Coverage thresholds in PyTest

Introduction

Coverage thresholds help ensure your tests check enough of your code. They warn you if tests miss too much code.

When you want to make sure your tests cover at least 80% of your code.

When you want to prevent new code from lowering your test coverage.

When you want to fail a test run if coverage is below a set limit.

When you want to track coverage progress over time.

When you want to enforce quality rules in a team project.

Syntax

PyTest

[run]
branch = True

[report]
fail_under = 80

This example is from a pytest.ini file configuring coverage.

fail_under sets the minimum coverage percent required.

Examples

Set coverage threshold to 90%. Test run fails if coverage is below 90%.

PyTest

[run]
branch = True

[report]
fail_under = 90

Set threshold to 75% and show which lines are missing coverage.

PyTest

[run]
branch = True

[report]
fail_under = 75
show_missing = True

Set threshold to 80% and skip files fully covered in the report.

PyTest

[run]
branch = True

[report]
fail_under = 80
skip_covered = True

Sample Program

This config sets a coverage threshold of 80%. When you run pytest with coverage, it checks if coverage is at least 80%. If not, the test run fails.

PyTest

[run]
branch = True

[report]
fail_under = 80

# Run pytest with coverage:
# pytest --cov=my_module

# If coverage is below 80%, pytest will fail the test run.

Important Notes

Coverage thresholds help keep your tests effective over time.

You can set different thresholds for lines, branches, or functions.

Use pytest --cov-report=term-missing to see which lines lack tests.

Summary

Coverage thresholds set a minimum test coverage percentage.

If coverage is below the threshold, the test run fails.

This helps maintain good test quality and catch missing tests early.