Skip to content

Latest commit

 

History

History
53 lines (36 loc) · 2.06 KB

testing.md

File metadata and controls

53 lines (36 loc) · 2.06 KB

Playwright Testing

This short guide summarizes the current test runtime patterns used by the integration suite.

Route-Based Test Content (no php -S)

  • Use tests/Support/RouteServerTestTrait to fulfill requests directly via Playwright routing.
  • This removes the need to spawn a PHP built-in web server in tests.

Example

use Playwright\Tests\Support\RouteServerTestTrait;

// Inside a test case using PlaywrightTestCaseTrait
$this->installRouteServer($this->page, [
    '/index.html' => '<h1>Hello</h1>',
    '/script.js'  => 'window.ok = true;',
]);

$this->page->goto($this->routeUrl('/index.html'));

Notes

  • routeUrl('/path') returns a stable URL (e.g. http://localhost:31817/path).
  • Prefer page->route(...) inside the test if you need to override a response for a single test.

Optional Tracing

  • Tracing is disabled by default to keep the suite fast and light.
  • Enable per-run by setting PW_TRACE=1 (screenshots and snapshots enabled for each test).
    • macOS/Linux: PW_TRACE=1 composer test:integration
    • Windows (PowerShell): $env:PW_TRACE='1'; composer test:integration
  • On failure, artifacts are written to test-failures/.

Shared Lifecycle for Speed

  • The test trait reuses a single Playwright client and browser per test class by default.
  • Each test still uses a fresh BrowserContext + Page, which are closed in tearDown().
  • If a test needs a custom PlaywrightConfig (e.g., screenshotDir), pass it to setUpPlaywright(...) — in that case, the test runs with a dedicated client/browser (bypasses sharing) to respect the custom config.

Performance Tips

  • Keep PW_TRACE off unless diagnosing failures.
  • Prefer Playwright waits (e.g., waitForSelector, waitForResponse) over usleep().
  • When iterating locally, run integration tests without coverage for further speedups.

Expected Processes

  • A single Node.js Playwright server process (spawned by the transport).
  • A browser process (Chromium/Firefox/WebKit) per launched browser.
  • Minimal short‑lived processes in specific tests (e.g., ProcessLauncherTest).