Testing Guide Skill

This skill guides you through the testing process in PowerPlay. While there is no strict "Test Driven Development" mandate, writing tests for complex logic (especially parsers and utilities) is highly recommended.

1. 🧪 Test Stack

• •Framework: Vitest (Compatible with Jest API)
• •Environment: jsdom (for React components), node (for logic)
• •Testing Library: @testing-library/react
• •Location: tests/ directory (or co-located __tests__ folders)

2. 🏃 Running Tests

Since there is no default test script in package.json, use:

npx vitest

Or for a specific file:

npx vitest tests/my-test.test.ts

3. 📂 Structure & Strategy

A. Unit Tests (Recommended)

• •Target: Utility functions, Parsers, Helpers.
• •Path: tests/unit/ or next to the file e.g., src/utils/__tests__/parser.test.ts
• •Example: Testing parseMatchText function.

import { describe, it, expect } from 'vitest';
import { parseMatchText } from '@/utils/parser';

describe('parseMatchText', () => {
  it('should extract date correctly', () => {
    const result = parseMatchText('10/10 22:00 ...');
    expect(result.date).toBe('2024-10-10');
  });
});

B. Server Action Tests (Advanced)

• •Target: src/app/actions/*.ts
• •Challenge: Requires mocking Supabase.
• •Strategy:
  1. •Mock createClient from @/lib/supabase/server.
  2. •Use vi.mock to intercept DB calls.

import { vi, describe, it, expect } from 'vitest';
import { getMatches } from '@/app/actions/match';

// Mock Supabase
vi.mock('@/lib/supabase/server', () => ({
  createClient: vi.fn(() => ({
    from: () => ({
      select: () => ({ data: [], error: null })
    })
  }))
}));

4. 🤖 Dealing with AI Non-Determinism in Tests

When testing AI-generated content or parsers:

• •Loose Assertions: Use expect.stringContaining(), expect.any(String).
• •avoid: Strict equality check on long generated strings.

5. ⚠️ Key Rules

1. •Don't test the framework: Don't test if Next.js routiing works. Test YOUR logic.
2. •Mock External APIs: Never call real Naver Maps or OpenAI API in tests.
3. •Clean up: Where possible, use beforeEach / afterEach to reset mocks.