Bug 19604 - Re-enable 7 working layout tests
Summary: Re-enable 7 working layout tests
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.5
: P3 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks: 19569
  Show dependency treegraph
 
Reported: 2008-06-17 08:30 PDT by Jean-Charles Verdié
Modified: 2010-06-10 16:06 PDT (History)
1 user (show)

See Also:


Attachments
re-enable 7 layout tests (51.40 KB, patch)
2008-06-17 08:32 PDT, Jean-Charles Verdié
darin: review-
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Jean-Charles Verdié 2008-06-17 08:30:16 PDT
this patch re-enabled 7 working layout tests
Comment 1 Jean-Charles Verdié 2008-06-17 08:32:36 PDT
Created attachment 21755 [details]
re-enable 7 layout tests
Comment 2 Alexey Proskuryakov 2008-06-25 03:28:32 PDT
+        * http/tests/xmlhttprequest/web-apps/006-expected.txt: Added.

Why is this file added without any changes to the test itself (not that I think that this test should be enabled)
Comment 3 Darin Adler 2008-10-12 16:47:18 PDT
Comment on attachment 21755 [details]
re-enable 7 layout tests

I don't think you did enough research on why these tests were disabled.

For example, the two counters tests were disabled in <http://trac.webkit.org/changeset/18642> with this comment: "disabled 3 tests with results that are too font-dependent to be used as part of the test suite". So we can't just re-enable those tests unless we resolve the issue about getting appropriate fonts on everyone's computer.

And fast/frames/iframe-scroll-page-up-down.html fails when I run it on my machine, so I don't agree that it's working.

The crash that fast/images/svg-background-crash-on-refresh.html was causing was something that affected tests downstream. I'm not confident that bringing this patch back won't bring back the reason we turned it off in the first place; I'll need more evidence before doing that.

The create-list-1.html test does not give the expected result that's checked in, so if we wanted to bring it back we'd have to fix the expected result.

Overall, this patch isn't usable as-is. Changing to review-.