Bug 186838 – [WPE][GTK] ASSERTION FAILED: url == m_string in UserAgentQuirks test

• Status: RESOLVED
• Resolution: FIXED
• Component: Platform

[Rob Buis]

ASSERTION FAILED: url == m_string
../../Source/WebCore/platform/URL.cpp(101) : WebCore::URL::URL(WebCore::ParsedURLStringTag, const WTF::String&)
1   0x581c541 WebKitBuild/Debug/bin/TestWebKitAPI/WebCore/TestWebCore(WTFCrash+0x9) [0x581c541]
2   0x797297b WebKitBuild/Debug/bin/TestWebKitAPI/WebCore/TestWebCore(_ZN7WebCore3URLC1ENS_18ParsedURLStringTagERKN3WTF6StringE+0x103) [0x797297b]
3   0x58121f9 WebKitBuild/Debug/bin/TestWebKitAPI/WebCore/TestWebCore() [0x58121f9]
4   0x5813490 WebKitBuild/Debug/bin/TestWebKitAPI/WebCore/TestWebCore(_ZN13TestWebKitAPI25UserAgentTest_Quirks_Test8TestBodyEv+0x166) [0x5813490]

[Rob Buis]

Created attachment 343142 [details]
Patch

[Frédéric Wang (:fredw)]

Comment on attachment 343142 [details]
Patch

OK, I think the slash is not important in the test for user agent quirks and other URLs use it, so let's do it. The core issue seems to be bug 167093 though, it would be nice to fix it at some point...

[Frédéric Wang (:fredw)]

(In reply to Frédéric Wang (:fredw) from comment #2)
> Comment on attachment 343142 [details]
> Patch
> 
> OK, I think the slash is not important in the test for user agent quirks and
> other URLs use it, so let's do it. The core issue seems to be bug 167093
> though, it would be nice to fix it at some point...

Actually per bug 167093 comment 3, it is wrong to pass a non-canonicalized URL so we must take this patch.

[WebKit Commit Bot]

Comment on attachment 343142 [details]
Patch

Clearing flags on attachment: 343142

Committed r233008: <https://trac.webkit.org/changeset/233008>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/41283884>

[Michael Catanzaro]

Thanks for this fix. Please remember to CC [e-mail] so that we all see the bug! It's done automatically if you select the 'WebKit Gtk' or 'WebKit WPE' Bugzilla components, but when you pick other components it has to be done manually.

[Rob Buis]

Will do! May be even better if it was done automatically given the [GTK] prefix.

[Michael Catanzaro]

Bugzilla is not smart enough for that, I'm afraid....

[Rob Buis]

Reopening to attach new patch.

[Rob Buis]

Created attachment 344059 [details]
Patch

[Michael Catanzaro]

Comment on attachment 344059 [details]
Patch

Sigh, we really need to improve our gardening so that we can get a handle on these issues.  No doubt the test started failing immediately after I introduced this bug, but probably nobody has checked the API tests on the debug bot ever since.

[EWS Watchlist]

Comment on attachment 344059 [details]
Patch

Attachment 344059 [details] did not pass win-ews (win):
Output: https://webkit-queues.webkit.org/results/8406605

New failing tests:
http/tests/security/contentSecurityPolicy/userAgentShadowDOM/allow-audio.html

[EWS Watchlist]

Created attachment 344063 [details]
Archive of layout-test-results from ews206 for win-future

The attached test failures were seen while running run-webkit-tests on the win-ews.
Bot: ews206  Port: win-future  Platform: CYGWIN_NT-6.1-2.9.0-0.318-5-3-x86_64-64bit

[WebKit Commit Bot]

Comment on attachment 344059 [details]
Patch

Clearing flags on attachment: 344059

Committed r233416: <https://trac.webkit.org/changeset/233416>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.