Bug 93911 – AutoTableLayout truncates preferred widths for cells when it needs to ceil them to contain the contents

• Status: RESOLVED
• Resolution: FIXED
• Component: Layout and Rendering

[Levi Weintraub]

Repros in various places on Facebook.com

[Eric Seidel (no email)]

our table code needs some no-holds-barred lovin.

[Levi Weintraub]

Created attachment 158168 [details]
Patch

[Eric Seidel (no email)]

It's hard for me to tell if this is an improvement in the various rendering results.

[WebKit Review Bot]

Comment on attachment 158168 [details]
Patch

Attachment 158168 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/13488455

New failing tests:
svg/zoom/page/zoom-img-preserveAspectRatio-support-1.html
svg/zoom/page/zoom-svg-through-object-with-percentage-size.xhtml
fast/transforms/bounding-rect-zoom.html
tables/mozilla_expected_failures/bugs/bug89315.html
svg/zoom/page/zoom-replaced-intrinsic-ratio-001.htm

[WebKit Review Bot]

Created attachment 158188 [details]
Archive of layout-test-results from gce-cr-linux-03

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: gce-cr-linux-03  Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'>  Platform: Linux-2.6.39-gcg-201203291735-x86_64-with-Ubuntu-10.04-lucid

[Levi Weintraub]

(In reply to comment #3)
> It's hard for me to tell if this is an improvement in the various rendering results.

A handfull of them are incorrectly wrapping in the current test expectations. A couple are difficult to tell.

[Levi Weintraub]

(In reply to comment #4)
> (From update of attachment 158168 [details])
> Attachment 158168 [details] did not pass chromium-ews (chromium-xvfb):
> Output: http://queues.webkit.org/results/13488455
> 
> New failing tests:
> svg/zoom/page/zoom-img-preserveAspectRatio-support-1.html
> svg/zoom/page/zoom-svg-through-object-with-percentage-size.xhtml
> fast/transforms/bounding-rect-zoom.html
> tables/mozilla_expected_failures/bugs/bug89315.html
> svg/zoom/page/zoom-replaced-intrinsic-ratio-001.htm

Also, all these tests are marked pass/fail on Linux as part of this change. I find it confounding that the EWS bot seemingly ignores changes to TestExpectations on the test pass! Or, perhaps I'm doing it wrong...

Either way, I'm confident this change is for the better. Eric, let me know if you want to discuss :)

[Eric Seidel (no email)]

Comment on attachment 158168 [details]
Patch

OK.  Please update your ChangeLog to explain why all these tests are progressions, how they only affect subpixel ports, and how this was a regression from turning on subpixel.  Ideally you'd encorporate the cr-linux results, but I understand our tools don't make that easy for you.  I guess you're stuck with garden-o-matic for now. :)

[Levi Weintraub]

Comment on attachment 158168 [details]
Patch

Clearing flags.

[Levi Weintraub]

Landed in http://trac.webkit.org/changeset/125694