Harden front-end tracker E2E assertions against attribute-order changes

[Copilot]

• Identify failing test: front-end-tracker.spec.ts:98 — toContainText returns "" for <script> elements (Playwright limitation for non-visible elements)
• Fix: replace toContainText with evaluate() → el.textContent + plain expect().toContain() on line 98

[acicovic]

@coderabbitai review

Make sure that the tests have been modified in a way that still keeps them thorough and doesn't introduce any lacks compared to their previous state.

[coderabbitai]

Tip

For best results, initiate chat on the files or code changes.

@acicovic After thoroughly comparing the old and new assertions across both tests, the refactored tests maintain full coverage with no gaps. Here's the analysis:

---

Test 1 — "Should work as expected when only Site ID is provided"

Old assertion	New equivalent	Notes
content.toContain('<link rel="dns-prefetch" …>')	CSS selector link[rel="dns-prefetch"][href="…"] + toHaveCount(1)	Equivalent; DOM-based is more robust
Single string match for <script … id="parsely-cfg"> (attribute-order sensitive)	toHaveCount(1) + individual toHaveAttribute() calls for data-parsely-site and src	Strictly better — each attribute is independently verified, order-agnostic
Single string match for <script … id="wp-parsely-loader-js">	toHaveCount(1) + toHaveAttribute('src', …)	Equivalent
not.toContain("<script id='wp-parsely-loader-js-before'>")	script#wp-parsely-loader-js-before + toHaveCount(0)	Equivalent
not.toContain('window.wpParselySiteId =')	page.locator('script', { hasText: 'window.wpParselySiteId =' }) + toHaveCount(0)	Equivalent; scoping to <script> elements is appropriate since this string can only appear there

Test 2 — "Should work as expected when a Site ID and API Secret are provided"

All shared checks mirror Test 1 (same analysis above). The key change:

Old assertion	New equivalent	Notes
content.toContain('<script id="wp-parsely-loader-js-before">')	script#wp-parsely-loader-js-before + toHaveCount(1)	Equivalent
content.toContain(window.wpParselySiteId = '${VALID_SITE_ID}')	loaderInlineScript.evaluate(el => el.textContent) + expect(scriptContent).toContain(…)	Correct Playwright workaround for non-visible elements; same semantic coverage

---

Conclusion: No coverage gaps were introduced. The refactoring is sound — every original assertion has a direct, semantically equivalent replacement, and the attribute decomposition approach is actually an improvement over the brittle single-string HTML matches.