Summary: | REGRESSION(81374, 81384): editing/deleting/5206311-1.html hits assertion on non-Mac platforms | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Ryosuke Niwa <rniwa> | ||||||||
Component: | HTML Editing | Assignee: | Nobody <webkit-unassigned> | ||||||||
Status: | RESOLVED FIXED | ||||||||||
Severity: | Normal | CC: | adele, apavlov, aroben, darin, enrica, eric, justin.garcia, leviw, mrobinson, ossy, tonikitoo, tony | ||||||||
Priority: | P2 | Keywords: | LayoutTestFailure | ||||||||
Version: | 528+ (Nightly build) | ||||||||||
Hardware: | All | ||||||||||
OS: | All | ||||||||||
Attachments: |
|
Description
Ryosuke Niwa
2011-03-17 15:03:22 PDT
Committed r81397: <http://trac.webkit.org/changeset/81397> Now editing/deleting/5206311-2.html is asserting. It seems an earlier test is to blame. Created attachment 86135 [details]
Patch
Comment on attachment 86135 [details]
Patch
What about removing the test from the Skipped files?
(In reply to comment #4) > (From update of attachment 86135 [details]) > What about removing the test from the Skipped files? Oops, let me do that. Created attachment 86136 [details]
Patch
Tried skipping the previous test (5168598.html) on Windows instead Committed r81444: <http://trac.webkit.org/changeset/81444> (In reply to comment #6) > Created an attachment (id=86136) [details] > Patch I'm not going to fix the purple EWSs as removing lines from Skipped / text_expectations.txt is bound to encounter some conflicts. (In reply to comment #7) > Tried skipping the previous test (5168598.html) on Windows instead > > Committed r81444: <http://trac.webkit.org/changeset/81444> This seems to have made the Windows XP bot stop asserting. So it seems 5168598.html is the problem. The failure is now occurring on all DEBUG Chromium platforms (the faulty revision range is 81379:81398), I'm suppressing them now. (In reply to comment #9) > (In reply to comment #7) > > Tried skipping the previous test (5168598.html) on Windows instead > > > > Committed r81444: <http://trac.webkit.org/changeset/81444> > > This seems to have made the Windows XP bot stop asserting. So it seems 5168598.html is the problem. (In reply to comment #10) > The failure is now occurring on all DEBUG Chromium platforms (the faulty revision range is 81379:81398), I'm suppressing them now. It seems like there are some flakiness on the bot but it shouldn't affect the correctness of my patch. (In reply to comment #9) > This seems to have made the Windows XP bot stop asserting. So it seems 5168598.html is the problem. There is some flakiness in the tests that adds hyper links. Maybe there's some bug in the code that figures out visited-ness(?) of a URL. Created attachment 86182 [details]
updated Chromium test_expectations.txt
Comment on attachment 86182 [details]
updated Chromium test_expectations.txt
OK.
Thanks for the review! Committed r81518: <http://trac.webkit.org/changeset/81518> |