[chromium] REGRESSION: Incorrect preferred width calculation for table cells

• Component: Layout and Rendering
• Status: RESOLVED FIXED

[Emil A Eklund]

In fixing bug 93911 r125694 caused a problem with the preferred width calculations for table cells with a fractional padding where the logic to ensure that the cell is wide enough for the content returns different results for cells with a fixed width and cells with an automatic width.

[Emil A Eklund]

Created attachment 165471 [details]
Patch

[Emil A Eklund]

Downstream chromium bug: http://code.google.com/p/chromium/issues/detail?id=147496

[Levi Weintraub]

Comment on attachment 165471 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=165471&action=review

This LGTM other than coverage for Fixed tables. I'd feel better if Julien can give it a look as well before we land.

> Source/WebCore/ChangeLog:12
> +        In fixing bug 93911 r125694 caused a problem with the preferred
> +        width calculations for table cells with a fractional padding
> +        where the logic to ensure that the cell is wide enough for the
> +        content returns different results for cells with a fixed width
> +        and cells with an automatic width.

This change will also affect FixedTableLayout when we specify a fractional width, as it will be ceiled. Probably worth mentioning in the ChangeLog, and it wouldn't be a bad idea to have a specific test for that just so we note the behavior and track if it changes.

[Julien Chaffraix]

Comment on attachment 165471 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=165471&action=review

This change is fine. I would r+ if we had the proper fixed table layout testing.

>> Source/WebCore/ChangeLog:12
>> +        and cells with an automatic width.
> 
> This change will also affect FixedTableLayout when we specify a fractional width, as it will be ceiled. Probably worth mentioning in the ChangeLog, and it wouldn't be a bad idea to have a specific test for that just so we note the behavior and track if it changes.

Spot on, this should prevent a similar issue with fixed table layout so it should be tested.

[Emil A Eklund]

Cool, I'll add a test for fixed layout and re-upload.
Thanks for the quick review!

[Emil A Eklund]

Created attachment 165500 [details]
Patch

[Build Bot]

Comment on attachment 165500 [details]
Patch

Attachment 165500 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/13992858

New failing tests:
fast/sub-pixel/table-cells-have-stable-width.html

[Emil A Eklund]

Created attachment 165625 [details]
Patch

[Levi Weintraub]

Comment on attachment 165625 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=165625&action=review

Fix that changelog and we're good to go :)

> Source/WebCore/ChangeLog:14
> +        This also affects

Trailed off there ;)

[Emil A Eklund]

(In reply to comment #9)
> (From update of attachment 165625 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=165625&action=review
> 
> Fix that changelog and we're good to go :)
> 
> > Source/WebCore/ChangeLog:14
> > +        This also affects
> 
> Trailed off there ;)

Good catch, thanks!

[Emil A Eklund]

Committed r129529: <http://trac.webkit.org/changeset/129529>