[chrome-devtools-patches] Update Chrome DevTools to M150 (Chrome 150)#14573
[chrome-devtools-patches] Update Chrome DevTools to M150 (Chrome 150)#14573petebacondarwin wants to merge 1 commit into
Conversation
🦋 Changeset detectedLatest commit: 1fe8421 The changes in this PR will be included in the next version bump. This PR includes changesets to release 1 package
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
|
UnknownError: ProviderInitError |
|
@petebacondarwin Bonk workflow failed. Check the logs for details. View workflow run · To retry, trigger Bonk again. |
|
✅ All changesets look good |
@cloudflare/autoconfig
create-cloudflare
@cloudflare/deploy-helpers
@cloudflare/kv-asset-handler
miniflare
@cloudflare/pages-shared
@cloudflare/unenv-preset
@cloudflare/vite-plugin
@cloudflare/vitest-pool-workers
@cloudflare/workers-auth
@cloudflare/workers-editor-shared
@cloudflare/workers-utils
wrangler
commit: |
| @@ -60,26 +60,26 @@ index 893ac3533f..937e646ab5 100644 | |||
| import '../../panels/network/network-meta.js'; | |||
There was a problem hiding this comment.
🔍 Duplicate network-meta.js import will be present after patch application
Patch 0006 adds import '../../panels/network/network-meta.js'; at line 58, but the context lines at line 60 show this import already exists in the upstream file. After applying the patch, js_app.ts will contain the same side-effect import twice. This is harmless at runtime (JS modules are singletons and only execute once), but it's unnecessary code that suggests the patch was not updated when upstream added this import. This is pre-existing (identical in the old patch version), but since this PR rebases all patches, it's a good opportunity to remove the duplicate line from the patch.
(Refers to lines 58-60)
Was this helpful? React with 👍 or 👎 to provide feedback.
| @@ -74,8 +61,12 @@ index 64d55d4c14..07972be81f 100644 | |||
| + display: 'none', | |||
There was a problem hiding this comment.
🔍 CSS rule uses invalid quoted value and comma instead of semicolon
The CSS rule at line 61 reads display: 'none', — this has two problems: (1) CSS property values should not be quoted with single quotes ('none' is not a valid CSS value; it should be none), and (2) the declaration ends with a comma instead of a semicolon. As a result, this rule will be silently ignored by browsers, meaning duplicate .cm-editor elements won't be hidden as intended. This is pre-existing (unchanged from the old patch), but since the patches are being rebased, it's worth fixing.
Was this helpful? React with 👍 or 👎 to provide feedback.
569912c to
0a34f44
Compare
|
The Wrangler DevTools preview is now live. You can access it directly at: https://1cc95e87-cloudflare-devtools.devprod.workers.dev/js_app In order to test the DevTools preview in
|
0a34f44 to
1fe8421
Compare
| + Root.Runtime.ExperimentName.AUTHORED_DEPLOYED_GROUPING, | ||
| + Root.Runtime.ExperimentName.JUST_MY_CODE, |
There was a problem hiding this comment.
No longer needed?
Periodic rebase of our maintained Workers-specific Chrome DevTools patches onto a newer upstream release, aligned with Chrome 150 stable (
devtools-frontendcommit1d67dc0dafa344bbd6ca75c124e2d6d9d53074d8, up fromb66ecf5b).This is the DevTools frontend used by
wrangler devand the Workers Playground (seepackages/chrome-devtools-patches). All 8 maintained patches were rebased and regenerated; applying the regenerated patch set withgit amonto the new base reproduces the rebased tree exactly.Notable adaptations required by upstream churn:
SessionRouter/CDPConnectionObserverobserver model; the ping now uses the typedconnection.send("Runtime.getIsolateId", ...)API.JUST_MY_CODEandAUTHORED_DEPLOYED_GROUPINGexperiments graduated upstream into thenavigator-just-my-codeandnavigator-group-by-authoredsettings. We now default those settings to on (insources-meta.ts) instead of force-enabling experiments (inMainImpl.ts), preserving the previous behaviour.ProfilesPaneldropped itsprofileTypesconstructor parameter, so we override the new virtualget profileTypes()getter to expose only the heap-snapshot type.script-src.Verification status (this is a draft):
git amof the regenerated patches applies cleanly onto the new base and reproduces the rebased tree exactly.make devtools-frontend/out/Default/gen/front_end).preview:chrome-devtools-patches) and manual smoke test (dark-mode console contrast,?theme=param, "Cloudflare" sources navigator, network tab with disable-cache/throttling hidden, heap profiling limited to the snapshot type).git amreproducing the patched tree exactly, a local DevTools build, the CI preview build, and a manual smoke test of the patched behaviours (theme query param, "Cloudflare" sources navigator, network-tab customisations, and heap profiling limited to the snapshot type).A picture of a cute animal (not mandatory, but encouraged)