[Webkit-unassigned] [Bug 112598] [chromium] http/tests/security/mixedContent/insecure-audio-video-in-main-frame.html is flaky because of console output

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=112598





--- Comment #2 from Mike West <mkwst at chromium.org>  2013-03-20 06:25:03 PST ---
(In reply to comment #1)
> (In reply to comment #0)
> > The test http/tests/security/mixedContent/insecure-audio-video-in-main-frame.html is flaky on all platforms because occasionally the console output attributes the error message to a line number. See http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20-%20chromium.org&showExpectations=true&tests=http%2Ftests%2Fsecurity%2FmixedContent%2Finsecure-audio-video-in-main-frame.html .
> > 
> > Marking this test as flaky.
> 
> For some reason, "document->parsing()" is periodically returning true when the onload event is triggered. This causes PageConsole::addMessage to check whether the scriptableDocumentParser isWaitingForScripts or isExecutingScripts. Neither is true, so we grab the line number, believing it to be a message tied to something in parsing.
> 
> I'll dig a bit deeper.

I think this is happening because we defer loading of media sources via 'HTMLMediaElement::scheduleDelayedAction(LoadMediaResource)'. This means that the mixed content and CSP checks (among others) will happen at some indeterminate point; perhaps before the onload event, perhaps after.

This is different from what we do for images, for instance: those checks happen synchronously (via ImageLoader::updateFromElement()), which makes it possible to pull a line number directly from the parser.

I'll have to think a bit about how to address this. Nothing brilliant is coming to mind. :/

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.