all_tests_common.yml

name: Common All Tests

on:
  workflow_call:
    inputs:
      platform:
        required: true
        type: string
        description: Platform name (e.g., cuda, default)
      ci_train_image:
        required: false
        type: string
        description: Override train image (e.g., newly built image). Falls back to platform config if not set.
        default: ""
      ci_inference_image:
        required: false
        type: string
        description: Override inference image. Falls back to platform config if not set.
        default: ""
      runs_on:
        required: false
        type: string
        description: Override runs_on. Falls back to platform config if not set.
        default: ""
      container_volumes:
        required: false
        type: string
        description: Override container_volumes. Falls back to platform config if not set.
        default: ""

jobs:
  checkout_and_config:
    defaults:
      run:
        shell: bash
    runs-on: ubuntu-latest
    outputs:
      # ci_image: ${{ steps.config.outputs.ci_image }}
      ci_train_image: ${{ steps.config.outputs.ci_train_image }}
      ci_inference_image: ${{ steps.config.outputs.ci_inference_image }}
      runs_on: ${{ steps.config.outputs.runs_on }}
      container_volumes: ${{ steps.config.outputs.container_volumes }}
      container_options: ${{ steps.config.outputs.container_options }}
      device_types: ${{ steps.config.outputs.device_types }}
      train_test_matrix: ${{ steps.config.outputs.train_test_matrix }}
      hetero_train_test_matrix: ${{ steps.config.outputs.hetero_train_test_matrix }}
      inference_test_matrix: ${{ steps.config.outputs.inference_test_matrix }}
      serve_test_matrix: ${{ steps.config.outputs.serve_test_matrix }}
      rl_test_matrix: ${{ steps.config.outputs.rl_test_matrix }}
      benchmark_test_matrix: ${{ steps.config.outputs.benchmark_test_matrix }}
      pkg_mgr: ${{ steps.config.outputs.pkg_mgr }}
      env_path: ${{ steps.config.outputs.env_path }}
      env_name_train: ${{ steps.config.outputs.env_name_train }}
      env_name_inference: ${{ steps.config.outputs.env_name_inference }}
      env_name_serve: ${{ steps.config.outputs.env_name_serve }}
      env_name_rl: ${{ steps.config.outputs.env_name_rl }}
    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
          ref: ${{ github.event.pull_request.head.ref || github.ref }}
          ssh-strict: true
          ssh-user: git
          persist-credentials: false
          clean: true
          sparse-checkout-cone-mode: true
          fetch-tags: false
          show-progress: true
          lfs: false
          submodules: false
          set-safe-directory: true

      - name: Cache source code
        id: cache-source
        uses: actions/cache@v4
        with:
          path: .
          key: source-code-${{ github.sha }}

      - name: Create source code tarball for distribution
        run: |
          tar -czf /tmp/flagscale-source.tar.gz \
            --exclude='.git' \
            --exclude='*.pyc' \
            --exclude='__pycache__' \
            .

      - name: Upload source code artifact
        uses: actions/upload-artifact@v4
        with:
          name: flagscale-source-${{ github.sha }}
          path: /tmp/flagscale-source.tar.gz
          retention-days: 1
          compression-level: 0

      - name: Load platform configuration
        id: config
        run: |
          set -euo pipefail

          PLATFORM="${{ inputs.platform }}"
          CONFIG_FILE=".github/configs/${PLATFORM}.yml"

          if [ ! -f "$CONFIG_FILE" ]; then
            echo "❌ Error: Platform configuration file not found: $CONFIG_FILE"
            echo "Available configs:"
            ls -la .github/configs/
            exit 1
          fi

          # Install mikefarah/yq (v4) for YAML parsing
          sudo wget -qO /usr/local/bin/yq https://github.com/mikefarah/yq/releases/download/v4.45.1/yq_linux_amd64
          sudo chmod +x /usr/local/bin/yq
          /usr/local/bin/yq --version

          # Source the platform config loading script
          source ./tools/install/utils/load_platform_config.sh

          # Load configuration and group tests by task
          load_platform_config "$PLATFORM"

          # Override images if provided as inputs
          if [ -n "${{ inputs.ci_train_image }}" ]; then
            echo "ci_train_image=${{ inputs.ci_train_image }}" >> $GITHUB_OUTPUT
          fi
          if [ -n "${{ inputs.ci_inference_image }}" ]; then
            echo "ci_inference_image=${{ inputs.ci_inference_image }}" >> $GITHUB_OUTPUT
          fi
          # Use single-quoted assignment so bash treats the JSON value literally
          # (double quotes inside the JSON would break echo "runs_on=${{ inputs.runs_on }}")
          RUNS_ON_INPUT='${{ inputs.runs_on }}'
          if [ -n "$RUNS_ON_INPUT" ]; then
            { echo 'runs_on<<EOFRUNSON_INPUT'; echo "$RUNS_ON_INPUT"; echo 'EOFRUNSON_INPUT'; } >> $GITHUB_OUTPUT
          fi
          # (double quotes inside the JSON would break echo "container_volumes=${{ inputs.container_volumes }}")
          CONTAINER_VOLUMES_INPUT='${{ inputs.container_volumes }}'
          if [ -n "$CONTAINER_VOLUMES_INPUT" ]; then
            { echo 'container_volumes<<EOFRUNSON_INPUT'; echo "$CONTAINER_VOLUMES_INPUT"; echo 'EOFRUNSON_INPUT'; } >> $GITHUB_OUTPUT
          fi

  # CLI validation runs first (outside virtual env) as a gate for all subsequent tests
  cli_validation:
    needs: checkout_and_config
    uses: ./.github/workflows/functional_tests_cli.yml
    with:
      platform: ${{ inputs.platform }}
      image: ${{ needs.checkout_and_config.outputs.ci_train_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  unit_tests:
    needs:
      - checkout_and_config
      - cli_validation
    if: fromJson(needs.checkout_and_config.outputs.device_types)[0] != null
    strategy:
      fail-fast: false
      matrix:
        device: ${{ fromJson(needs.checkout_and_config.outputs.device_types) }}
    uses: ./.github/workflows/unit_tests_common.yml
    name: unit_tests
    with:
      platform: ${{ inputs.platform }}
      device: ${{ matrix.device }}
      image: ${{ needs.checkout_and_config.outputs.ci_train_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  functional_tests_train:
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
    if: fromJson(needs.checkout_and_config.outputs.train_test_matrix)[0] != null
    uses: ./.github/workflows/functional_tests_train.yml
    with:
      platform: ${{ inputs.platform }}
      test_matrix: ${{ needs.checkout_and_config.outputs.train_test_matrix }}
      image: ${{ needs.checkout_and_config.outputs.ci_train_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  functional_tests_hetero_train:
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
    if: fromJson(needs.checkout_and_config.outputs.hetero_train_test_matrix)[0] != null
    uses: ./.github/workflows/functional_tests_hetero_train.yml
    with:
      platform: ${{ inputs.platform }}
      test_matrix: ${{ needs.checkout_and_config.outputs.hetero_train_test_matrix }}
      image: ${{ needs.checkout_and_config.outputs.ci_train_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  functional_tests_inference:
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
    if: fromJson(needs.checkout_and_config.outputs.inference_test_matrix)[0] != null
    uses: ./.github/workflows/functional_tests_inference.yml
    with:
      platform: ${{ inputs.platform }}
      test_matrix: ${{ needs.checkout_and_config.outputs.inference_test_matrix }}
      image: ${{ needs.checkout_and_config.outputs.ci_inference_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_inference }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  functional_tests_serve:
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
    if: fromJson(needs.checkout_and_config.outputs.serve_test_matrix)[0] != null
    uses: ./.github/workflows/functional_tests_serve.yml
    with:
      platform: ${{ inputs.platform }}
      test_matrix: ${{ needs.checkout_and_config.outputs.serve_test_matrix }}
      image: ${{ needs.checkout_and_config.outputs.ci_inference_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_serve }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  # NOTE: rl functional tests are temporarily disabled
  # functional_tests_rl:
  #   needs:
  #     - checkout_and_config
  #     - unit_tests
  #   if: fromJson(needs.checkout_and_config.outputs.rl_test_matrix)[0] != null
  #   uses: ./.github/workflows/functional_tests_rl.yml
  #   with:
  #     platform: ${{ inputs.platform }}
  #     test_matrix: ${{ needs.checkout_and_config.outputs.rl_test_matrix }}
  #     image: ${{ needs.checkout_and_config.outputs.ci_image }}
  #     runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
  #     container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
  #     container_options: ${{ needs.checkout_and_config.outputs.container_options }}
  #     source_artifact: flagscale-source-${{ github.sha }}
  #     pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
  #     env_name: ${{ needs.checkout_and_config.outputs.env_name_rl }}
  #     env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  functional_tests_benchmark:
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
    if: fromJson(needs.checkout_and_config.outputs.benchmark_test_matrix)[0] != null
    uses: ./.github/workflows/functional_tests_benchmark.yml
    with:
      platform: ${{ inputs.platform }}
      test_matrix: ${{ needs.checkout_and_config.outputs.benchmark_test_matrix }}
      image: ${{ needs.checkout_and_config.outputs.ci_train_image }}
      runs_on: ${{ needs.checkout_and_config.outputs.runs_on }}
      container_volumes: ${{ needs.checkout_and_config.outputs.container_volumes }}
      container_options: ${{ needs.checkout_and_config.outputs.container_options }}
      source_artifact: flagscale-source-${{ github.sha }}
      pkg_mgr: ${{ needs.checkout_and_config.outputs.pkg_mgr }}
      env_name: ${{ needs.checkout_and_config.outputs.env_name_train }}
      env_path: ${{ needs.checkout_and_config.outputs.env_path }}

  all_tests_complete:
    defaults:
      run:
        shell: bash
    needs:
      - checkout_and_config
      - cli_validation
      - unit_tests
      - functional_tests_train
      - functional_tests_hetero_train
      - functional_tests_benchmark
      - functional_tests_inference
      - functional_tests_serve
      # NOTE: Disabled tests removed from needs
      # - functional_tests_rl
    runs-on: ubuntu-latest
    if: always()
    steps:
      - name: Verify all tests passed
        run: |
          # Check all test jobs (skip if not run)
          failed=false

          if [ "${{ needs.unit_tests.result }}" != "success" ] && \
             [ "${{ needs.unit_tests.result }}" != "skipped" ]; then
            echo "❌ Unit tests failed"
            failed=true
          fi

          if [ "${{ needs.cli_validation.result }}" != "success" ] && \
             [ "${{ needs.cli_validation.result }}" != "skipped" ]; then
            echo "❌ CLI validation failed"
            failed=true
          fi

          # Only check functional tests if they ran
          if [ "${{ needs.functional_tests_train.result }}" != "success" ] && \
             [ "${{ needs.functional_tests_train.result }}" != "skipped" ]; then
            echo "❌ Training functional tests failed"
            failed=true
          fi

          if [ "${{ needs.functional_tests_hetero_train.result }}" != "success" ] && \
             [ "${{ needs.functional_tests_hetero_train.result }}" != "skipped" ]; then
            echo "❌ Heterogeneous training functional tests failed"
            failed=true
          fi

          if [ "${{ needs.functional_tests_benchmark.result }}" != "success" ] && \
             [ "${{ needs.functional_tests_benchmark.result }}" != "skipped" ]; then
            echo "❌ Benchmark tests failed"
            failed=true
          fi

          if [ "${{ needs.functional_tests_inference.result }}" != "success" ] && \
             [ "${{ needs.functional_tests_inference.result }}" != "skipped" ]; then
            echo "❌ Inference functional tests failed"
            failed=true
          fi

          if [ "${{ needs.functional_tests_serve.result }}" != "success" ] && \
             [ "${{ needs.functional_tests_serve.result }}" != "skipped" ]; then
            echo "❌ Serve functional tests failed"
            failed=true
          fi

          # NOTE: Inference, serve, and rl checks disabled
          # if [ "${{ needs.functional_tests_rl.result }}" != "success" ] && \
          #    [ "${{ needs.functional_tests_rl.result }}" != "skipped" ]; then
          #   echo "❌ RL functional tests failed"
          #   failed=true
          # fi

          if [ "$failed" = "true" ]; then
            exit 1
          fi

          echo "✅ All tests completed successfully!"