# Test Data and Resetting State — Playwright & Cypress

Source: https://www.skillbyai.com/en/e2e-testing/r-state

> Every test owns its world.

## Isolation by design

Both tools isolate the browser by default: Playwright gives each test a new browser context, and Cypress (since version 12) enables `testIsolation`, clearing cookies and storage and resetting the page before each test. **Backend state** is your responsibility. Strategies: create the data a test needs through the API in a hook, use unique identifiers so parallel tests do not clash, reset or reseed a dedicated test database before a run, or wrap tests in tenants or accounts created per worker. Avoid relying on test order or on data created by another test. Control time with Playwright's clock API (`page.clock`, in recent versions) or Cypress `cy.clock()` and `cy.tick()`.

## Per-test data through the API

Playwright fixture that creates and deletes a project.

```typescript
import { test as base, expect } from '@playwright/test';

type Project = { id: string; name: string };

export const test = base.extend<{ project: Project }>({
  project: async ({ request }, use, testInfo) => {
    const name = `proj-${testInfo.workerIndex}-${Date.now()}`;
    const res = await request.post('/api/projects', { data: { name } });
    expect(res.ok()).toBeTruthy();
    const project: Project = await res.json();

    await use(project);

    await request.delete(`/api/projects/${project.id}`);
  },
});

test('renames a project', async ({ page, project }) => {
  await page.goto(`/projects/${project.id}`);
  await page.getByLabel('Name').fill(`${project.name}-renamed`);
  await page.getByRole('button', { name: 'Save' }).click();
  await expect(page.getByRole('heading')).toHaveText(`${project.name}-renamed`);
});
```

## Clean up before, not only after

If a run crashes, teardown may never happen. Resetting or seeding at the start of a run (or per test) keeps the next run predictable.

**Quiz:** Why add the worker index or a timestamp to created data names?

- [ ] Because the API rejects short names
- [ ] To make the HTML report sort alphabetically
- [x] So tests running in parallel do not collide on the same records
- [ ] To slow tests down deliberately

*Answer:* So tests running in parallel do not collide on the same records. Unique data enables parallel, order-independent tests.