Strategie testów end-to-end

Welcome to End-to-End (E2E) Testing! This strategy simulates a real user interacting with your application from start to finish. It covers the entire stack: your frontend, backend, database, and network.

E2E tests ensure that all components work together seamlessly, just as they would in a live environment. Think of it as a robot user clicking, typing, and verifying your app's behavior.

While tRPC provides excellent end-to-end type safety, E2E tests serve a different, crucial purpose. Type safety helps prevent bugs at compile time, but E2E tests catch issues that might arise at runtime.

These include deployment problems, configuration errors, network issues, and integration bugs that static analysis simply can't detect. E2E testing validates the actual user experience across your entire tRPC-powered application.

To write E2E tests, you'll need a testing framework. Popular choices include Playwright and Cypress. Both allow you to automate browser interactions and assert application states.

• Playwright: Known for its multi-browser support (Chromium, Firefox, WebKit), strong parallelism, and robust auto-waiting capabilities.
• Cypress: Offers an excellent developer experience with real-time reloads and powerful debugging tools directly in the browser.

For our examples, we'll focus on Playwright due to its versatility.

Setting up Playwright is straightforward. You typically initialize it in your project and it handles browser installations. It then creates a configuration file and an example test.

To begin, open your terminal in your project's root and run:

npm init playwright@latest
# Follow prompts: choose TypeScript, add a test file

This command installs Playwright, browser binaries, and sets up a playwright.config.ts file for your test settings.

Playwright allows you to simulate user actions like clicking buttons, typing into fields, and navigating pages. You use 'locators' to find elements on the page.

Here's a snippet showing how to navigate and fill out a login form:

// tests/login.spec.ts
import { test, expect } from '@playwright/test';

test('should navigate to login and fill form', async ({ page }) => {
  await page.goto('http://localhost:3000/login');
  await page.locator('input[type="email"]').fill('test@example.com');
  await page.locator('input[type="password"]').fill('password123');
  await page.locator('button', { hasText: 'Login' }).click();
  // Further assertions would go here
});

It's important to remember that E2E tests do not directly call your tRPC backend procedures. Instead, they interact with your frontend user interface (UI).

When your UI triggers a tRPC query or mutation (e.g., submitting a form), the E2E test observes the resulting changes on the page. You verify that the UI updates correctly, indicating that the underlying tRPC call and backend logic worked as expected.

Let's walk through an example: testing the creation of a new post in your application. The E2E test will simulate a user navigating to the 'create post' page, filling out a form, submitting it, and then verifying that the new post appears on a list.

// tests/post-creation.spec.ts
import { test, expect } from '@playwright/test';

test('should allow a user to create a new post', async ({ page }) => {
  await page.goto('http://localhost:3000/posts/new');
  await page.locator('input[placeholder="Title"]').fill('My New E2E Post');
  await page.locator('textarea[placeholder="Content"]').fill('This is content from an E2E test.');
  await page.locator('button', { hasText: 'Create Post' }).click();

  // Assert redirection and new post visibility
  await expect(page).toHaveURL(/posts$/); // Should redirect to /posts
  await expect(page.locator('h2', { hasText: 'My New E2E Post' })).toBeVisible();
});

For reliable E2E tests, you need a predictable starting state. This means managing your test data carefully. You'll often need to 'seed' your database with specific data before a test runs and 'clean up' that data afterwards.

Strategies include: calling special API endpoints (e.g., /api/test/reset-db) before each test, or directly interacting with a test database instance. This ensures each test starts with a clean slate and is isolated from others.

After performing actions, E2E tests use assertions to check if the application is in the expected state. Playwright provides powerful assertion methods with automatic waiting.

This 'auto-waiting' feature is crucial for testing asynchronous UI updates, ensuring that elements are visible or actionable before an assertion or action is attempted.

// Example assertions after actions
await expect(page.locator('.success-message')).toBeVisible();
await expect(page.locator('#item-count')).toHaveText('10');
await expect(page).toHaveURL(/dashboard/); // Checks the current URL
await expect(page.locator('button', { hasText: 'Delete' })).toBeDisabled();

We've explored End-to-End testing, understanding its role in validating the complete user journey in tRPC applications. By simulating real user interactions with tools like Playwright, you can ensure your frontend and backend seamlessly communicate and deliver the expected experience.

E2E tests are your final line of defense, catching integration issues and ensuring your application works as intended from the user's perspective. Keep practicing these strategies to build robust and reliable tRPC apps!