run.yml

# On every push this script is executed
on:
  workflow_dispatch:
  schedule:
    - cron: "0 3 * * *"

concurrency: data

name: Build data
jobs:
  generate-matrix:
    runs-on: ubuntu-latest
    env:
      UV_CACHE_DIR: /tmp/.uv-cache
    steps:
      - name: checkout
        uses: actions/checkout@v4

      - uses: actions/setup-python@v5
        with:
          python-version: '3.11'
          cache: pip

      - name: Set up uv
        # Install latest uv version using the installer
        run: curl -LsSf https://astral.sh/uv/install.sh | sh

      - name: Restore uv cache
        uses: actions/cache@v4
        with:
          path: /tmp/.uv-cache
          key: uv-${{ runner.os }}-${{ hashFiles('uv.lock') }}
          restore-keys: |
            uv-${{ runner.os }}-${{ hashFiles('uv.lock') }}
            uv-${{ runner.os }}      

      - name: Generate token
        id: generate_token
        uses: pypi-data/github-app-token@v1
        with:
          app_id: ${{ secrets.APP_ID }}
          private_key: ${{ secrets.APP_PRIVATE_KEY }}

      - name: Install deps
        run: uv sync --frozen

      - env:
          GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
        run: |
          uv run pypi-data print-git-urls > links/repositories.txt
          uv run pypi-data print-git-urls --ssh-urls > links/repositories_ssh.txt

      - uses: EndBug/add-and-commit@v9
        with:
          add: |
            links/repositories.txt
            links/repositories_ssh.txt
          author_email: "41898282+github-actions[bot]@users.noreply.github.com"
          author_name: "commit-bot"
          message: "Add repository URLs"
          push: true
          fetch: true
          pull: '--rebase --autostash'

      - env:
          GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
        id: groups
        run: |
          mkdir output/
          uv run pypi-data group-index-urls output/
          echo "matrix=$(cat output/groups.json)" >> "$GITHUB_OUTPUT"

      - uses: actions/upload-artifact@v4
        with:
          name: groups
          path: output/groups/
          retention-days: 1

      - name: "Set current date as env variable"
        run: |
          echo "tag_name=$(date +'%Y-%m-%d-%H-%M')" >> $GITHUB_OUTPUT
        id: version

      - name: Create Release
        id: create-release
        uses: shogo82148/actions-create-release@v1
        with:
          draft: true
          release_name: ${{ steps.version.outputs.tag_name }}
          tag_name: ${{ steps.version.outputs.tag_name }}
          commitish: ${{ github.sha }}

      - name: Release URL
        run: echo ${{ steps.create-release.html_url }}

      - name: Minimize uv cache
        run: uv cache prune --ci

    outputs:
      matrix: ${{ steps.groups.outputs.matrix }}
      upload_url: ${{ steps.create-release.outputs.upload_url }}
      release_id: ${{ steps.create-release.outputs.id }}

  combine:
    needs: [generate-matrix]
    runs-on: ubuntu-latest
    strategy:
      fail-fast: false
      matrix:
        index: ${{fromJson(needs.generate-matrix.outputs.matrix)}}
    steps:
      - name: Maximize build space
        uses: easimon/maximize-build-space@master
        with:
          remove-dotnet: 'true'
          remove-android: 'true'
          remove-haskell: 'true'
          remove-codeql: 'true'
          remove-docker-images: 'true'

      - name: checkout
        uses: actions/checkout@v4

      - uses: actions/download-artifact@v4
        with:
          name: groups

      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-on-failure: 'false'

      - name: Install parquet cli from crates.io
        uses: baptiste0928/cargo-install@v3
        with:
          crate: parquet
          features: cli

      - name: Download links
        run: cat ${{ matrix.index }} | jq -rc '.[]'

      - name: Debug
        run: |
          echo "Links for ${{ matrix.index }}" 
          cat ${{ matrix.index }} | jq -rc '.[] | [.name, .id] | @tsv'

      - name: Download
        run: |
          mkdir input/
          cat ${{ matrix.index }} | jq -rc '.[] | [.name, .id] | @tsv' | parallel --colsep '\t' wget --tries=2 -nv -O input/{2}.parquet {1}

      - run: ls -la ${{ github.workspace }}/input/

      - name: Combine
        run: parquet-concat ${{ github.workspace }}/merged.parquet ${{ github.workspace }}/input/*.parquet

      - name: Merged size
        run: du -hs ${{ github.workspace }}/merged.parquet

      - name: Rewrite
        run: |
          parquet-rewrite \
            --compression=zstd \
            --writer-version=2.0 \
            --dictionary-enabled=true \
            --statistics-enabled=page \
            --bloom-filter-enabled=true \
            --bloom-filter-fpp 0.2 \
            --input=${{ github.workspace }}/merged.parquet \
            --output=${{ github.workspace }}/output.parquet 

      - name: Output size
        run: du -hs ${{ github.workspace }}/output.parquet

      - name: Upload Assets
        uses: shogo82148/actions-upload-release-asset@v1
        with:
          upload_url: ${{ needs.generate-matrix.outputs.upload_url }}
          asset_path: ${{ github.workspace }}/output.parquet
          asset_name: index-${{ matrix.index }}.parquet

  makepublic:
    needs: [generate-matrix, combine]
    runs-on: ubuntu-latest
    steps:
      - name: checkout
        uses: actions/checkout@v4

      - name: Publish release
        uses: StuYarrow/publish-release@v1.1.2
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        with:
          id: ${{ needs.generate-matrix.outputs.release_id }}

      - name: Get download links
        id: get_download_links
        uses: actions/github-script@v7
        with:
          github-token: ${{ secrets.GITHUB_TOKEN }}
          script: |
            const response = await github.rest.repos.listReleaseAssets({
              owner: context.repo.owner,
              repo: context.repo.repo,
              release_id: ${{ needs.generate-matrix.outputs.release_id }},
            });
            const assets = response.data;
            // Save the download links to a file
            require('fs').writeFileSync('links/dataset.txt', assets.map(asset => asset.browser_download_url).join('\n'));

      - run: cat links/dataset.txt

      - uses: EndBug/add-and-commit@v9
        with:
          add: 'links/dataset.txt'
          author_email: "41898282+github-actions[bot]@users.noreply.github.com"
          author_name: "commit-bot"
          message: "Add download links for asset ${{ needs.generate-matrix.outputs.release_id }}"
          push: true
          fetch: true
          pull: '--rebase --autostash'