[webkit-reviews] review granted: [Bug 83116] Add helpers to create anonymous table parts. : [Attachment 135611] Patch

[bugzilla-daemon at webkit.org]

Julien Chaffraix <jchaffraix at webkit.org> has granted Abhishek Arya
<inferno at chromium.org>'s request for review:
Bug 83116: Add helpers to create anonymous table parts.
https://bugs.webkit.org/show_bug.cgi?id=83116

Attachment 135611: Patch
https://bugs.webkit.org/attachment.cgi?id=135611&action=review

------- Additional Comments from Julien Chaffraix <jchaffraix at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=135611&action=review


> Source/WebCore/ChangeLog:9
> +	   introducing a new static function
createAnonymousWithParentRendererAndDisplay.

s/createAnonymousWithParentRendererAndDisplay/createAnonymousWithParentRenderer
/

> Source/WebCore/ChangeLog:10
> +	   The function builds a new anonymous wrapper inheriting style
properties from

a new anonymous wrapper of the same type as the class, inheriting style
properties...

> Source/WebCore/rendering/RenderBlock.cpp:7422
> +    // FIXME: Why are we forcing a BOX display for old flex box here ? Is
this a hack for RenderButton ?

I think it should be stated more as:

// FIXME: Do we need to convert all our inline displays to block-type in the
anonymous logic?

There is a function already doing that on CSSStyleSelector that we could reuse
if that was going to happen (equivalentBlockDisplay).

> LayoutTests/ChangeLog:11
> +	   *
platform/chromium-mac/fast/css/bidi-override-in-anonymous-block-expected.txt:
> +	   *
platform/chromium-win/fast/css/bidi-override-in-anonymous-block-expected.txt:

This needs to be skipped on all other platforms as you will be breaking them if
not. For Chromium something like (to be double checked against our baselines):

WK83116 LINUX WIN : fast/css/bidi-override-in-anonymous-block.html = TEXT

For other platforms, it should just be added to the Skipped file.