Testing & Debugging

Unit Testing with pytest

pytest discovers test_ functions and assert statements with rich failure output. Fixtures inject dependencies and setup/teardown via yield.

Parametrize runs cases across inputs. conftest.py shares fixtures across modules.

Mark slow or integration tests for selective CI runs with -m markers.

• One logical assertion concept per test function
• Use tmp_path fixture for filesystem tests
• Run pytest --cov for coverage trends, not vanity metrics

def test_discount():
    assert price_after_discount(100, 0.2) == 80

@pytest.mark.parametrize("n,expected", [(1,1),(2,4)])
def test_square(n, expected):
    assert square(n) == expected

Mocking and Debugging

unittest.mock.patch replaces functions during tests. pytest-mock provides mocker fixture wrapping pytest style.

logging module configures structured logs better than print for production. pdb.set_trace() or breakpoint() pauses execution in dev.

IDE debuggers integrate breakpoints and variable inspection for complex flows.

• Mock at boundaries: HTTP clients, database, clock
• Use caplog fixture to assert log messages
• Reproduce bugs with minimal failing tests first