reuse-get-test-matrix.yml

# Copyright (C) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

# Support push and pull_request events
name: Get Test Matrix
permissions: read-all
on:
  workflow_call:
    outputs:
      run_matrix:
        description: "The matrix string"
        value: ${{ jobs.job1.outputs.run_matrix }}

jobs:
  job1:
    name: Get-test-matrix
    runs-on: ubuntu-latest
    outputs:
      run_matrix: ${{ steps.get-test-matrix.outputs.run_matrix }}
    steps:
      - name: Get checkout ref
        run: |
          if [ "${{ github.event_name }}" == "pull_request" ] || [ "${{ github.event_name }}" == "pull_request_target" ]; then
            echo "CHECKOUT_REF=refs/pull/${{ github.event.number }}/merge" >> $GITHUB_ENV
          else
            echo "CHECKOUT_REF=${{ github.ref }}" >> $GITHUB_ENV
          fi
          echo "checkout ref ${{ env.CHECKOUT_REF }}"

      - name: Checkout out Repo
        uses: actions/checkout@v4
        with:
          ref: ${{ env.CHECKOUT_REF }}
          fetch-depth: 0

      - name: Get test matrix
        id: get-test-matrix
        run: |
          set -xe
          if [ "${{ github.event_name }}" == "pull_request" ] || [ "${{ github.event_name }}" == "pull_request_target" ]; then
            base_commit=${{ github.event.pull_request.base.sha }}
          else
            base_commit=$(git rev-parse HEAD~1) # push event
          fi
          merged_commit=$(git log -1 --format='%H')

          changed_files="$(git diff --name-only ${base_commit} ${merged_commit} | \
          grep 'comps/' | grep -vE '*.md|*.txt|comps/cores')" || true
          services=$(printf '%s\n' "${changed_files[@]}" | cut -d'/' -f2 | grep -vE '*.py' | sort -u) || true
          path_level_1=("asr" "tts")
          path_level_3=("llms/summarization" "llms/text-generation" "dataprep/redis" "retrievers/langchain")
          run_matrix="{\"include\":["
          for service in ${services}; do
              hardware="gaudi" # default hardware, set based on the changed files
              if [[ "${path_level_1[@]}" =~ "${service}" ]]; then
                  run_matrix="${run_matrix}{\"service\":\"${service}\",\"hardware\":\"${hardware}\"},"
              else
                  vendors=$(printf '%s\n' "${changed_files[@]}" | grep ${service} | cut -d'/' -f3 | grep -vE '*.py|Dockerfile|*.md|*.sh' | sort -u)
                  for vendor in ${vendors}; do
                      if [[ "${path_level_3[@]}" =~ "${service}/${vendor}" ]]; then
                          sub_vendors=$(printf '%s\n' "${changed_files[@]}" | grep ${service} | grep ${vendor} | cut -d'/' -f4 | grep -vE '*.py' | sort -u)
                          for sub_vendor in ${sub_vendors}; do
                              run_matrix="${run_matrix}{\"service\":\"${service}/${vendor}/${sub_vendor}\",\"hardware\":\"${hardware}\"},"
                          done
                      else
                          run_matrix="${run_matrix}{\"service\":\"${service}/${vendor}\",\"hardware\":\"${hardware}\"},"
                      fi
                  done
              fi
          done

          # add test for comps/dataprep/utils.py
          if [[ "${changed_files[@]}" =~ "comps/dataprep/utils.py" ]]; then
              service_list=("dataprep/qdrant" "dataprep/redis/langchain")
              hardware="gaudi" # default hardware, set based on the changed files
              for service in ${service_list[@]}; do
                  if [ $(echo ${run_matrix} | grep -c ${service}) == 0 ]; then
                      run_matrix="${run_matrix}{\"service\":\"${service}\",\"hardware\":\"${hardware}\"},"
                  fi
              done
          fi

          # add test for test scripts update
          changed_files="$(git diff --name-only ${base_commit} ${merged_commit} | \
          grep 'tests/' | grep -vE '*.md|*.txt|tests/cores')" || true
          run_matrix_match=$(echo $run_matrix | tr '/' '_')
          test_files=$(printf '%s\n' "${changed_files[@]}" | grep -E "tests/test_*" | grep -E "*.sh") || true
          for test_file in ${test_files}; do
              service=$(echo $test_file | cut -d'/' -f2 | cut -d'.' -f1 | cut -c6-)
              hardware="gaudi" # default hardware, set based on the changed files
              if [ $(echo ${run_matrix_match} | grep -c ${service}) == 0 ]; then
                  run_matrix="${run_matrix}{\"service\":\"${service}\",\"hardware\":\"${hardware}\"},"
              fi
          done

          run_matrix=$run_matrix"]}"
          echo "run_matrix=${run_matrix}" >> $GITHUB_OUTPUT