Comprehensive Testing

Comprehensive Testing

Based on Anthropic's Claude Code Best Practices and community patterns

Core Philosophy

"Claude performs best when it has a clear target to iterate against—a test case provides concrete success criteria."

Testing is not about proving code works; it's about designing code that is testable and documenting expected behavior.

Test Pyramid

Copy code
        /\
       /  \        E2E Tests (10%)
      /----\       - Full user flows
     /      \      - Slowest, most brittle
    /--------\
   /          \    Integration Tests (20%)
  /------------\   - Component interaction
 /              \  - Real dependencies
/----------------\
       Unit Tests (70%)
       - Single function/method
       - Fast, isolated, many

TDD Workflow (Anthropic Recommended)

The 6-Step Process

Copy code
1. WRITE TESTS FIRST
   ↓
2. VERIFY TESTS FAIL
   ↓
3. COMMIT TEST SUITE
   ↓
4. IMPLEMENT CODE
   ↓
5. VERIFY WITH SUBAGENT
   ↓
6. COMMIT IMPLEMENTATION

Step 1: Write Tests First

markdownCopy code
Be EXPLICIT about TDD to avoid mock implementations:

"I want to implement [feature] using TDD.
First, write tests for [expected behavior] with these input/output pairs:
- Input: X → Expected: Y
- Input: A → Expected: B
Do NOT create any implementation yet."

Step 2: Verify Tests Fail

bashCopy code
# Run tests and confirm they fail for the RIGHT reason
npm test  # or pytest, go test, etc.

# Expected: "function not found" or "undefined"
# NOT: syntax error, wrong import

Step 3: Commit Test Suite

bashCopy code
git add tests/
git commit -m "test: Add tests for [feature] (RED phase)"

Step 4: Implement Incrementally

markdownCopy code
"Now implement the code to make these tests pass.
Do NOT modify the tests.
Run tests after each change until all pass."

Claude will enter an autonomous loop:

Copy code
Write code → Run tests → Analyze failures → Adjust → Repeat

Step 5: Verify with Subagent

markdownCopy code
"Use a subagent to independently verify the implementation:
- Is it overfitting to tests?
- Are edge cases handled?
- Is the code maintainable?"

Step 6: Commit Implementation

bashCopy code
git add src/
git commit -m "feat: Implement [feature] (GREEN phase)"

Test Structure Patterns

AAA Pattern (Arrange-Act-Assert)

typescriptCopy code
describe('UserService', () => {
  it('should create user with valid email', async () => {
    // Arrange - Setup test data and dependencies
    const userRepo = new InMemoryUserRepository();
    const service = new UserService(userRepo);
    const input = { email: 'test@example.com', name: 'Test' };

    // Act - Execute the code under test
    const user = await service.create(input);

    // Assert - Verify the results
    expect(user.email).toBe('test@example.com');
    expect(user.id).toBeDefined();
    expect(await userRepo.findById(user.id)).toEqual(user);
  });
});

Given-When-Then Pattern

pythonCopy code
def test_order_total_with_discount():
    """
    Given an order with items totaling $100
    When a 20% discount is applied
    Then the total should be $80
    """
    # Given
    order = Order()
    order.add_item(Item(price=50))
    order.add_item(Item(price=50))

    # When
    order.apply_discount(Percentage(20))

    # Then
    assert order.total == Money(80)

Framework Best Practices

Jest / Vitest (JavaScript/TypeScript)

typescriptCopy code
// Structure
describe('ModuleName', () => {
  describe('methodName', () => {
    it('should [expected behavior] when [condition]', () => {});
  });
});

// Setup/Teardown
beforeAll(async () => { /* one-time setup */ });
beforeEach(() => { /* per-test setup */ });
afterEach(() => { /* per-test cleanup */ });
afterAll(async () => { /* one-time cleanup */ });

// Async testing
it('handles async operations', async () => {
  const result = await asyncFunction();
  expect(result).toBe(expected);
});

// Error testing
it('throws on invalid input', () => {
  expect(() => validate(null)).toThrow('Input required');
});

// Snapshot testing (use sparingly)
it('renders correctly', () => {
  const tree = renderer.create(<Component />).toJSON();
  expect(tree).toMatchSnapshot();
});

// Table-driven tests
it.each([
  [1, 1, 2],
  [2, 2, 4],
  [0, 0, 0],
])('add(%i, %i) = %i', (a, b, expected) => {
  expect(add(a, b)).toBe(expected);
});

vitest.config.ts:

typescriptCopy code
export default defineConfig({
  test: {
    globals: true,
    environment: 'node',
    coverage: {
      provider: 'v8',
      reporter: ['text', 'json', 'html'],
      thresholds: {
        lines: 80,
        branches: 70,
        functions: 80,
      },
    },
  },
});

pytest (Python)

pythonCopy code
import pytest
from mymodule import Calculator

# Fixtures for dependency injection
@pytest.fixture
def calculator():
    return Calculator()

@pytest.fixture
def database():
    db = TestDatabase()
    yield db
    db.cleanup()

# Parametrized tests
@pytest.mark.parametrize("a,b,expected", [
    (1, 1, 2),
    (2, 2, 4),
    (0, 0, 0),
    (-1, 1, 0),
])
def test_add(calculator, a, b, expected):
    assert calculator.add(a, b) == expected

# Exception testing
def test_divide_by_zero(calculator):
    with pytest.raises(ZeroDivisionError):
        calculator.divide(1, 0)

# Async testing
@pytest.mark.asyncio
async def test_async_operation():
    result = await async_function()
    assert result == expected

# Markers for categorization
@pytest.mark.slow
@pytest.mark.integration
def test_database_connection(database):
    assert database.is_connected()

conftest.py:

pythonCopy code
import pytest

@pytest.fixture(scope="session")
def database_url():
    return "postgresql://test:test@localhost/test"

@pytest.fixture(autouse=True)
def reset_database(database):
    yield
    database.rollback()

pytest.ini:

iniCopy code
[pytest]
testpaths = tests
python_files = test_*.py
python_functions = test_*
addopts = -v --cov=src --cov-report=term-missing
markers =
    slow: marks tests as slow
    integration: marks tests as integration tests

Go Testing

goCopy code
package mypackage

import (
    "testing"
    "github.com/stretchr/testify/assert"
    "github.com/stretchr/testify/require"
)

func TestAdd(t *testing.T) {
    tests := []struct {
        name     string
        a, b     int
        expected int
    }{
        {"positive numbers", 1, 2, 3},
        {"zero values", 0, 0, 0},
        {"negative numbers", -1, -2, -3},
    }

    for _, tt := range tests {
        t.Run(tt.name, func(t *testing.T) {
            result := Add(tt.a, tt.b)
            assert.Equal(t, tt.expected, result)
        })
    }
}

// Table-driven with subtests
func TestUserService_Create(t *testing.T) {
    t.Run("creates user with valid input", func(t *testing.T) {
        repo := NewInMemoryRepo()
        svc := NewUserService(repo)

        user, err := svc.Create(CreateUserInput{Email: "test@example.com"})

        require.NoError(t, err)
        assert.NotEmpty(t, user.ID)
        assert.Equal(t, "test@example.com", user.Email)
    })

    t.Run("returns error for invalid email", func(t *testing.T) {
        repo := NewInMemoryRepo()
        svc := NewUserService(repo)

        _, err := svc.Create(CreateUserInput{Email: "invalid"})

        require.Error(t, err)
        assert.Contains(t, err.Error(), "invalid email")
    })
}

Mock Strategy

When to Mock

How to Mock

typescriptCopy code
// Jest - Mock module
jest.mock('./emailService', () => ({
  sendEmail: jest.fn().mockResolvedValue({ success: true }),
}));

// Jest - Mock function
const mockCallback = jest.fn();
mockCallback.mockReturnValue(42);

// Vitest - Spy
import { vi } from 'vitest';
const spy = vi.spyOn(console, 'log');

// Time mocking
beforeEach(() => {
  vi.useFakeTimers();
  vi.setSystemTime(new Date('2024-01-01'));
});

afterEach(() => {
  vi.useRealTimers();
});

pythonCopy code
# pytest - Mock with unittest.mock
from unittest.mock import Mock, patch, MagicMock

@patch('mymodule.external_api.fetch')
def test_with_mocked_api(mock_fetch):
    mock_fetch.return_value = {'data': 'mocked'}
    result = my_function()
    assert result == expected
    mock_fetch.assert_called_once_with('expected_arg')

# Fixture-based mock
@pytest.fixture
def mock_email_service():
    service = Mock()
    service.send.return_value = True
    return service

Prefer Test Doubles Over Mocks

typescriptCopy code
// ❌ Heavy mocking
const mockRepo = {
  findById: jest.fn().mockResolvedValue({ id: '1', name: 'Test' }),
  save: jest.fn().mockResolvedValue(undefined),
  delete: jest.fn().mockResolvedValue(undefined),
};

// ✅ In-memory implementation (test double)
class InMemoryUserRepository implements UserRepository {
  private users: Map<string, User> = new Map();

  async findById(id: string): Promise<User | null> {
    return this.users.get(id) || null;
  }

  async save(user: User): Promise<User> {
    this.users.set(user.id, user);
    return user;
  }

  async delete(id: string): Promise<void> {
    this.users.delete(id);
  }
}

Extended Reference

Detailed material starting at ## Boundary Testing has been moved to reference/extended.md to keep this skill concise. Load that reference when the task requires the moved examples, command catalogs, checklists, platform details, or implementation templates.