Bug 51786
| Summary: | [Windows 7 Release Tests] svg/dom/length-list-parser.html fails since r74788 | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Jessie Berlin <jberlin> |
| Component: | SVG | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED CONFIGURATION CHANGED | ||
| Severity: | Normal | CC: | ahmad.saleem792, jberlin, jschuh, krit, zimmermann |
| Priority: | P2 | Keywords: | LayoutTestFailure, PlatformOnly, Regression |
| Version: | 528+ (Nightly build) | ||
| Hardware: | All | ||
| OS: | Windows 7 | ||
Jessie Berlin
It appears that http://trac.webkit.org/changeset/74788 broke the Windows 7 Release Tests:
http://build.webkit.org/results/Windows%207%20Release%20(Tests)/r74833%20(7845)/svg/dom/length-list-parser-pretty-diff.html
I am not sure what is causing the extra 0 to appear, or if it really makes any difference. Thoughts?
I will be landing the failing results soon in order to get the bot green.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Jessie Berlin
Failing results landed in http://trac.webkit.org/changeset/74836
Justin Schuh
Thanks. We had to make the same change on the Chromium Windows bots. and I'm not in the habit of checking the non-core bots or I would have remedied that myself. I'll make sure to check them in the future.
I assume the cause is that the SVGLength handling code doesn't specify the leading zeroes for scientific notation, and the default is different on Windows. So, modifying length-list-parser.html just happened to tickle a difference that had always been there. I'll make time over the next few days to try and track this down and then remove the platform-specific expectations.
Ahmad Saleem
We don't have any platform specific expectation for this test while searching on 'searchfox.org' (wubkat).
So nothing to do here.
Marking as 'RESOLVED CONFIGURATION CHANGED'.