Bug 290876 – New Test(292363@main): [ iOS ] imported/w3c/web-platform-tests/trusted-types/DedicatedWorker-constructor-from-SharedWorker.html is a flaky failure.

• Component: New Bugs
• Status: RESOLVED DUPLICATE

[Yasmin Karimi]

imported/w3c/web-platform-tests/trusted-types/DedicatedWorker-constructor-from-SharedWorker.html

This test is a flaky failure on iOS.

HISTORY:

https://results.webkit.org/?suite=layout-tests&test=imported%2Fw3c%2Fweb-platform-tests%2Ftrusted-types%2FDedicatedWorker-constructor-from-SharedWorker.html

TEXT DIFF:

@@ -1,7 +1,4 @@
 
-FAIL Creating a Worker from a string should throw (shared worker scope) assert_throws_js: Creating a Worker threw function "() => { new Worker("w"); }" threw object "ReferenceError: Can't find variable: Worker" ("ReferenceError") expected instance of function "function TypeError() {
-    [native code]
-}" ("TypeError")
-FAIL Creating a Worker from a TrustedScriptURL should not throw (shared worker scope) Can't find variable: Worker
-FAIL Creating a Worker from a string with a default policy should not throw (shared worker scope) Can't find variable: Worker
+Harness Error (TIMEOUT), message = null
 
+

DIFF URL:

https://build.webkit.org/results/Apple-iOS-18-Simulator-Release-WK2-Tests/293017@main%20(1244)/imported/w3c/web-platform-tests/trusted-types/DedicatedWorker-constructor-from-SharedWorker-pretty-diff.html

REPRODUCIBILITY:

I was unable to reproduce on the iOS 18.2 simulator.

REGRESSION:

This test appears to be failing since it was imported at 292363@main.

[Radar WebKit Bug Importer]

<rdar://problem/148376545>

[EWS]

Test gardening commit 293049@main (b19a20d52837): <https://commits.webkit.org/293049@main>

Reviewed commits have been landed. Closing PR #43419 and removing active labels.

[Anne van Kesteren]

Luke, could you take a look at this?

[Luke Warlow]

I'm also unable to reproduce this using the iOS 18.4 simulator and minibrowser. This test isn't expected to actually pass because only Gecko seems to actually expose Workers inside of SharedWorkers, but I would expect a consistent set of errors as is marked in the expected file.

Given we don't expect this to actually pass I'd be inclined to Skip it if it's a flaky timeout. Running it doesn't really provide much?

[Anne van Kesteren]

I see, reportedly there is a problem with fetch_tests_from_worker at the harness-level so Skip seems acceptable.

[Luke Warlow]

*** This bug has been marked as a duplicate of bug 291345 ***