Unit Testing Skill

This skill provides instructions on how to run and add unit tests in the doug-dashboard project.

Running Tests

To run all unit tests, execute the following command from the project root:

uv run python -m pytest src/tests

To run a specific test file:

uv run python -m pytest src/tests/path/to/test_file.py

Creating New Tests

When adding new functionality, you must create unit tests to verify it.

1. Directory Structure & Naming

Mirror the src/ directory structure within src/tests/.

• •Source: src/assets/work/github/github.py
• •Test: src/tests/assets/work/github/test_github.py

2. Test Data Management (Fixtures)

Do not hardcode large data structures in test functions. Use pytest fixtures.

• •Global Fixtures (src/tests/conftest.py): Place mocks for shared resources (e.g., dagster.AssetExecutionContext, s3fs).
• •Domain Fixtures (src/tests/.../conftest.py): Place dataset-specific fixtures (e.g., sample GitHub events, Spotify API responses) in a conftest.py within the closest relevant subdirectory.

3. Mocking External Dependencies

Use unittest.mock to mock external API calls and resources.

• •Avoid making real network requests during tests.
• •Mock Dagster resources (GithubResource, SpotifyResource) to return the fixture data.

4. Example Pattern

def test_process_event(mock_resource, mock_context, sample_data):
    # Setup
    mock_resource.get_item.return_value = sample_data
    
    # Execution
    result = process_event(mock_resource, mock_context)
    
    # Verification
    assert result["status"] == "success"
    mock_context.log.info.assert_called()