How to Measure Code Coverage in Python Easily

You can measure code coverage in Python using the coverage tool, which tracks how much of your code is tested by running your tests with coverage run and then generating a report with coverage report or coverage html. This helps you see which parts of your code are covered by tests and which are not.

Syntax

To measure code coverage, you use the coverage command-line tool with these main steps:

• coverage run -m unittest runs your tests while tracking coverage.
• coverage report shows a summary of coverage in the terminal.
• coverage html creates a detailed HTML report you can open in a browser.

Each part means:

• coverage run: Runs your test code and collects coverage data.
• -m unittest: Runs Python's built-in test framework (you can replace this with your test command).
• coverage report: Prints coverage stats in the terminal.
• coverage html: Generates a visual report in HTML format.

bash

coverage run -m unittest
coverage report
coverage html

Example

This example shows how to measure coverage for a simple Python function and its test using coverage and unittest.

python

def add(a, b):
    return a + b

import unittest

class TestAdd(unittest.TestCase):
    def test_add(self):
        self.assertEqual(add(2, 3), 5)

if __name__ == '__main__':
    import coverage
    cov = coverage.Coverage()
    cov.start()

    unittest.main(exit=False)

    cov.stop()
    cov.save()
    cov.report()

Output

Name        Stmts   Miss  Cover
--------------------------------
__main__        7      0   100%

----------------------------------------------------------------------
Ran 1 test in 0.000s

OK

Common Pitfalls

Common mistakes when measuring code coverage include:

• Not running tests through coverage run, so no coverage data is collected.
• Forgetting to save coverage data with cov.save() when using the API.
• Running tests without coverage tracking, leading to zero coverage reports.
• Ignoring generated HTML reports which provide clearer insights than terminal output.

Always ensure your tests run inside the coverage tool and check reports carefully.

python

## Wrong way (no coverage tracking):
import unittest

class TestAdd(unittest.TestCase):
    def test_add(self):
        self.assertEqual(1 + 1, 2)

unittest.main()


## Right way (with coverage):
import coverage
cov = coverage.Coverage()
cov.start()

import unittest

class TestAdd(unittest.TestCase):
    def test_add(self):
        self.assertEqual(1 + 1, 2)

unittest.main(exit=False)

cov.stop()
cov.save()
cov.report()

Quick Reference

Here is a quick cheat sheet for measuring code coverage in Python:

Command	Description
coverage run -m unittest	Run tests with coverage tracking
coverage report	Show coverage summary in terminal
coverage html	Generate detailed HTML coverage report
coverage erase	Clear previous coverage data
coverage combine	Combine data from multiple runs

Key Takeaways

Use the coverage.py tool to track which parts of your Python code are tested.

Run your tests with coverage using 'coverage run' to collect data.

Generate reports with 'coverage report' for terminal output or 'coverage html' for detailed browser view.

Always ensure coverage data is saved when using the API to avoid empty reports.

Check HTML reports for clear visualization of covered and missed code lines.