Optimize preparing packages for PROD image builds (#38539)

With this change provider and airflow packages are prepared only
once for all PROD image builds and uploaded as artifacts. This saves
~ 4 minutes per PROD image build (which is multiplied by the number
of Python versions we have and by all the additional PROD image
checks we have.

.github/workflows/additional-prod-image-tests.yml

@@ -16,7 +16,7 @@
 # under the License.
 #
 ---
-name: All PROD image tests
+name: Additional PROD image tests
 on:  # yamllint disable-line rule:truthy
   workflow_call:
     inputs:

.github/workflows/build-images.yml

@@ -158,6 +158,7 @@ jobs:
           PR_LABELS: ${{ steps.get-latest-pr-labels.outputs.pull-request-labels }}
           GITHUB_CONTEXT: ${{ toJson(github) }}
 
+
   build-ci-images:
     name: Build CI images
     permissions:
@@ -203,6 +204,7 @@ jobs:
       runs-on: '["ubuntu-22.04"]'
       build-type: "Regular"
       do-build: ${{ needs.build-info.outputs.ci-image-build }}
+      upload-package-artifact: "true"
       target-commit-sha: ${{ needs.build-info.outputs.target-commit-sha }}
       pull-request-target: "true"
       is-committer-build: ${{ needs.build-info.outputs.is-committer-build }}

.github/workflows/ci.yml

@@ -489,6 +489,7 @@ jobs:
       runs-on: '["ubuntu-22.04"]'
       build-type: "Regular"
       do-build: ${{ needs.build-info.outputs.in-workflow-build }}
+      upload-package-artifact: "true"
       image-tag: ${{ needs.build-info.outputs.image-tag }}
       ci-image-tag: ${{ needs.build-info.outputs.image-tag }}
       platform: "linux/amd64"

.github/workflows/prod-image-build.yml

@@ -38,6 +38,12 @@ on:  # yamllint disable-line rule:truthy
         required: false
         default: "true"
         type: string
+      upload-package-artifact:
+        description: >
+          Whether to upload package artifacts (true/false). If false, the job will rely on artifacts prepared
+          by the main prod-image build job.
+        required: true
+        type: string
       target-commit-sha:
         description: "The commit SHA to checkout for the build"
         required: false
@@ -110,6 +116,74 @@ on:  # yamllint disable-line rule:truthy
         required: true
         type: string
 jobs:
+
+  build-prod-packages:
+    name: "${{ inputs.do-build == 'true' && 'Build' || 'Skip building' }} Airflow and provider packages"
+    timeout-minutes: 10
+    runs-on: ["ubuntu-22.04"]
+    env:
+      PYTHON_MAJOR_MINOR_VERSION: "${{ inputs.default-python-version }}"
+    steps:
+      - name: "Cleanup repo"
+        shell: bash
+        run: docker run -v "${GITHUB_WORKSPACE}:/workspace" -u 0:0 bash -c "rm -rf /workspace/*"
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - name: "Checkout ${{ github.ref }} ( ${{ github.sha }} )"
+        uses: actions/checkout@v4
+        with:
+          persist-credentials: false
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - name: Cleanup docker
+        uses: ./.github/actions/cleanup-docker
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "${{ inputs.default-python-version }}"
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - name: "Cleanup dist and context file"
+        shell: bash
+        run: rm -fv ./dist/* ./docker-context-files/*
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - name: "Install Breeze"
+        uses: ./.github/actions/breeze
+        if: inputs.do-build == 'true' && inputs.upload-package-artifact == 'true'
+      - name: "Prepare providers packages"
+        shell: bash
+        run: >
+          breeze release-management prepare-provider-packages
+          --package-list-file ./prod_image_installed_providers.txt
+          --package-format wheel --version-suffix-for-pypi dev0
+        if: >
+          inputs.do-build == 'true' &&
+          inputs.upload-package-artifact == 'true' &&
+          inputs.build-provider-packages == 'true'
+      - name: "Prepare chicken-eggs provider packages"
+        # In case of provider packages which use latest dev0 version of providers, we should prepare them
+        # from the source code, not from the PyPI because they have apache-airflow>=X.Y.Z dependency
+        # And when we prepare them from sources they will have apache-airflow>=X.Y.Z.dev0
+        shell: bash
+        run: >
+          breeze release-management prepare-provider-packages
+          --package-format wheel --version-suffix-for-pypi dev0 ${{ inputs.chicken-egg-providers }}
+        if: >
+          inputs.do-build  == 'true' &&
+          inputs.upload-package-artifact == 'true' &&
+          inputs.chicken-egg-providers != ''
+      - name: "Prepare airflow package"
+        shell: bash
+        run: >
+          breeze release-management prepare-airflow-package
+          --package-format wheel --version-suffix-for-pypi dev0
+        if: inputs.do-build  == 'true' && inputs.upload-package-artifact == 'true'
+      - name: "Upload prepared packages as artifacts"
+        uses: actions/upload-artifact@v4
+        with:
+          name: prod-packages
+          path: ./dist
+          retention-days: 7
+          if-no-files-found: error
+        if: inputs.do-build  == 'true' && inputs.upload-package-artifact == 'true'
+
   build-prod-images:
     strategy:
       fail-fast: true
@@ -123,6 +197,8 @@ PROD ${{ inputs.build-type }} image\
 ${{ matrix.python-version }}${{ inputs.do-build == 'true' && ':' || '' }}\
 ${{ inputs.do-build == 'true' && inputs.image-tag || '' }}"
     runs-on: ${{ fromJSON(inputs.runs-on) }}
+    needs:
+      - build-prod-packages
     env:
       BACKEND: sqlite
       DEFAULT_BRANCH: ${{ inputs.branch }}
@@ -209,33 +285,11 @@ ${{ inputs.do-build == 'true' && inputs.image-tag || '' }}"
         shell: bash
         run: rm -fv ./dist/* ./docker-context-files/*
         if: inputs.do-build == 'true'
-      - name: "Prepare providers packages"
-        shell: bash
-        run: >
-          breeze release-management prepare-provider-packages
-          --package-list-file ./prod_image_installed_providers.txt
-          --package-format wheel --version-suffix-for-pypi dev0
-        if: inputs.do-build == 'true' && inputs.build-provider-packages == 'true'
-      - name: "Prepare chicken-eggs provider packages"
-        # In case of provider packages which use latest dev0 version of providers, we should prepare them
-        # from the source code, not from the PyPI because they have apache-airflow>=X.Y.Z dependency
-        # And when we prepare them from sources they will have apache-airflow>=X.Y.Z.dev0
-        shell: bash
-        run: >
-          breeze release-management prepare-provider-packages
-          --package-format wheel --version-suffix-for-pypi dev0 ${{ inputs.chicken-egg-providers }}
-        if: >
-          inputs.do-build == 'true' && inputs.build-provider-packages != 'true' &&
-          inputs.chicken-egg-providers != ''
-      - name: "Prepare airflow package"
-        shell: bash
-        run: >
-          breeze release-management prepare-airflow-package
-          --package-format wheel --version-suffix-for-pypi dev0
-        if: inputs.do-build == 'true'
-      - name: "Copy dist packages to docker-context files"
-        shell: bash
-        run: cp -v --no-preserve=mode,ownership ./dist/*.whl ./docker-context-files
+      - name: "Download packages prepared as artifacts"
+        uses: actions/download-artifact@v4
+        with:
+          name: prod-packages
+          path: ./docker-context-files
         if: inputs.do-build == 'true'
       - name: "Download constraints from the CI build"
         uses: actions/download-artifact@v4

.github/workflows/prod-image-extra-checks.yml

@@ -56,6 +56,7 @@ jobs:
     uses: ./.github/workflows/prod-image-build.yml
     with:
       build-type: "Bullseye"
+      upload-package-artifact: "false"
       image-tag: bullseye-${{ inputs.image-tag }}
       ci-image-tag: ${{ inputs.image-tag }}
       debian-version: "bullseye"
@@ -75,6 +76,7 @@ jobs:
     uses: ./.github/workflows/prod-image-build.yml
     with:
       build-type: "MySQL Client"
+      upload-package-artifact: "false"
       image-tag: mysql-${{ inputs.image-tag }}
       ci-image-tag: ${{ inputs.image-tag }}
       install-mysql-client-type: "mysql"
@@ -94,6 +96,7 @@ jobs:
     uses: ./.github/workflows/prod-image-build.yml
     with:
       build-type: "pip"
+      upload-package-artifact: "false"
       image-tag: mysql-${{ inputs.image-tag }}
       ci-image-tag: ${{ inputs.image-tag }}
       install-mysql-client-type: "mysql"

.github/workflows/push-image-cache.yml

@@ -174,15 +174,11 @@ jobs:
         uses: ./.github/actions/breeze
       - name: "Cleanup dist and context file"
         run: rm -fv ./dist/* ./docker-context-files/*
-      - name: "Prepare airflow package for PROD build"
-        run: breeze release-management prepare-airflow-package --package-format wheel
-      - name: "Prepare providers packages for PROD build"
-        run: >
-          breeze release-management prepare-provider-packages
-          --package-list-file ./prod_image_installed_providers.txt
-          --package-format wheel
-      - name: "Copy dist packages to docker-context files"
-        run: cp -v --no-preserve=mode,ownership ./dist/*.whl ./docker-context-files
+      - name: "Download packages prepared as artifacts"
+        uses: actions/download-artifact@v4
+        with:
+          name: prod-packages
+          path: ./docker-context-files
       - name: "Start ARM instance"
         run: ./scripts/ci/images/ci_start_arm_instance_and_connect_to_docker.sh
         if: inputs.platform == 'linux/arm64'