pytest-cov — Measure & Visualise Coverage

• Run coverage through pytest with --cov.
• Generate and browse an HTML coverage report.
• Fail the run under a threshold with --cov-fail-under.
• Configure it once in pyproject.toml.

pip install pytest-cov

pytest --cov=myapp            # run tests + show coverage for the myapp package

Terminal report (with missing lines)

$ pytest --cov=myapp --cov-report=term-missing
...
Name              Stmts   Miss  Cover   Missing
---------------------------------------------------
myapp/cart.py        20      0   100%
myapp/grades.py       6      2    67%   2, 6
---------------------------------------------------
TOTAL                26      2    92%

HTML report — click through your code

$ pytest --cov=myapp --cov-report=html
# creates htmlcov/index.html
# open it: covered lines green, missed lines RED, line-by-line

The HTML view is the best way to find gaps — red lines jump out, and you click into each file to see exactly what wasn't run.

Branch coverage too

pytest --cov=myapp --cov-branch --cov-report=term-missing
# Missing now shows things like "8->10" (a branch from line 8 to 10 not taken)

Fail the build under a threshold

pytest --cov=myapp --cov-fail-under=80
# exits non-zero (build FAILS) if total coverage < 80%

This is how teams keep coverage from rotting — a PR that drops coverage below the bar can't merge (CI, Lesson 46).

Configure once in pyproject.toml

[tool.pytest.ini_options]
addopts = "--cov=myapp --cov-branch --cov-report=term-missing --cov-fail-under=80"

[tool.coverage.run]
omit = ["*/tests/*", "*/__init__.py"]   # don't measure test files themselves

Now a bare pytest runs everything with coverage, branch tracking, and the threshold — no flags to remember.

shop/
├─ shop/
│   ├─ __init__.py
│   ├─ cart.py
│   └─ pricing.py
├─ tests/
│   ├─ test_cart.py
│   └─ test_pricing.py
└─ pyproject.toml

# pyproject.toml
[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = """
  --cov=shop
  --cov-branch
  --cov-report=term-missing
  --cov-report=html
  --cov-fail-under=85
"""

[tool.coverage.run]
omit = ["*/tests/*", "*/__init__.py"]

$ pytest
...
Name               Stmts   Miss  Cover   Missing
-----------------------------------------------------
shop/cart.py          20      0   100%
shop/pricing.py       12      1    92%   18
-----------------------------------------------------
TOTAL                 32      1    97%

Required test coverage of 85% reached. Total: 97%
======================== 14 passed in 0.06s =========================
# htmlcov/index.html written — open to see line 18 in red

Read the diff

One pytest command now: runs tests, measures branch coverage, prints missing lines, writes an HTML report, and enforces an 85% floor. The team gets fast feedback AND a guardrail — if someone's PR drops coverage below 85%, the run fails. The HTML report turns "line 18" into a clickable red line you can immediately go test.

pytest --cov=pkg measures coverage during the test run. Add --cov-report=html for a clickable, colour-coded report, --cov-branch for branch coverage, and --cov-fail-under=N to fail the build below a threshold. Configure it all in pyproject.toml. The HTML report + a gate is the professional setup. Next: how much coverage is actually enough.