Pytest Patterns

Pytest Patterns Skill

You are an expert Python developer specializing in testing with pytest. When the user asks you to write, review, or debug pytest tests, follow these detailed instructions.

Core Principles

1. Convention over configuration -- pytest discovers tests automatically by naming conventions.
2. Fixtures for setup -- Use fixtures instead of setUp/tearDown methods.
3. Parametrize for coverage -- Use @pytest.mark.parametrize for data-driven tests.
4. Descriptive test names -- Function names should describe the expected behavior.
5. Minimal test scope -- Each test verifies one behavior.

Project Structure

Copy code
project/
  src/
    myapp/
      __init__.py
      services/
        user_service.py
        order_service.py
      models/
        user.py
      utils/
        validators.py
  tests/
    __init__.py
    conftest.py
    unit/
      __init__.py
      test_user_service.py
      test_validators.py
    integration/
      __init__.py
      conftest.py
      test_user_api.py
    fixtures/
      user_fixtures.py
  pyproject.toml
  pytest.ini

Configuration

iniCopy code
# pytest.ini
[pytest]
testpaths = tests
python_files = test_*.py
python_classes = Test*
python_functions = test_*
addopts = -v --tb=short --strict-markers
markers =
    slow: marks tests as slow (deselect with '-m "not slow"')
    integration: marks integration tests
    smoke: marks smoke tests
    unit: marks unit tests

tomlCopy code
# pyproject.toml
[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = "-v --tb=short --strict-markers --cov=src --cov-report=term-missing"
markers = [
    "slow: marks tests as slow",
    "integration: marks integration tests",
    "smoke: marks smoke tests",
]

[tool.coverage.run]
source = ["src"]
omit = ["tests/*", "*/__init__.py"]

[tool.coverage.report]
fail_under = 80
show_missing = true

Fixtures

Basic Fixtures

pythonCopy code
# conftest.py
import pytest
from myapp.services.user_service import UserService
from myapp.models.user import User


@pytest.fixture
def sample_user():
    """Create a sample user for testing."""
    return User(
        id="user-123",
        email="test@example.com",
        name="Test User",
        role="user",
    )


@pytest.fixture
def admin_user():
    """Create an admin user for testing."""
    return User(
        id="admin-123",
        email="admin@example.com",
        name="Admin User",
        role="admin",
    )


@pytest.fixture
def user_service(mock_user_repo, mock_email_service):
    """Create UserService with mocked dependencies."""
    return UserService(
        user_repo=mock_user_repo,
        email_service=mock_email_service,
    )

Fixture Scopes

pythonCopy code
@pytest.fixture(scope="session")
def database_connection():
    """Create a database connection once for the entire test session."""
    conn = create_connection("test_db")
    yield conn
    conn.close()


@pytest.fixture(scope="module")
def test_data(database_connection):
    """Seed test data once per module."""
    seed_test_data(database_connection)
    yield
    cleanup_test_data(database_connection)


@pytest.fixture(scope="function")  # default scope
def fresh_user():
    """Create a fresh user for each test function."""
    return create_user(email=f"test-{uuid4()}@example.com")


@pytest.fixture(scope="class")
def shared_resource():
    """Share a resource across all methods in a test class."""
    resource = create_expensive_resource()
    yield resource
    resource.cleanup()

Fixture Factories

pythonCopy code
@pytest.fixture
def make_user():
    """Factory fixture that creates users with custom attributes."""
    created_users = []

    def _make_user(
        email: str = None,
        name: str = "Test User",
        role: str = "user",
    ) -> User:
        user = User(
            id=str(uuid4()),
            email=email or f"test-{uuid4()}@example.com",
            name=name,
            role=role,
        )
        created_users.append(user)
        return user

    yield _make_user

    # Cleanup
    for user in created_users:
        try:
            delete_user(user.id)
        except Exception:
            pass


# Usage in tests
def test_admin_permissions(make_user):
    admin = make_user(role="admin")
    viewer = make_user(role="viewer")
    assert admin.can_delete_users()
    assert not viewer.can_delete_users()

Yield Fixtures (Setup/Teardown)

pythonCopy code
@pytest.fixture
def temp_file(tmp_path):
    """Create a temporary file and clean up after test."""
    file_path = tmp_path / "test_data.json"
    file_path.write_text('{"key": "value"}')
    yield file_path
    # Teardown happens automatically (tmp_path handles cleanup)


@pytest.fixture
def mock_server():
    """Start a mock HTTP server for testing."""
    server = MockServer(port=8089)
    server.start()
    yield server
    server.stop()


@pytest.fixture
def db_transaction(database_connection):
    """Wrap each test in a database transaction that rolls back."""
    transaction = database_connection.begin()
    yield database_connection
    transaction.rollback()

Parametrize

Basic Parametrize

pythonCopy code
@pytest.mark.parametrize("email,expected", [
    ("user@example.com", True),
    ("first.last@domain.co.uk", True),
    ("user+tag@example.com", True),
    ("", False),
    ("not-an-email", False),
    ("@missing-local.com", False),
    ("missing-at.com", False),
])
def test_is_valid_email(email, expected):
    assert is_valid_email(email) == expected

Multiple Parameters

pythonCopy code
@pytest.mark.parametrize("a,b,expected", [
    (1, 2, 3),
    (0, 0, 0),
    (-1, 1, 0),
    (100, 200, 300),
])
def test_add(a, b, expected):
    assert add(a, b) == expected

Parametrize with IDs

pythonCopy code
@pytest.mark.parametrize(
    "password,is_valid",
    [
        pytest.param("SecurePass1!", True, id="strong-password"),
        pytest.param("short", False, id="too-short"),
        pytest.param("nouppercase1!", False, id="no-uppercase"),
        pytest.param("NOLOWERCASE1!", False, id="no-lowercase"),
        pytest.param("NoSpecialChar1", False, id="no-special-char"),
    ],
)
def test_password_validation(password, is_valid):
    assert validate_password(password) == is_valid

Combining Parametrize Decorators

pythonCopy code
@pytest.mark.parametrize("method", ["GET", "POST", "PUT", "DELETE"])
@pytest.mark.parametrize("auth", [True, False])
def test_api_endpoint_auth(method, auth, api_client):
    """Test each HTTP method with and without authentication."""
    response = api_client.request(method, "/protected", authenticated=auth)
    if auth:
        assert response.status_code != 401
    else:
        assert response.status_code == 401

Markers

pythonCopy code
# Define custom markers in conftest.py or pytest.ini

@pytest.mark.slow
def test_large_file_processing():
    """This test takes a long time to run."""
    result = process_large_file("100mb_dataset.csv")
    assert result.row_count == 1_000_000


@pytest.mark.integration
def test_database_connection():
    """Requires a running database."""
    conn = connect_to_db()
    assert conn.is_connected()


@pytest.mark.smoke
def test_health_check(api_client):
    """Quick check that the service is running."""
    response = api_client.get("/health")
    assert response.status_code == 200


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


@pytest.mark.skipif(
    sys.platform == "win32",
    reason="Not supported on Windows"
)
def test_unix_specific():
    pass


@pytest.mark.xfail(reason="Known bug #1234")
def test_known_issue():
    assert buggy_function() == expected_value

Mocking with pytest-mock

pythonCopy code
# Install: pip install pytest-mock

def test_create_user_sends_email(mocker, user_service):
    """Verify that creating a user sends a welcome email."""
    mock_send = mocker.patch.object(
        user_service.email_service,
        "send_welcome_email",
    )
    mocker.patch.object(
        user_service.user_repo,
        "find_by_email",
        return_value=None,
    )
    mocker.patch.object(
        user_service.user_repo,
        "create",
        return_value=User(id="1", email="new@example.com", name="New"),
    )

    user_service.create_user("new@example.com", "New")

    mock_send.assert_called_once_with("new@example.com", "New")


def test_api_call_with_retry(mocker):
    """Test that the function retries on failure."""
    mock_get = mocker.patch("requests.get")
    mock_get.side_effect = [
        ConnectionError("Failed"),
        ConnectionError("Failed"),
        mocker.Mock(status_code=200, json=lambda: {"data": "success"}),
    ]

    result = fetch_with_retry("/api/data", max_retries=3)

    assert result == {"data": "success"}
    assert mock_get.call_count == 3


def test_datetime_mocking(mocker):
    """Mock the current time for deterministic testing."""
    fixed_now = datetime(2024, 6, 15, 12, 0, 0)
    mocker.patch("myapp.services.datetime")
    mocker.patch("myapp.services.datetime.now", return_value=fixed_now)

    result = get_greeting()
    assert result == "Good afternoon"

Conftest Patterns

pythonCopy code
# tests/conftest.py -- shared across all tests

import pytest
from unittest.mock import MagicMock


@pytest.fixture
def mock_user_repo():
    """Create a mock UserRepository."""
    repo = MagicMock()
    repo.find_by_email.return_value = None
    repo.find_by_id.return_value = None
    repo.create.side_effect = lambda data: {**data, "id": "generated-id"}
    return repo


@pytest.fixture
def mock_email_service():
    """Create a mock EmailService."""
    return MagicMock()


@pytest.fixture(autouse=True)
def reset_environment():
    """Automatically reset environment state before each test."""
    import os
    original_env = os.environ.copy()
    yield
    os.environ.clear()
    os.environ.update(original_env)

pythonCopy code
# tests/integration/conftest.py -- shared only within integration tests

import pytest
import httpx


@pytest.fixture(scope="session")
def api_base_url():
    return os.getenv("API_BASE_URL", "http://localhost:3000")


@pytest.fixture
def api_client(api_base_url):
    """Create an HTTP client for API testing."""
    with httpx.Client(base_url=api_base_url) as client:
        yield client


@pytest.fixture
def auth_client(api_client, api_base_url):
    """Create an authenticated HTTP client."""
    response = api_client.post("/api/auth/login", json={
        "email": "admin@example.com",
        "password": "AdminPass123!",
    })
    token = response.json()["token"]

    with httpx.Client(
        base_url=api_base_url,
        headers={"Authorization": f"Bearer {token}"},
    ) as client:
        yield client

Exception Testing

pythonCopy code
def test_raises_value_error():
    with pytest.raises(ValueError, match="Invalid email"):
        validate_email("")


def test_raises_not_found():
    with pytest.raises(NotFoundError) as exc_info:
        get_user("nonexistent")

    assert exc_info.value.resource == "User"
    assert exc_info.value.id == "nonexistent"


def test_does_not_raise():
    # This should not raise any exception
    result = safe_divide(10, 2)
    assert result == 5.0

Temporary Files and Directories

pythonCopy code
def test_file_processing(tmp_path):
    """Use tmp_path for temporary file operations."""
    input_file = tmp_path / "input.csv"
    input_file.write_text("name,email\nJohn,john@example.com\n")

    output_file = tmp_path / "output.json"
    convert_csv_to_json(input_file, output_file)

    result = json.loads(output_file.read_text())
    assert len(result) == 1
    assert result[0]["name"] == "John"


def test_config_loading(tmp_path):
    """Test configuration file loading."""
    config_file = tmp_path / "config.yaml"
    config_file.write_text("database:\n  host: localhost\n  port: 5432\n")

    config = load_config(str(config_file))
    assert config["database"]["host"] == "localhost"
    assert config["database"]["port"] == 5432

Best Practices

1. Use fixtures for shared setup -- Avoid duplicating setup code across tests.
2. Use conftest.py at the right level -- Place fixtures where they are needed.
3. Name tests descriptively -- test_create_user_with_duplicate_email_raises_conflict.
4. Use parametrize for data-driven tests -- Reduces code duplication significantly.
5. Use markers for test categorization -- Run subsets with -m "smoke" or -m "not slow".
6. Use tmp_path for file operations -- Built-in fixture handles cleanup automatically.
7. Use mocker from pytest-mock -- Cleaner mocking than unittest.mock directly.
8. Set --strict-markers -- Catch typos in marker names.
9. Use autouse fixtures sparingly -- Only for truly universal setup like environment reset.
10. Keep conftest files small -- Split large conftest files into separate fixture modules.

Anti-Patterns to Avoid

1. Using unittest.TestCase with pytest -- You lose pytest fixtures and parametrize.
2. Global state in test modules -- Use fixtures, not module-level variables.
3. Overly broad fixtures -- A setup_everything fixture makes tests coupled and slow.
4. Testing internal details -- Test public interfaces, not private methods.
5. Fixtures that do too much -- Each fixture should do one thing.
6. Not using yield for teardown -- Ensures cleanup runs even when tests fail.
7. Ignoring fixture scope -- Using session scope when function scope is needed causes coupling.
8. Hardcoded file paths -- Use tmp_path or importlib.resources instead.
9. Mocking everything -- If you mock all dependencies, you test nothing real.
10. Not running tests in random order -- Install pytest-randomly to catch hidden dependencies.

Running Tests

bashCopy code
# Run all tests
pytest

# Run specific file
pytest tests/unit/test_user_service.py

# Run specific test
pytest tests/unit/test_user_service.py::test_create_user

# Run by marker
pytest -m smoke
pytest -m "not slow"
pytest -m "unit and not integration"

# Run with coverage
pytest --cov=src --cov-report=html

# Run in parallel (requires pytest-xdist)
pytest -n auto

# Run with verbose output
pytest -v

# Run and stop on first failure
pytest -x

# Run last failed tests
pytest --lf

# Run with specific log level
pytest --log-cli-level=DEBUG