Coverage 101 — What Lines Got Hit?

• Explain what line coverage measures (and what it doesn't).
• Run coverage.py and read the percentage report.
• Find the exact unexecuted lines (Missing column).
• Understand branch coverage vs line coverage.

def grade(score):
    if score >= 80:
        return "A"           # ← did any test hit this?
    elif score >= 50:
        return "B"           # ← or this?
    else:
        return "F"           # ← or this?

# a test for grade(90) only runs the first branch.
# coverage shows you the others were NEVER executed.

Install & run

pip install coverage

coverage run -m pytest        # run tests, recording coverage
coverage report               # print the summary
coverage report -m            # ALSO show the missing line numbers

Reading the report

Name          Stmts   Miss  Cover   Missing
-------------------------------------------------
grades.py        7      2    71%     6-7
cart.py         20      0   100%
-------------------------------------------------
TOTAL           27      2    93%

Stmts    total executable statements
Miss     statements NOT run by any test
Cover    percentage executed
Missing  the exact line numbers never run  ← go test these!

Line coverage vs branch coverage

def f(x):
    if x > 0:
        log("positive")     # line coverage: this line ran
    return x                # but did we test the x <= 0 case (no log)?

line coverage    "was this LINE executed?"
branch coverage  "was each BRANCH (if-taken AND if-skipped) executed?"
                 stricter — catches untested "else" / fall-through paths

coverage run --branch -m pytest    # enable branch coverage

What coverage does NOT mean

100% coverage ≠ bug-free!
A line can run without your test ASSERTING anything about it.
Coverage finds UNtested code; it can't prove tested code is CORRECT.

# grades.py
def grade(score):
    if score < 0 or score > 100:
        raise ValueError("0-100 only")
    if score >= 80: return "A"
    if score >= 50: return "B"
    return "F"

# test_grades.py — incomplete on purpose
from grades import grade
def test_a(): assert grade(90) == "A"
def test_b(): assert grade(60) == "B"

$ coverage run -m pytest && coverage report -m
Name        Stmts   Miss  Cover   Missing
-------------------------------------------------
grades.py       6      2    67%     2, 6
-------------------------------------------------
# line 2 = the ValueError guard; line 6 = the "F" return
# → never tested!

Coverage names the gaps: the error guard and the "F" case. Close them:

import pytest
def test_fail(): assert grade(20) == "F"          # covers line 6
def test_out_of_range():                          # covers line 2
    with pytest.raises(ValueError):
        grade(150)

$ coverage run -m pytest && coverage report
grades.py       6      0   100%

Read the diff

Coverage turned "I think my grade function is tested" into "lines 2 and 6 are NOT" — then guided you to exactly the two tests needed. That's the workflow: run coverage, read Missing, write a test for each gap, repeat. Next lesson plugs this into pytest directly with pytest-cov.

coverage run -m pytest + coverage report -m shows which lines your tests executed and which they missed. Branch coverage (--branch) is stricter, catching untested if/else paths. Use the Missing column to find gaps and write targeted tests. But remember: coverage finds untested code — it can't prove tested code is correct. Next: integrating this with pytest via pytest-cov.