unstructured/.github/workflows/ci.yml

name: CI

on:
  # NOTE(robinson) - We are limiting when we run CI avoid exceeding our 2,000 min/month limt.
  # We can switch to running on push if we make this repo public or are fine with
  # paying for CI minutes.
  push:
    branches: [ main ]
  pull_request:
    branches: [ main ]

env:
  PYTHON_VERSION: 3.8

jobs:
  setup:
    runs-on: ubuntu-latest
    env:
      NLTK_DATA: ${{ github.workspace }}/nltk_data
    steps:
    - uses: actions/checkout@v3
    - uses: actions/cache@v3
      id: virtualenv-cache
      with:
        path: |
          .venv
          nltk_data
        key: unstructured-${{ runner.os }}-${{ env.PYTHON_VERSION }}-${{ hashFiles('requirements/*.txt') }}
    - name: Set up Python ${{ env.PYTHON_VERSION }}
      uses: actions/setup-python@v4
      with:
        python-version: ${{ env.PYTHON_VERSION }}
    - name: Setup virtual environment (no cache hit)
      if: steps.virtualenv-cache.outputs.cache-hit != 'true'
      run: |
        python${{ env.PYTHON_VERSION }} -m venv .venv
        source .venv/bin/activate
        make install-ci

  lint:
    runs-on: ubuntu-latest
    needs: setup
    steps:
    - uses: actions/checkout@v3
    - uses: actions/cache@v3
      id: virtualenv-cache
      with:
        path: .venv
        key: unstructured-${{ runner.os }}-${{ env.PYTHON_VERSION }}-${{ hashFiles('requirements/*.txt') }}
    # NOTE(robinson) - This is a fallback in case the lint job does not find the cache.
    # We can take this out when we implement the fix in CORE-99
    - name: Set up Python ${{ env.PYTHON_VERSION }}
      uses: actions/setup-python@v4
      with:
        python-version: ${{ env.PYTHON_VERSION }}
    - name: Setup virtual environment (no cache hit)
      if: steps.virtualenv-cache.outputs.cache-hit != 'true'
      run: |
        python${{ env.PYTHON_VERSION }} -m venv .venv
        source .venv/bin/activate
        make install-ci
    - name: Lint
      run: |
        source .venv/bin/activate
        make check

  shellcheck:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2
      - name: ShellCheck
        uses: ludeeus/action-shellcheck@master

  test:
    runs-on: ubuntu-latest
    env:
      NLTK_DATA: ${{ github.workspace }}/nltk_data
    needs: [setup, lint]
    steps:
    - uses: actions/checkout@v3
    - uses: actions/cache@v3
      id: virtualenv-cache
      with:
        path: |
          .venv
          nltk_data
        key: unstructured-${{ runner.os }}-${{ env.PYTHON_VERSION }}-${{ hashFiles('requirements/*.txt') }}
    # NOTE(robinson) - This is a fallback in case the lint job does not find the cache.
    # We can take this out when we implement the fix in CORE-99
    - name: Setup virtual environment (no cache hit)
      if: steps.virtualenv-cache.outputs.cache-hit != 'true'
      run: |
        python${{ env.PYTHON_VERSION }} -m venv .venv
        source .venv/bin/activate
        make install-ci
    - name: Test
      run: |
        source .venv/bin/activate
        make install-nltk-models
        make install-detectron2
        sudo apt-get install -y libmagic-dev poppler-utils tesseract-ocr libreoffice
        make test
        make check-coverage
        make install-ingest-s3
        ./test_unstructured_ingest/test-ingest.sh

  changelog:
    runs-on: ubuntu-latest
    steps:
    - if: github.ref != 'refs/heads/main'
      uses: dorny/paths-filter@v2
      id: changes
      with:
        filters: |
          src:
            - 'unstructured/**'

    - if: steps.changes.outputs.src == 'true' && github.ref != 'refs/heads/main'
      uses: dangoslen/changelog-enforcer@v3