Amazon Playwright Testing Interview Questions & Answers

A browser is the engine instance; a context is an isolated session (cookies, storage); a page is a tab inside a context.

Workflow: checkout → setup-node → npm ci → npx playwright install --with-deps → npx playwright test → upload playwright-report as an artifact.

Before acting, Playwright waits for the element to be attached, visible, stable, enabled and able to receive events — so manual waits are rarely needed.

The action runs without waiting, so the next step may run first — causing flaky failures or 'Target closed' errors. The @typescript-eslint/no-floating-promises rule catches it.

Syntax over Promises: an async function returns a Promise and await pauses until it resolves. Every Playwright action returns a Promise, so you await each step.

Use page.route to intercept the call and fulfil it with fixed JSON so tests do not depend on external systems.

Destructuring pulls values from arrays or objects into variables; spread expands an array or object into another, useful for copying and merging config.

POST the query and variables as JSON to the single endpoint and assert both data and errors fields in the response.

Placeholders for types that make reusable code type-safe, for example a helper function returning the same type of item that it receives.

Run in parallel with sharding, reuse authentication state, set up data through APIs, avoid unnecessary waits, run smoke on every commit and the full suite nightly.

Use device descriptors such as devices['iPhone 13'] in a project config to emulate viewport, user agent and touch.

Use request.post with multipart: { file: { name, mimeType, buffer } } and assert the response and stored file details.

Record timestamps around the request or use response timing, then assert the duration is under an agreed threshold.

Register page.on('dialog', d => d.accept()) before the action triggers the dialog.

List, dot, line, HTML, JSON and JUnit reporters, plus integrations such as Allure, configurable per environment.

Integrate axe-core through @axe-core/playwright, run an analysis on the page and assert that there are no violations.

undefined means a variable has not been assigned; null is an intentional empty value.

Scripts that run once before or after the whole run, for seeding data, starting services or creating authentication state.

Assert the status with expect(res.status()).toBe(200), parse res.json() and check fields, using schema validation for the structure.

test.step groups actions into named steps in reports; annotations such as test.skip, test.fixme and test.slow communicate intent and adjust behaviour.

Fixtures set up and tear down test dependencies. Extend with test.extend({ loggedInPage: async ({ page }, use) => { ...; await use(page); } }).

Set extraHTTPHeaders in the config or pass headers per request, for example Authorization: Bearer token fetched in a setup step.

User-facing ones: getByRole, getByText, getByLabel, getByPlaceholder and getByTestId, which are more resilient than CSS or XPath.

Use try-catch with await, or .catch() on promises, and make sure rejected promises are never left unhandled.

Use --debug or the Inspector, page.pause(), UI mode, the trace viewer and headed runs with slowMo.

const [download] = await Promise.all([ page.waitForEvent('download'), page.getByText('Export').click(), ]); await download.saveAs('reports/file.csv');

Set them in the context options (geolocation, permissions, timezoneId, locale) to simulate users from other regions.

Use expect(page).toHaveScreenshot() to compare with stored baselines, with thresholds and masks for dynamic regions.

A Locator is lazy, re-resolved on every action and auto-waits (recommended). An ElementHandle points to one DOM node and can go stale.

Rely on auto-waiting plus a web-first assertion on the final state, disable animations in config for tests and avoid fixed timeouts.

Fetch tokens in a fixture with expiry tracking and refresh it when it is about to expire, keeping tests unaware of the details.

map transforms each item, filter keeps items that match a condition, and reduce combines all items into a single value.

Set retries in the config or CLI and failed tests re-run in a fresh context, with trace recorded on first retry to diagnose flakiness.

Log in with request.post, call request.storageState(), and create a browser context using that state.

beforeEach runs the same code before every test in scope; fixtures are lazy, composable and only set up what a test uses.

Playwright supports multiple browsers, tabs, languages and has auto-waiting and tracing; Selenium has the widest ecosystem; Cypress runs in-browser with a great UI but has more limits across tabs.

Use the request fixture: const res = await request.get('/api/users'); expect(res.ok()).toBeTruthy(); const body = await res.json();

Define projects for chromium, firefox and webkit in playwright.config.ts, set workers, then run: npx playwright test

await expect(page).toHaveScreenshot('home.png') compares against a baseline image; tune maxDiffPixels and mask dynamic areas like dates.

TypeScript + POM + custom fixtures, environment configs, API helpers for test data, tags like @smoke, sharding across CI machines, retries with traces, HTML/Allure reports, lint and review rules.

Separate tests, page objects, fixtures, utilities, test data and config; use tags, projects, shared auth setup, linting, code review and a CI pipeline with sharding and reports.

Use page.frameLocator('#frame').getByRole(...) so actions work inside the iframe with auto-waiting.

Create a class that takes a page, defines locators as properties and exposes action methods; instantiate it in a fixture for clean tests.

Send invalid payloads or missing auth and assert 400, 401 or 404 codes and the error message bodies.

test('login', async ({ page }) => { await page.goto('/login'); await page.getByLabel('Email').fill('a@b.com'); await page.getByLabel('Password').fill('secret'); await page.getByRole('button', { name: 'Login' }).click(); await expect(page.getByRole('heading', { name: 'Dashboard' })).toBeVisible(); });

Turn on trace: 'on-first-retry', open the trace viewer, check network and timing, replace hard waits with web-first assertions, isolate test data and check for sharding conflicts.

Cover happy path, validation errors, auth and permissions, boundary values, idempotency, schema, performance and data cleanup.

Auto-waiting, a built-in test runner, Chromium/Firefox/WebKit with one API, network interception, the trace viewer and parallel runs out of the box — usually faster and less flaky.

API testing checks services directly and is fast and stable; UI testing checks the full user experience in the browser and is slower.

Files run in parallel by default; set workers in the config and use test.describe.configure({ mode: 'parallel' }) for tests inside a file.

Use the official Playwright image which contains browsers and dependencies, mount the project and run npx playwright test for consistent CI environments.

Declarations of var and functions are moved to the top of their scope at compile time; let and const are hoisted but stay in the temporal dead zone until declared.

Log in once in a setup project, save storageState to a file and reuse it in other projects or tests.

Use npx playwright test --shard=1/4 in each CI job and merge reports afterwards to reduce total run time.

In a fixture or beforeEach, call the API (request.post) to create the user or order, use it in the UI test, then delete it in teardown. It is faster and more reliable than creating data through the UI.

click simulates a real user click with actionability checks; dispatchEvent fires a DOM event directly without those checks.

Strings in backticks that allow embedded expressions like `Hello ${name}` and multi-line text, useful for building dynamic locators and URLs.

Both describe the shape of data. Interfaces are extendable and suit object contracts; type aliases can also model unions and primitives.

page.route('**/api/items', route => route.fulfill({ json: [...] })) intercepts and replies with test data; route.abort() simulates failures.

Capture the download event, save it with download.saveAs, then read the file in Node and assert on name, size or content.

Add tags in the title (@smoke) or use the tag option, and run with --grep @smoke or --grep-invert.

Functions that take other functions as arguments or return them, such as map, filter and reduce, or a helper that wraps retry logic around a test step.