ci.yml

name: CI

on:
  pull_request:
    branches: [ main ]
    paths:
      - '.github/workflows/ci.yml'
      - 'install.sh'
      - 'src/**'
      - 'tests/**'
      - 'pyproject.toml'

jobs:
  test:
    name: Test on ${{ matrix.os }} / Python ${{ matrix.python-version }}
    runs-on: ${{ matrix.os }}
    strategy:
      fail-fast: false
      matrix:
        os: [ubuntu-latest, macos-15, macos-26]
        python-version: ['3.11', '3.12', '3.13']

    steps:
    - name: Free Disk Space (Ubuntu)
      if: runner.os == 'Linux'
      uses: jlumbroso/free-disk-space@v1.3.1
      with:
        tool-cache: false
        android: true
        dotnet: true
        haskell: true
        large-packages: true
        docker-images: true
        swap-storage: false

    - name: Checkout code
      uses: actions/checkout@v4

    - name: Set up Python ${{ matrix.python-version }}
      uses: actions/setup-python@v5
      with:
        python-version: ${{ matrix.python-version }}

    - name: Cache pip dependencies
      uses: actions/cache@v4
      with:
        path: ~/.cache/pip
        key: ${{ matrix.os }}-pip-${{ matrix.python-version }}-${{ hashFiles('pyproject.toml') }}
        restore-keys: |
          ${{ matrix.os }}-pip-${{ matrix.python-version }}-
          ${{ matrix.os }}-pip-

    - name: Install dependencies
      run: |
        python -m pip install --upgrade pip
        # Install extras dependencies based on matrix variable
        if [[ "${{ runner.os }}" == "Linux" ]]; then
          pip install -e ".[gpu, dev]"
        else
          pip install -e ".[mac, dev]"
        fi

    - name: Run Unit Tests
      shell: bash
      run: |
        pytest tests/ -v --cov=src/parallax --cov-report=xml

    - name: Upload coverage to Codecov
      if: matrix.os == 'macos-15' && matrix.python-version == '3.11'
      uses: codecov/codecov-action@v4
      with:
        file: ./coverage.xml
        fail_ci_if_error: false
        token: ${{ secrets.CODECOV_TOKEN }}

    - name: Install Parallax and vLLM Rust frontend
      if: matrix.os == 'macos-26' && matrix.python-version == '3.12'
      shell: bash
      run: |
        ./install.sh --extras mac --python "${{ matrix.python-version }}"
        test -x .venv/bin/vllm-rs

    - name: Run E2E tests (macOS only)
      if: matrix.os == 'macos-26' && matrix.python-version == '3.12'
      shell: bash
      env:
        TERM: xterm-256color
      run: |
        # Start the server
        .venv/bin/python src/parallax/launch.py \
          --model-path Qwen/Qwen3-0.6B \
          --max-num-tokens-per-batch 16384 \
          --kv-block-size 32 \
          --max-batch-size 128 \
          --start-layer 0 \
          --end-layer 28 &
        PID=$!

        echo "Waiting for server to start..."
        # Poll to check if the port is ready (wait up to 120 seconds)
        for i in {1..60}; do
          # If curl succeeds (200) or returns 405 (Method Not Allowed), the port is open
          if curl -s --max-time 5 --connect-timeout 2 -o /dev/null -w "%{http_code}" http://localhost:3000/v1/chat/completions | grep -qE "200|400|405"; then
             echo "Server is up!"
             break
          fi

          # Check if the process is still alive
          if ! kill -0 $PID 2>/dev/null; then
             echo "Server process died prematurely"
             exit 1
          fi

          if [ $i -eq 60 ]; then
             echo "Server failed to start within 120 seconds"
             kill $PID 2>/dev/null
             exit 1
          fi
          sleep 2
        done

        # Wait additional time to ensure server is fully ready (model loaded, KV cache allocated, etc.)
        echo "Waiting for server to be fully ready..."
        sleep 10

        echo "Sending test request..."
        # Capture the response with increased timeout (5 minutes for model inference)
        RESPONSE=$(curl --fail --silent --show-error --location \
          --max-time 300 \
          --connect-timeout 10 \
          'http://localhost:3000/v1/chat/completions' \
          --header 'Content-Type: application/json' \
          --data '{
            "model": "Qwen/Qwen3-0.6B",
            "messages": [
              {
                "role": "user",
                "content": "What is the capital of France"
              }
            ],
            "stream": false,
            "max_tokens": 1024,
            "chat_template_kwargs": {"enable_thinking": false},
            "top_k": 3
        }')

        echo "Response received:"
        echo "$RESPONSE"

        # Check if the response contains "Paris" (case-insensitive)
        if echo "$RESPONSE" | grep -iq "Paris"; then
            echo "Test passed: Response contains 'Paris'"
        else
            echo "Test failed: Response does not contain 'Paris'"
            kill $PID 2>/dev/null || true
            exit 1
        fi

        # Clean up process
        kill $PID 2>/dev/null || true