pytest-testing

Comprehensive guide for writing effective pytest tests in Python projects.

Quick Start

Installation

pip install pytest pytest-cov
# Or with uv
uv pip install pytest pytest-cov

Basic Test Structure

# tests/test_example.py
def test_addition():
    assert 1 + 1 == 2

def test_string_concatenation():
    assert "hello" + " " + "world" == "hello world"

Running Tests

pytest                    # Run all tests
pytest tests/             # Run tests in directory
pytest tests/test_models.py::TestClass::test_method  # Run specific test
pytest -v                # Verbose output
pytest --cov=package     # With coverage

Test Organization

File Structure

project/
├── tests/
│   ├── __init__.py
│   ├── conftest.py          # Shared fixtures
│   ├── test_models.py       # Model tests
│   ├── test_api.py          # API tests
│   └── test_cli.py          # CLI tests
└── pytest.ini              # Configuration

Naming Conventions

• Test files: test_*.py or *_test.py
• Test classes: Test*
• Test functions: test_*

Fixtures

Fixtures provide reusable test data and setup/teardown logic.

Basic Fixture

# conftest.py
import pytest

@pytest.fixture
def sample_data():
    """Create sample test data"""
    return {"key": "value"}

# test_file.py
def test_something(sample_data):
    assert sample_data["key"] == "value"

Fixture Scopes

• function (default): Each test function
• class: Each test class
• module: Each test module
• session: Entire test session

@pytest.fixture(scope="session")
def expensive_setup():
    """Run once per test session"""
    return expensive_operation()

Built-in Fixtures

• tmp_path: Temporary directory (Path object)
• capsys: Capture stdout/stderr
• monkeypatch: Temporarily modify environment
• request: Access test context

Fixture with Dependencies

@pytest.fixture
def sample_file(tmp_path):
    """Create temporary file"""
    file_path = tmp_path / "test.txt"
    file_path.write_text("test content")
    return str(file_path)

Assertions

Pytest uses Python's assert statement with detailed failure information.

def test_assertions():
    # Basic
    assert 1 + 1 == 2
    
    # Membership
    assert "test" in "this is a test"
    
    # Type checking
    assert isinstance([1, 2, 3], list)
    
    # Exception testing
    with pytest.raises(ValueError):
        int("not a number")
    
    # Exception message matching
    with pytest.raises(ValueError, match="invalid"):
        raise ValueError("invalid input")

Parametrized Tests

Run the same test with multiple inputs:

@pytest.mark.parametrize("input,expected", [
    ("hello", 5),
    ("world", 5),
    ("", 0),
])
def test_string_length(input, expected):
    assert len(input) == expected

Mocking and Patching

Use unittest.mock to isolate code under test:

from unittest.mock import patch, MagicMock

def test_with_mock(capsys):
    with patch('sys.argv', ['script', 'arg1']):
        main()
        captured = capsys.readouterr()
        assert "expected" in captured.out

def test_mock_object():
    mock_obj = MagicMock()
    mock_obj.method.return_value = "result"
    assert mock_obj.method() == "result"
    mock_obj.method.assert_called_once()

Test Coverage

Running Coverage

pytest --cov=package_name --cov-report=term-missing
pytest --cov=package_name --cov-report=html

Configuration (pytest.ini)

[pytest]
testpaths = tests
python_files = test_*.py
python_classes = Test*
python_functions = test_*
addopts = 
    -v
    --cov=package_name
    --cov-report=term-missing
    --cov-report=html

Best Practices

1. Test Independence

Each test should be independent and not rely on execution order:

# Good
def test_create():
    obj = create_object()
    assert obj is not None

# Bad - depends on global state
global_obj = None
def test_create():
    global global_obj
    global_obj = create_object()

2. Use Fixtures, Not Globals

# Good
@pytest.fixture
def sample_data():
    return {"key": "value"}

def test_something(sample_data):
    assert sample_data["key"] == "value"

3. Test Edge Cases

Test both happy paths and edge cases:

def test_normal_case():
    result = function("normal input")
    assert result == expected

def test_empty_input():
    result = function("")
    assert result == default

def test_none_input():
    with pytest.raises(ValueError):
        function(None)

4. Descriptive Test Names

Use clear, descriptive names:

# Good
def test_find_user_by_email_returns_user():
    pass

# Bad
def test1():
    pass

5. Test Documentation

Add docstrings to explain test purpose:

def test_find_by_name_fuzzy():
    """Test fuzzy name search finds partial matches"""
    # test code

Common Patterns

Testing CLI Commands

from unittest.mock import patch

def test_cli_command(capsys):
    with patch('sys.argv', ['script', 'command', 'arg']):
        main()
        captured = capsys.readouterr()
        assert "expected output" in captured.out

Testing Exceptions

def test_invalid_input():
    with pytest.raises(ValueError) as exc_info:
        function("invalid")
    assert "error message" in str(exc_info.value)

Testing File Operations

def test_file_operations(tmp_path):
    file_path = tmp_path / "test.txt"
    file_path.write_text("content")
    assert file_path.read_text() == "content"

Testing with Database

@pytest.fixture
def db_session():
    # Setup database
    session = create_session()
    yield session
    # Teardown
    session.close()

Advanced Features

Markers

@pytest.mark.slow
def test_expensive_operation():
    pass

# Run only fast tests
pytest -m "not slow"

Skipping Tests

@pytest.mark.skip(reason="Not implemented yet")
def test_future_feature():
    pass

@pytest.mark.skipif(sys.version_info < (3, 8), reason="Requires Python 3.8+")
def test_python38_feature():
    pass

Expected Failures

@pytest.mark.xfail(reason="Known bug, fixing")
def test_known_bug():
    assert False

Debugging Tests

Run with Output

pytest -s              # Show print statements
pytest -v              # Verbose output
pytest --pdb           # Drop into debugger on failure
pytest --tb=short      # Shorter traceback

Run Last Failed

pytest --lf            # Run last failed tests
pytest --ff            # Run failed first, then others

Configuration Examples

pytest.ini

[pytest]
testpaths = tests
python_files = test_*.py
python_classes = Test*
python_functions = test_*
addopts = 
    -v
    --strict-markers
    --tb=short
    --cov=package_name
    --cov-report=term-missing
    --cov-report=html

markers =
    slow: marks tests as slow
    integration: marks tests as integration tests
    unit: marks tests as unit tests

pyproject.toml

[tool.pytest.ini_options]
testpaths = ["tests"]
python_files = ["test_*.py"]
addopts = "-v --tb=short"

Common Issues and Solutions

Import Errors

Ensure test directory is in Python path or use -e install:

pip install -e .

Fixture Not Found

Check fixture is defined in conftest.py or same test file.

Coverage Not Working

Ensure pytest-cov is installed and package name is correct.

Tests Not Running

Check file naming (test_*.py) and function naming (test_*).

Integration with CI/CD

GitHub Actions Example

- name: Run tests
  run: |
    pip install pytest pytest-cov
    pytest --cov=package --cov-report=xml

References

• Pytest Documentation
• Pytest Best Practices
• See references/advanced-patterns.md for complex testing scenarios
• See references/coverage-improvement.md for coverage optimization strategies