Portfolio Docs

Appearance

Testing ​

E2E Tests (Playwright) ​

End-to-end tests live in e2e/ and use Playwright with Chromium.

Configuration ​

typescript
// playwright.config.ts
{
  testDir: './e2e',
  fullyParallel: true,
  retries: process.env.CI ? 2 : 0,
  workers: process.env.CI ? 1 : undefined,
  baseURL: 'http://localhost:6173',
  webServer: {
    command: 'npm run dev',
    url: 'http://localhost:6173',
    reuseExistingServer: !process.env.CI,
  },
}

Running Tests ​

bash
# Run all E2E tests
npm test

# Interactive UI mode
npm run test:ui

Playwright automatically starts the dev server if it isn't running.

Test Files ​

Tests cover the chat widget, page navigation, and adversarial scenarios:

e2e/
├── chat-adversarial.spec.ts   # Security and edge case tests
├── chat.spec.ts               # Core chat functionality
├── ...

Chat Test Bypass ​

E2E tests bypass rate limiting via the X-Bypass-Rate-Limit header with the RATE_LIMIT_BYPASS_TOKEN value.

Worker Tests (Vitest) ​

Chat worker tests use Vitest with @cloudflare/vitest-pool-workers to run tests in a Workers-like environment.

Configuration ​

typescript
// chat-worker/vitest.config.ts
defineWorkersConfig({
  test: {
    poolOptions: {
      workers: {
        wrangler: { configPath: './wrangler.toml' },
        isolatedStorage: false,
        miniflare: {
          bindings: {
            SLACK_BOT_TOKEN: 'xoxb-test-token',
            SLACK_SIGNING_SECRET: 'test-signing-secret-for-ci',
            // ... other test bindings
          },
        },
      },
    },
  },
});

Running Worker Tests ​

bash
cd chat-worker
npm test

Test Files ​

chat-worker/test/
├── routes.test.ts         # HTTP endpoint tests
├── chat-session.test.ts   # Durable Object lifecycle tests
├── ai-engine.test.ts      # AI engine unit tests
├── slack.test.ts           # Slack API integration tests
├── env.d.ts               # Test environment types
└── tsconfig.json          # Test TypeScript config

Isolated Storage

Isolated storage is disabled (isolatedStorage: false) because Durable Objects with WebSockets and alarms perform async storage writes that outlive individual test boundaries.