python-testing

Use @pytest.mark.parametrize with argument names and a list of value tuples to run one test function against multiple inputs. You can add ids to label each case, making failures easier to identify in output.

Define a fixture that creates an in-memory SQLite session, yields it to the test, and closes it afterward. Choose a scope such as function, class, module, or session depending on how often setup should run.

unittest.mock provides Mock, patch, and MagicMock from the standard library, while pytest-mock wraps these in a mocker fixture with cleaner syntax. Both let you stub dependencies and assert on call counts and arguments.

Yes, with the pytest-asyncio plugin you mark coroutine tests with @pytest.mark.asyncio and can define async fixtures that yield connected clients. Without the plugin, async tests are skipped or fail.

Run pytest with --cov=src and --cov-fail-under=80, or set these in pytest.ini or pyproject.toml addopts. The test run fails if coverage drops below the configured percentage.

Use hypothesis for property-based testing when you want generated inputs to verify invariants like commutativity or idempotence across many cases. Use parametrize when you have specific known inputs and expected outputs.