RESOLVED DUPLICATE of bug 40879 36210
plugins/resize-from-plugin.html fails on Qt
https://bugs.webkit.org/show_bug.cgi?id=36210
Summary plugins/resize-from-plugin.html fails on Qt
Maciej Stachowiak
Reported 2010-03-17 00:56:51 PDT
Some tests failing or crashing on Gtk/Qt with r56079
Attachments
proposed patch (5.69 KB, patch)
2010-03-18 10:55 PDT, Alexey Proskuryakov
no flags
Alexey Proskuryakov
Comment 1 2010-03-17 10:16:16 PDT
The new test added in r56079 was plugins/resize-from-plugin.html. It's failing due to a pre-existing issue, NPP_SetWindow seemingly isn't ever called: -x: 18, y: 52, width: 200, height: 200, clipRect: (18, 52, 200, 142) +x: 6619246, y: 7536756, width: 6357091, height: 6619248, clipRect: (0, 0, 0, 0) Height and width should equal 200, and the plug-in size should not change on scroll. Another test that's skipped on Gtk with a reference to this bug is plugins/return-error-from-new-stream-doesnt-invoke-destroy-stream.html. It was added in r34988, and r56079 couldn't affect it in any way. It does seem to crash randomly, and needs a separate bug.
Alexey Proskuryakov
Comment 2 2010-03-17 14:31:12 PDT
Filed bug 36228 for the old crashing test.
Alexey Proskuryakov
Comment 3 2010-03-18 10:47:27 PDT
It also fails on Windows. I have a patch that should get us closer.
Alexey Proskuryakov
Comment 4 2010-03-18 10:55:11 PDT
Created attachment 51051 [details] proposed patch
Adam Roben (:aroben)
Comment 5 2010-03-18 10:58:07 PDT
Comment on attachment 51051 [details] proposed patch r=me, though you could ask Anders Carlsson or Jon Honeycutt to look at the different result on Windows.
Alexey Proskuryakov
Comment 6 2010-03-18 11:08:06 PDT
Anders came up with an explanation, which I've put into ChangeLog. Committed <http://trac.webkit.org/changeset/56172>, and closing. We can reopen if bots are still unhappy.
Alexey Proskuryakov
Comment 7 2010-03-18 11:35:02 PDT
I'll land new results for Gtk - it gets a different clip rect, which may or may not be correct. Will also take results for Qt from buildbot, but these a definite failure.
Eric Seidel (no email)
Comment 8 2010-03-24 14:33:39 PDT
Attachment 51051 [details] was posted by a committer and has review+, assigning to Alexey Proskuryakov for commit.
Alexey Proskuryakov
Comment 9 2010-03-24 14:44:33 PDT
Comment on attachment 51051 [details] proposed patch Clearing review on a landed patch.
Alexey Proskuryakov
Comment 10 2010-03-24 14:45:02 PDT
This bug now represents a Qt problem with this test.
Andreas Kling
Comment 11 2010-06-30 03:04:06 PDT
*** This bug has been marked as a duplicate of bug 40879 ***
Note You need to log in before you can comment on or make changes to this bug.