Build And Push Docker Image workflow (Unstructured-IO/unstructured-api)

name: Build And Push Docker Image
 
on:
  push:
    branches:
      - main
  workflow_dispatch:
 
env:
  DOCKER_REPOSITORY: quay.io/unstructured-io/unstructured-api
  DOCKER_BUILD_REPOSITORY: quay.io/unstructured-io/build-unstructured-api
  PACKAGE: "unstructured-api"
  PIPELINE_FAMILY: "general"
 
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true
 
jobs:
  set-short-sha:
    timeout-minutes: 30
    runs-on: opensource-linux-8core
    outputs:
      short_sha: ${{ steps.set_short_sha.outputs.short_sha }}
    steps:
      - name: Set Short SHA
        id: set_short_sha
        run: echo "short_sha=$(echo ${{ github.sha }} | cut -c1-7)" >> $GITHUB_OUTPUT
  build-images:
    timeout-minutes: 30
    strategy:
      matrix:
        arch: ["arm64", "amd64"]
    runs-on: ${{ matrix.arch == 'arm64' && 'ubuntu-24.04-arm' || 'opensource-linux-8core' }}
    needs: set-short-sha
    env:
      SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
      DOCKER_PLATFORM: linux/${{ matrix.arch }}
    steps:
    - name: Set up Docker Buildx
      uses: docker/setup-buildx-action@v4
      with:
        driver: docker
    - name: Checkout code
      uses: actions/checkout@v5
    - name: Read Python version from .python-version
      run: echo "PYTHON_VERSION=$(cat .python-version)" >> $GITHUB_ENV
    - name: Login to Quay.io
      uses: docker/login-action@v4
      with:
        registry: quay.io
        username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
        password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
    - name: Free up disk space
      run: |
        # Clear some space (https://github.com/actions/runner-images/issues/2840)
        echo "Disk usage before cleanup:"
        df -h
 
        # Remove unnecessary pre-installed software
        sudo rm -rf /usr/share/dotnet
        sudo rm -rf /opt/ghc
        sudo rm -rf /usr/local/share/boost
        sudo rm -rf /usr/local/lib/android
        sudo rm -rf /opt/hostedtoolcache/CodeQL
        sudo rm -rf /usr/local/.ghcup
        sudo rm -rf /usr/share/swift
 
        # Clean up docker to ensure we start fresh
        docker system prune -af --volumes
 
        echo "Disk usage after cleanup:"
        df -h
    - name: Build image
      run: |
        DOCKER_BUILDKIT=1 docker buildx build --load -f Dockerfile \
          --platform=$DOCKER_PLATFORM \
          --build-arg BUILDKIT_INLINE_CACHE=1 \
          --build-arg PIPELINE_PACKAGE=${{ env.PIPELINE_FAMILY }} \
          --provenance=false \
          --progress plain \
          --cache-from $DOCKER_BUILD_REPOSITORY:${{ matrix.arch }} \
          -t $DOCKER_BUILD_REPOSITORY:${{ matrix.arch }}-$SHORT_SHA .
    - name: Install uv
      uses: astral-sh/setup-uv@v7
      with:
        enable-cache: true
        cache-dependency-glob: "uv.lock"
    - name: Set up Python ${{ env.PYTHON_VERSION }}
      run: uv python install ${{ env.PYTHON_VERSION }}
    - name: Install test dependencies
      run: uv sync --group test --locked
    - name: Test image
      run: |
        export DOCKER_IMAGE="$DOCKER_BUILD_REPOSITORY:${{ matrix.arch }}-$SHORT_SHA"
        SKIP_INFERENCE_TESTS=true make docker-test
    - name: Push image
      run: |
        # write to the build repository to cache for the publish-images job
        docker push $DOCKER_BUILD_REPOSITORY:${{ matrix.arch }}-$SHORT_SHA
  publish-images:
    timeout-minutes: 30
    runs-on: opensource-linux-8core
    needs: [set-short-sha, build-images]
    env:
      SHORT_SHA: ${{ needs.set-short-sha.outputs.short_sha }}
    steps:
    - name: Checkout code
      uses: actions/checkout@v5
    - name: Set SHORT_SHA
      run: echo "SHORT_SHA=$(git rev-parse --short HEAD)" >> $GITHUB_ENV
    - name: Login to Quay.io
      uses: docker/login-action@v4
      with:
        registry: quay.io
        username: ${{ secrets.QUAY_IO_ROBOT_USERNAME }}
        password: ${{ secrets.QUAY_IO_ROBOT_TOKEN }}
    - name: Pull AMD image
      run: |
        docker pull $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA
    - name: Pull ARM image
      run: |
        docker pull $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA
    - name: Push AMD and ARM tags
      run: |
        # these are used to construct the final manifest but also cache-from in subsequent runs
        docker tag $DOCKER_BUILD_REPOSITORY:amd64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:amd64
        docker push $DOCKER_BUILD_REPOSITORY:amd64
        docker tag $DOCKER_BUILD_REPOSITORY:arm64-$SHORT_SHA $DOCKER_BUILD_REPOSITORY:arm64
        docker push $DOCKER_BUILD_REPOSITORY:arm64
    - name: Push multiarch manifest
      run: |
        VERSION=$(grep -oP '(?<=__version__ = ")[^"]+' prepline_general/api/__version__.py)
        docker buildx imagetools create \
          -t ${DOCKER_REPOSITORY}:latest \
          -t ${DOCKER_REPOSITORY}:$SHORT_SHA \
          -t ${DOCKER_REPOSITORY}:$VERSION \
          $DOCKER_BUILD_REPOSITORY:amd64 \
          $DOCKER_BUILD_REPOSITORY:arm64