $VERSION [skip ci]" || echo "No version changes to commit"
          git push origin "$BRANCH"

          git tag -a "v$VERSION" -m "Release $VERSION"
          git push origin "v$VERSION"

          PRERELEASE_FLAG=""
          TITLE=""
          if [ "$TYPE" = "alpha" ]; then
            PRERELEASE_FLAG="--prerelease"
            TITLE="Nightly Alpha $VERSION"
          elif [ "$TYPE" = "beta" ]; then
            PRERELEASE_FLAG="--prerelease"
            TITLE="Beta Release $VERSION"
          else
            TITLE="DataFog v$VERSION"
          fi

          gh release create "v$VERSION" \
            --title "$TITLE" \
            --notes-file RELEASE_CHANGELOG.md \
            $PRERELEASE_FLAG \
            --target "$BRANCH" \
            dist/*

      - name: Dry run summary
        if: inputs.dry_run == true
        run: |
          echo "DRY RUN COMPLETE"
          echo "Would have published: ${{ steps.version.outputs.version }}"
          echo "Package contents:"
          ls -la dist/

  # ── 4. Cleanup old pre-releases ───────────────────────────────────────
  cleanup:
    needs: [determine-release, publish]
    if: needs.determine-release.outputs.release_type != 'stable' && inputs.dry_run != true
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Prune old alpha releases (keep 7)
        if: needs.determine-release.outputs.release_type == 'alpha'
        env:
          GITHUB_TOKEN: ${{ secrets.GH_PAT }}
        run: |
          echo "Cleaning up old alpha releases (keep last 7)..."
          ALPHA_RELEASES=$(gh release list --limit 50 | grep -i alpha | tail -n +8 | cut -f3)

          for release in $ALPHA_RELEASES; do
            echo "Deleting old alpha release: $release"
            gh release delete "$release" --yes || true
            git push --delete origin "$release" 2>/dev/null || true
          done

      - name: Prune old beta releases (keep 5)
        if: needs.determine-release.outputs.release_type == 'beta'
        env:
          GITHUB_TOKEN: ${{ secrets.GH_PAT }}
        run: |
          echo "Cleaning up old beta releases (keep last 5)..."
          BETA_RELEASES=$(gh release list --limit 30 | grep -i beta | tail -n +6 | cut -f3)

          for release in $BETA_RELEASES; do
            echo "Deleting old beta release: $release"
            gh release delete "$release" --yes || true
            git push --delete origin "$release" 2>/dev/null || true
          done