py-testing-async

Python Async Testing

Problem Statement

Async testing requires specific patterns. pytest-asyncio has modes that affect behavior. Database tests need isolation. Mocking async functions differs from sync. Get these wrong and tests are flaky or don't catch bugs.

---

Pattern: pytest-asyncio Configuration

Problem: Pytest needs configuration for async tests.

# pyproject.toml
[tool.pytest.ini_options]
asyncio_mode = "strict"  # Requires explicit @pytest.mark.asyncio
# OR
asyncio_mode = "auto"    # All async tests run automatically

import pytest

# With asyncio_mode = "strict" (this codebase)
@pytest.mark.asyncio
async def test_something():
    result = await some_async_function()
    assert result == expected

# Without the marker = test won't run as async!

---

Pattern: Async Fixtures

Problem: Fixtures that provide async resources need specific handling.

import pytest
from sqlalchemy.ext.asyncio import AsyncSession

# ✅ CORRECT: Async fixture for session
@pytest.fixture
async def session() -> AsyncGenerator[AsyncSession, None]:
    async with async_session() as session:
        yield session
        await session.rollback()  # Clean up after test

# ✅ CORRECT: Async fixture for test data
@pytest.fixture
async def test_user(session: AsyncSession) -> User:
    user = User(email="test@example.com", hashed_password="...")
    session.add(user)
    await session.commit()
    await session.refresh(user)
    return user

# ✅ CORRECT: Using async fixtures
@pytest.mark.asyncio
async def test_get_user(session: AsyncSession, test_user: User):
    result = await session.execute(
        select(User).where(User.id == test_user.id)
    )
    user = result.scalar_one()
    assert user.email == "test@example.com"

---

Pattern: Database Test Isolation

Problem: Tests polluting each other's database state.