[Webkit-unassigned] [Bug 80571] [mac] Restore color space switching code to run-webkit-tests

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=80571





--- Comment #10 from Tim Horton <timothy_horton at apple.com>  2012-03-09 13:28:03 PST ---
(In reply to comment #7)
> (In reply to comment #6)
> > Created an attachment (id=130886)
 --> (https://bugs.webkit.org/attachment.cgi?id=130886&action=review) [details] [details]
> > patch for safekeeping
> > 
> > Patch, but still needs investigation of buffer colorspace choice vs. Chromium and going forward; just for safekeeping.
> 
> I've generated most results of the SVG pixel tests by forcing Generic RGB profile on my iMac before.
> I've rebaselined to trunk, and here are my findings:
> 
> - When I switch to Generic RGB profile manually, all tests pass, except the ones you've added to platform/mac/test_expectations.txt (they're now failing)
> - When I switch to iMac profile, more tests are failing, but also including those that you marked as IMAGE failing in the test_expectations.txt file on mac.
> 
> It's quite frustrating for me, as the trunk pixel test baseline (before the recent changes) passed with tolerance 0 on my both 64 bit machines, one running Lion, one running SL, and I'm used to test this way.
> Anyhow, now it's broken and I don't know which route to take to fix it.
> What's your plan to move on here? Why can't we add this now for Mac, until we find consensus with the cr guys on how to hook it into their TestShell.

My plan to move on here is to make sure that our results match cr-mac in the very simplest of cases, and then go with it! This is one of my plans for today.

I agree that it's very unfortunate, that's why I brought it up and am fixing it :-D

Thanks for rebaselining all of the tests, by the way! Hopefully those results will stay stable.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.