fullscreen/exit-full-screen-iframe.html is failing

[Filip Pizlo]

This test is failing:


--- /Volumes/Data/pizlo/secondary/OpenSource/WebKitBuild/Debug/layout-test-results/fullscreen/exit-full-screen-iframe-expected.txt
+++ /Volumes/Data/pizlo/secondary/OpenSource/WebKitBuild/Debug/layout-test-results/fullscreen/exit-full-screen-iframe-actual.txt
@@ -5,6 +5,6 @@
 EVENT(webkitfullscreenchange)
 TEST(document.getElementById('frame').contentDocument.webkitIsFullScreen==true) OK
 EVENT(webkitfullscreenchange)
-TEST(document.getElementById('frame').contentDocument.webkitIsFullScreen==false) OK
+TEST(document.getElementById('frame').contentDocument.webkitIsFullScreen==false) FAIL
 END OF TEST

[Filip Pizlo]

Skipped in http://trac.webkit.org/changeset/122040

[Alexey Proskuryakov]

Also fails on chromium: bug 90704. Should r122035 be just rolled out?

Filip, please be sure to CC responsible parties when skipping tests.

[Max Feil]

(In reply to comment #2)
> Also fails on chromium: bug 90704. Should r122035 be just rolled out?

I believe this test should pass, and that the failure indicates a real problem that needs to be investigated.

After webkitCancelFullScreen() is called from JavaScript and the resulting webkitfullscreenchange event is received, webkitIsFullScreen should no longer be true.

Rolling out r122035 would revert the failing test, but it would also revert a valid BlackBerry code fix.

[Max Feil]

(In reply to comment #2)
> Also fails on chromium: bug 90704.

Note that this test causes EFL and GTK ports to crash, which is are real failures that need to be fixed. See also bug 89817 for my attempt at a patch which would have prevent these crashes but did not pass review.

[Max Feil]

(In reply to comment #4)
> Note that this test causes EFL and GTK ports to crash

GTK port crash is now fixed (bug 90719) and this test passes on GTK (see http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/ ).

[Antonio Gomes]

(In reply to comment #5)
> (In reply to comment #4)
> > Note that this test causes EFL and GTK ports to crash
> 
> GTK port crash is now fixed (bug 90719) and this test passes on GTK (see http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/ ).

EFL is also fixed in trunk.

[Eric Carlson]

(In reply to comment #6)
> (In reply to comment #5)
> > (In reply to comment #4)
> > > Note that this test causes EFL and GTK ports to crash
> > 
> > GTK port crash is now fixed (bug 90719) and this test passes on GTK (see http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/ ).
> 
> EFL is also fixed in trunk.

This test passes on the Mac port as well: http://build.webkit.org/results/Apple%20MountainLion%20Debug%20WK2%20(Tests)/r153825%20(11521)/results.html

Un-skipped in https://trac.webkit.org/r153834 .

Marking bug as RESOLVED/WORKSFORME.

[Eric Carlson]

(In reply to comment #7)
> (In reply to comment #6)
> > (In reply to comment #5)
> > > (In reply to comment #4)
> > > > Note that this test causes EFL and GTK ports to crash
> > > 
> > > GTK port crash is now fixed (bug 90719) and this test passes on GTK (see http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/ ).
> > 
> > EFL is also fixed in trunk.
> 
> This test passes on the Mac port as well: http://build.webkit.org/results/Apple%20MountainLion%20Debug%20WK2%20(Tests)/r153825%20(11521)/results.html
> 
> Un-skipped in https://trac.webkit.org/r153834 .
> 
> Marking bug as RESOLVED/WORKSFORME.

Actually the test is flakey on both Mac and EFL ports. Marked as flakey on Mac in https://trac.webkit.org/r153846.

[Alexey Proskuryakov]

*** This bug has been marked as a duplicate of bug 139779 ***