Python coverage: coverage.py, SlipCover, and Supercov

Choose by what you need to measure

coverage.py is a useful starting point for Python line and branch coverage. pytest-cov integrates that engine with pytest; it is not a separate coverage engine. SlipCover is an alternative collector for line and branch coverage. Supercov adds MC/DC and source-to-assertion review to a runner-aware workflow.

Coverage features

Feature	coverage.py7.16.0	SlipCover1.1.0	Supercov0.0.47
Line coverage	Supported	Supported	Supported
Branch coverage	Supported	Supported	Supported
MC/DC	Not built in	Not built in	Supported
Per-test coverage	Supported	Not built in	Supported
Assertion coverage	Not built in	Not built in	Supported

✓ Supported · × Not built in. Standard coverage workflows in the versions shown; custom plugins are outside this table.

Supercov requires CPython 3.12+ for full coverage. pytest-cov uses coverage.py; enable test contexts for per-test coverage.

Per-test coverage means individual tests, not just test files. Supercov needs a supported runner. Assertion coverage links source to checked properties through agent review; it is not a mutation score or a correctness guarantee.

coverage.py and pytest-cov

For an existing pytest project with both packages installed:

pytest --cov=src --cov-branch --cov-context=test

Replace src with your application’s package or directory. The last option records which tests exercised the code. This matters if you want to find the test responsible for a covered line, rather than see only a combined percentage. See pytest-cov’s contexts guide.

You can also run a unittest suite directly through coverage.py:

coverage run --branch -m unittest discover
coverage report -m

coverage.py supports measurement contexts. Supercov is not unique simply because it associates code with tests.

Where SlipCover fits

SlipCover offers another way to collect line and branch coverage, including a pytest launch path:

python -m slipcover --branch -m pytest

Its documentation describes supported Python versions and output options. If you evaluate it, use your real suite and the same source scope. A different collector’s percentage is not itself a test-quality score.

What branch coverage does not tell you

def can_access(signed_in, expired):
    if signed_in and not expired:
        return True
    return False

assert can_access(True, False)
assert not can_access(False, False)

This checks both return paths, but not an expired session. The missing case is assert not can_access(True, True).

On our equivalent unittest fixture, coverage.py 7.16.0 and SlipCover 1.1.0 both reported 100% line and branch coverage. Supercov 0.0.46 reported one of two independent conditions covered. MC/DC makes that missing independence case explicit.

When to use Supercov

With Supercov installed in your Python environment:

supercov -- python -m pytest
supercov runs latest gaps

Use it when your question goes beyond “did this line run?” to “did each condition matter?” or “which result did a test actually check?” Full Python measurement requires CPython 3.12 or newer. See Python setup.

Assertion coverage uses an agent’s review of recorded execution and passing assertions. It is not a mutation score or a correctness proof. And Supercov is not the only Python MC/DC option: dedicated analyzers such as pymcdc address that narrower question too.