|Summary:||Update image results for SVG as CSS border image tests|
|Product:||WebKit||Reporter:||Simon Fraser (smfr) <simon.fraser>|
|Component:||SVG||Assignee:||Simon Fraser (smfr) <simon.fraser>|
|Severity:||Normal||CC:||bdakin, eric, jparent, krit, xan.lopez, yael, zimmermann|
|Version:||528+ (Nightly build)|
|OS:||OS X 10.5|
Description Simon Fraser (smfr) 2010-04-02 10:16:01 PDT
The pixel results of LayoutTests/fast/borders/svg-as-border-image-2-diffs.html show that SVG as border image is broken.
Comment 1 Simon Fraser (smfr) 2010-04-02 10:17:09 PDT
fast/borders/svg-as-border-image.html also broken
Comment 2 Simon Fraser (smfr) 2010-04-02 10:51:03 PDT
This was broken by http://trac.webkit.org/changeset/54865
Comment 3 Yael 2010-04-02 11:42:49 PDT
I think the expected result is wrong, and the same is for fast/images/svg-as-tiled-background.html. The expected result of fast/borders/border-image-01.html shows that the image covers the whole div, so I don;t understand why this is not the case for SVG images.
Comment 4 Simon Fraser (smfr) 2010-04-28 22:58:12 PDT
Yes, I think the expected image are wrong. I have some better tests to land.
Comment 5 Simon Fraser (smfr) 2010-04-28 22:58:46 PDT
Created attachment 54678 [details] Better tests (zip)
Comment 6 Nikolas Zimmermann 2010-04-29 00:57:15 PDT
Nice tests! If you upload a patch, I'm happy to r+ them.
Comment 7 Nikolas Zimmermann 2010-04-29 00:58:21 PDT
*** Bug 35323 has been marked as a duplicate of this bug. ***
Comment 10 Eric Seidel (no email) 2010-04-30 11:01:52 PDT
It makes no sense to me, but it seems the Gtk bot seems more likely to crash in worker tests after this change (which seems impossible to be related). We'll have to monitor and see if it's just a bot issue.
Comment 11 Eric Seidel (no email) 2010-04-30 11:36:09 PDT
I don't understand it, but we've had 4 worker crashes in a row on the Gtk bot starting on this commit. http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r58580%20(2832)/results.html http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r58582%20(2834)/results.html http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r58583%20(2836)/results.html http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r58584%20(2838)/results.html
Comment 12 Eric Seidel (no email) 2010-04-30 12:08:07 PDT
The Gtk bot seems to have righted itself and stopped being mad at the world. I'll follow up with Xan to see if we've seen this before/how we can avoid false-positives like this in the future.
Comment 13 Eric Seidel (no email) 2010-05-02 19:31:07 PDT
Attachment 54777 [details] was posted by a committer and has review+, assigning to Simon Fraser for commit.