community_ci.yml

name: Community CI

concurrency:
  # This is the name of the concurrency group. It is used to prevent concurrent runs of the same workflow.
  #
  # - github.head_ref is only defined on PR runs, it makes sure that the concurrency group is unique for pull requests
  #  ensuring that only one run per pull request is active at a time.
  group: ${{ github.workflow }}-${{ github.head_ref }}
  cancel-in-progress: true

on:
  # The pull request target event is triggered on PR open and synchronize events.
  # The executed workflow code is the one from the target branch.
  # It guarantees that the workflow logic is not altered by the PR.
  pull_request_target:
    types: [opened, synchronize]

jobs:
  fail_on_protected_path_changes:
    name: "Check fork do not change protected paths"
    if: github.event.pull_request.head.repo.fork == true
    runs-on: ubuntu-latest
    permissions:
      pull-requests: read
    steps:
      - name: Check for changes in protected paths
        id: check_for_changes_in_protected_paths
        uses: dorny/paths-filter@v2
        with:
          filters: |
            protected_paths:
              - '.github/**'
              - 'airbyte-ci/**'

      - name: Fail if changes in protected paths
        if: steps.check_for_changes_in_protected_paths.outputs.protected_paths == 'true'
        run: |
          echo "The fork has changes in protected paths. This is not allowed."
          exit 1

  format_check:
    # IMPORTANT: This name must match the require check name on the branch protection settings
    name: "Check for formatting errors"
    if: github.event.pull_request.head.repo.fork == true
    environment: community-ci-auto
    runs-on: community-tooling-test-small
    needs: fail_on_protected_path_changes
    timeout-minutes: 30
    env:
      MAIN_BRANCH_NAME: "master"
    steps:
      # This checkouts a fork which can contain untrusted code
      # It's deemed safe as the formatter are not executing any checked out code
      - name: Checkout fork
        uses: actions/checkout@v4
        with:
          repository: ${{ github.event.pull_request.head.repo.full_name }}
          ref: ${{ github.event.pull_request.head.sha }}
          fetch-depth: 1

      # This will sync the .github folder of the main repo with the fork
      # This allows us to use up to date actions and CI logic from the main repo
      - name: Pull .github folder and internal packages from main repository
        id: pull_github_folder
        run: |
          git remote add main https://github.com/airbytehq/airbyte.git
          git fetch main ${MAIN_BRANCH_NAME}
          git checkout main/${MAIN_BRANCH_NAME} -- .github
          git checkout main/${MAIN_BRANCH_NAME} -- airbyte-ci

      - name: Run airbyte-ci format check all
        # This path refers to the fork .github folder.
        # We make sure its content is in sync with the main repo .github folder by pulling it in the previous step
        uses: ./.github/actions/run-airbyte-ci
        with:
          context: "pull_request"
          sentry_dsn: ${{ secrets.SENTRY_AIRBYTE_CI_DSN }}
          subcommand: "format check all"
          is_fork: "true"
  connectors_early_ci:
    name: Run connectors early CI on fork
    if: github.event.pull_request.head.repo.fork == true
    needs: fail_on_protected_path_changes
    environment: community-ci-auto
    runs-on: community-tooling-test-small
    timeout-minutes: 10
    env:
      MAIN_BRANCH_NAME: "master"
    permissions:
      statuses: write
    steps:
      # This checkouts a fork which can contain untrusted code
      # It's deemed safe as the static checks are not executing any checked out code
      - name: Checkout fork
        uses: actions/checkout@v4
        with:
          repository: ${{ github.event.pull_request.head.repo.full_name }}
          ref: ${{ github.event.pull_request.head.sha }}
          fetch-depth: 1

      # This will sync the .github folder of the main repo with the fork
      # This allows us to use up to date actions and CI logic from the main repo
      - name: Pull .github folder from main repository
        id: pull_github_folder
        run: |
          git remote add main https://github.com/airbytehq/airbyte.git
          git fetch main ${MAIN_BRANCH_NAME}
          git checkout main/${MAIN_BRANCH_NAME} -- .github
          git checkout main/${MAIN_BRANCH_NAME} -- airbyte-ci

      - name: Run airbyte-ci static checks and version increment checks on modified connectors
        # This path refers to the fork .github folder.
        # We make sure its content is in sync with the main repo .github folder by pulling it in the previous step
        uses: ./.github/actions/run-airbyte-ci
        with:
          context: "pull_request"
          sentry_dsn: ${{ secrets.SENTRY_AIRBYTE_CI_DSN }}
          subcommand: "connectors --modified test --only-step=qa_checks --only-step=version_inc_check --global-status-check-context='Connectors early CI checks' --global-status-check-description='Running early CI checks on connectors'"
          is_fork: "true"
          git_repo_url: ${{ github.event.pull_request.head.repo.clone_url }}
          git_branch: ${{ github.head_ref }}
          git_revision: ${{ github.event.pull_request.head.sha }}
          github_token: ${{ github.token }}
          docker_hub_password: ${{ secrets.DOCKER_HUB_PASSWORD }}
          docker_hub_username: ${{ secrets.DOCKER_HUB_USERNAME }}
      - name: Upload pipeline reports
        id: upload-artifact
        uses: actions/upload-artifact@v4
        with:
          name: early-ci-pipeline-reports
          path: /home/runner/work/airbyte/airbyte/airbyte-ci/connectors/pipelines/pipeline_reports/airbyte-ci/connectors/test/pull_request/**/output.html
          retention-days: 7

  connectors_full_ci:
    name: Run connectors full CI on fork
    if: github.event.pull_request.head.repo.fork == true
    # Deployment of jobs on the community-ci environment requires manual approval
    # This is something we set up in the GitHub environment settings:
    # https://github.com/airbytehq/airbyte/settings/environments/2091483613/edit
    # This is a safety measure to make sure the code running on our infrastructure has been reviewed before running on it
    needs: fail_on_protected_path_changes
    environment: community-ci
    runs-on: community-connector-test-large
    timeout-minutes: 180 # 3 hours
    permissions:
      statuses: write
    env:
      MAIN_BRANCH_NAME: "master"

    steps:
      # This checkouts a fork which can contain untrusted code
      # It's deemed safe as the community-ci environment requires manual reviewer approval to run
      - name: Checkout fork
        uses: actions/checkout@v4
        with:
          repository: ${{ github.event.pull_request.head.repo.full_name }}
          ref: ${{ github.event.pull_request.head.sha }}
          fetch-depth: 1

      # This will sync the .github folder of the main repo with the fork
      # This allows us to use up to date actions and CI logic from the main repo
      - name: Pull .github folder from main repository
        id: pull_github_folder
        run: |
          git remote add main https://github.com/airbytehq/airbyte.git
          git fetch main ${MAIN_BRANCH_NAME}
          git checkout main/${MAIN_BRANCH_NAME} -- .github
          git checkout main/${MAIN_BRANCH_NAME} -- airbyte-ci

      - name: Run airbyte-ci connectors test
        # This path refers to the fork .github folder.
        # We make sure its content is in sync with the main repo .github folder by pulling it in the previous step
        uses: ./.github/actions/run-airbyte-ci
        with:
          context: "pull_request"
          dagger_cloud_token: ${{ secrets.DAGGER_CLOUD_TOKEN_2 }}
          docker_hub_password: ${{ secrets.DOCKER_HUB_PASSWORD }}
          docker_hub_username: ${{ secrets.DOCKER_HUB_USERNAME }}
          gcp_gsm_credentials: ${{ secrets.GCP_GSM_CREDENTIALS }}
          sentry_dsn: ${{ secrets.SENTRY_AIRBYTE_CI_DSN }}
          git_repo_url: ${{ github.event.pull_request.head.repo.clone_url }}
          git_branch: ${{ github.head_ref }}
          git_revision: ${{ github.event.pull_request.head.sha }}
          github_token: ${{ github.token }}
          s3_build_cache_access_key_id: ${{ secrets.SELF_RUNNER_AWS_ACCESS_KEY_ID }}
          s3_build_cache_secret_key: ${{ secrets.SELF_RUNNER_AWS_SECRET_ACCESS_KEY }}
          subcommand: "connectors --modified test"
          is_fork: "true"
      - name: Upload pipeline reports
        id: upload-artifact
        uses: actions/upload-artifact@v4
        with:
          name: pipeline-reports
          path: /home/runner/work/airbyte/airbyte/airbyte-ci/connectors/pipelines/pipeline_reports/airbyte-ci/connectors/test/pull_request/**/output.html
          retention-days: 7

  internal_poetry_packages_ci:
    name: Internal Poetry packages CI
    if: github.event.pull_request.head.repo.fork == true
    # Deployment of jobs on the community-ci environment requires manual approval
    # This is something we set up in the GitHub environment settings:
    # https://github.com/airbytehq/airbyte/settings/environments/2091483613/edit
    # This is a safety measure to make sure the code running on our infrastructure has been reviewed before running on it
    needs: fail_on_protected_path_changes
    environment: community-ci
    runs-on: community-tooling-test-small
    timeout-minutes: 180 # 3 hours
    permissions:
      statuses: write
    env:
      MAIN_BRANCH_NAME: "master"

    steps:
      # This checkouts a fork which can contain untrusted code
      # It's deemed safe as the community-ci environment requires manual reviewer approval to run
      - name: Checkout fork
        uses: actions/checkout@v4
        with:
          repository: ${{ github.event.pull_request.head.repo.full_name }}
          ref: ${{ github.event.pull_request.head.sha }}
          fetch-depth: 1

      # This will sync the .github folder of the main repo with the fork
      # This allows us to use up to date actions and CI logic from the main repo
      - name: Pull .github folder from main repository
        id: pull_github_folder
        run: |
          git remote add main https://github.com/airbytehq/airbyte.git
          git fetch main ${MAIN_BRANCH_NAME}
          git checkout main/${MAIN_BRANCH_NAME} -- .github
          git checkout main/${MAIN_BRANCH_NAME} -- airbyte-ci

      - name: Run poe tasks for modified internal packages [PULL REQUEST]
        # This path refers to the fork .github folder.
        # We make sure its content is in sync with the main repo .github folder by pulling it in the previous step
        id: run-airbyte-ci-test-pr
        uses: ./.github/actions/run-airbyte-ci
        with:
          context: "pull_request"
          dagger_cloud_token: ${{ secrets.DAGGER_CLOUD_TOKEN_2 }}
          docker_hub_password: ${{ secrets.DOCKER_HUB_PASSWORD }}
          docker_hub_username: ${{ secrets.DOCKER_HUB_USERNAME }}
          gcp_gsm_credentials: ${{ secrets.GCP_GSM_CREDENTIALS }}
          gcs_credentials: ${{ secrets.METADATA_SERVICE_PROD_GCS_CREDENTIALS }}
          git_repo_url: ${{ github.event.pull_request.head.repo.clone_url }}
          git_branch: ${{ github.head_ref }}
          git_revision: ${{ github.event.pull_request.head.sha }}
          github_token: ${{ github.token }}
          subcommand: "test --modified"
          is_fork: "true"