Testing with pytest

Overview

pytest is the de facto Python testing framework, favored for its plain assert statements, rich failure introspection, and powerful fixture system. Test files named test_*.py and functions named test_* are collected automatically. Fixtures inject reusable setup, and parametrization runs one test across many inputs.

Syntax / Usage

Write a function starting with test_ and assert expected behavior. Run pytest from the project root.

# test_math.py

def add(a: int, b: int) -> int:
    return a + b

def test_add_positive():
    assert add(2, 3) == 5

def test_add_negative():
    assert add(-1, -1) == -2

# Assert that an exception is raised
import pytest

def test_divide_by_zero():
    with pytest.raises(ZeroDivisionError):
        _ = 1 / 0

pytest rewrites assert so a failure shows the actual and expected values without extra helpers.

Examples

Parametrize a test to cover many cases without duplication:

import pytest

def is_palindrome(s: str) -> bool:
    return s == s[::-1]

@pytest.mark.parametrize("value,expected", [
    ("racecar", True),
    ("hello", False),
    ("", True),
])
def test_is_palindrome(value, expected):
    assert is_palindrome(value) == expected

Share setup with a fixture; pytest passes it in by name:

import pytest

@pytest.fixture
def sample_users():
    return [{"name": "ada"}, {"name": "linus"}]

def test_user_count(sample_users):
    assert len(sample_users) == 2

def test_first_user(sample_users):
    assert sample_users[0]["name"] == "ada"

Common Mistakes

• Naming files or functions without the test_ prefix, so pytest never collects them
• Asserting pytest.raises without a with block, which never actually checks anything
• Sharing mutable state between tests instead of isolating with fixtures
• Overusing broad fixtures with scope="session" that leak state across tests
• Catching exceptions inside the test and asserting nothing, hiding real failures

See Also

python-type-hints python-exceptions python-functions