Bug 299830
| Summary: | REGRESSION(300617@main): [macOS] imported/w3c/web-platform-tests/storage-access-api/hasStorageAccess-insecure.sub.window.html is flaky | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Marta Darbinyan <darbinyan> |
| Component: | DOM | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | jbedard, m_finkel, sihui_liu, webkit-bot-watchers-bugzilla, webkit-bug-importer |
| Priority: | P2 | Keywords: | InRadar |
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| See Also: | https://bugs.webkit.org/show_bug.cgi?id=252268 | ||
Marta Darbinyan
The following test became flaky after 300617@main both in Release and Debug.
imported/w3c/web-platform-tests/storage-access-api/hasStorageAccess-insecure.sub.window.html
PASS [same-origin-frame] document.hasStorageAccess() should be supported on the document interface
PASS [same-origin-frame] document.hasStorageAccess() should be disallowed in insecure contexts
PASS [same-origin-frame] document.hasStorageAccess() should reject in a document that isn't fully active.
+PASS [cross-origin-frame] document.hasStorageAccess() should be supported on the document interface
+PASS [cross-origin-frame] document.hasStorageAccess() should be disallowed in insecure contexts
+PASS [cross-origin-frame] document.hasStorageAccess() should reject in a document that isn't fully active.
PASS [nested-same-origin-frame] document.hasStorageAccess() should be supported on the document interface
PASS [nested-same-origin-frame] document.hasStorageAccess() should be disallowed in insecure contexts
PASS [nested-same-origin-frame] document.hasStorageAccess() should reject in a document that isn't fully active.
-PASS [cross-origin-frame] document.hasStorageAccess() should be supported on the document interface
-PASS [cross-origin-frame] document.hasStorageAccess() should be disallowed in insecure contexts
-PASS [cross-origin-frame] document.hasStorageAccess() should reject in a document that isn't fully active.
PASS [nested-cross-origin-frame] document.hasStorageAccess() should be supported on the document interface
PASS [nested-cross-origin-frame] document.hasStorageAccess() should be disallowed in insecure contexts
PASS [nested-cross-origin-frame] document.hasStorageAccess() should reject in a document that isn't fully active.
Test history:
https://results.webkit.org/?suite=layout-tests&test=imported%2Fw3c%2Fweb-platform-tests%2Fstorage-access-api%2FhasStorageAccess-insecure.sub.window.html
Test result:
https://ews-build.webkit.org/#/builders/122/builds/38481
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Radar WebKit Bug Importer
<rdar://problem/161605161>
EWS
Test gardening commit 300765@main (c66650e9a65e): <https://commits.webkit.org/300765@main>
Reviewed commits have been landed. Closing PR #51572 and removing active labels.
Alexey Proskuryakov
This probably uncovered the issue by making the test run more correctly, and didn't cause it.
Jonathan Bedard
Can we just add `await` before each of the `RunTestsInIFrame` calls? I think that's an async testing call, it's not surprising that reliable ordering is an issue.
Sam Sneddon [:gsnedders]
This is again bug 252268, where we're having issues because we're relying on things that testharness.js API contract doesn't guarantee.