Block third-party trackers suite-wide (finish the #695 pilot) - #704
Open
hadobe wants to merge 1 commit into
Open
Block third-party trackers suite-wide (finish the #695 pilot)#704hadobe wants to merge 1 commit into
hadobe wants to merge 1 commit into
Conversation
Route the remaining 97 Nala block test files through the shared base test (nala/utils/test.cjs) so every suite gets the tracker-blocking page fixture, not just the 7 pilot blocks. Mechanical one-line import swap per file: require/import '@playwright/test' -> '../../utils/test.cjs'. All 104 test files now share the base. This removes the ~27 third-party ad/analytics tracker requests per page that were saturating the shared CI network and starving block decoration — the root cause of the "random block" first-run flakiness (pricing-cards, floating-button, etc.) that survived on blocks outside the pilot. Safe, verified: - Deny-list is trackers only; functional third-parties (embed's YouTube/ Vimeo, template APIs, IMS/SUSI) and Adobe infra/fonts are untouched. - instrument.js/DTM still load, so daa-* analytics attributes are still set — logo-row and quotes (which assert daa-*) pass unchanged. - Smoke-tested logo-row, quotes, embed, banner on chromium: 10/10 pass. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
hadobe
requested review from
echen-adobe,
maxn-adobe,
meganthecoder and
nateyolles
as code owners
August 21, 2026 01:11
|
Hello, I'm the AEM Code Sync Bot and I will run some actions to deploy your branch and validate page speed.
Commits
|
nateyolles
approved these changes
Aug 25, 2026
nateyolles
left a comment
nateyolles
left a comment
Contributor
There was a problem hiding this comment.
Looks like one test failed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Routes the remaining 97 Nala block test files through the shared base test (
nala/utils/test.cjs) so every suite gets the tracker-blockingpagefixture — not just the 7 pilot blocks from #695. All 104 test files now share the base. Mechanical one-line import swap per file (@playwright/test→../../utils/test.cjs); nothing else changed.Why
The #695 pilot proved that blocking the ~27 third-party ad/analytics tracker requests per page eliminates the network contention that starves block decoration on shared CI runners. The residual "random block" first-run flakiness (
pricing-cards,floating-button, …) was on blocks outside the pilot — they still loaded the full tracker stack. This finishes the rollout so they get the same relief.Safe — verified
embedblock's YouTube/Vimeo, template APIs, IMS/SUSI) and Adobe infra/fonts are untouched.instrument.js/DTM still load, sodaa-*attributes are still set —logo-rowandquotes(which assertdaa-*) pass unchanged.logo-row,quotes,embed,banner,template-x-promoon chromium: all pass.template-x-promo(anetworkidleuser) now settles in ~3.5s — with trackers gone,networkidlequiets down on its own.Not addressed here (separate follow-ups)
color-explore,color-extract) can still time out when they run concurrently on firefox/webkit — that's CPU/origin contention, not trackers (they already havemartech=off, so this fixture is a no-op for them). Pass solo and serial; fail clustered. Candidate fix: run the color suites at low concurrency. CIretries: 2currently absorbs them.networkidle→ deterministic waits: now lower priority since tracker-blocking makesnetworkidlesettle; do it only if something still hangs.🤖 Generated with Claude Code