[webkit-reviews] review granted: [Bug 99521] Remove an unneeded repaint call in RenderTableCell::setCellLogicalWidth : [Attachment 169051] Proposed removal, includes testing as I didn't find any coverage for such repainting.

[bugzilla-daemon at webkit.org]

Darin Adler <darin at apple.com> has granted Julien Chaffraix
<jchaffraix at webkit.org>'s request for review:
Bug 99521: Remove an unneeded repaint call in
RenderTableCell::setCellLogicalWidth
https://bugs.webkit.org/show_bug.cgi?id=99521

Attachment 169051: Proposed removal, includes testing as I didn't find any
coverage for such repainting.
https://bugs.webkit.org/attachment.cgi?id=169051&action=review

------- Additional Comments from Darin Adler <darin at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=169051&action=review


> Source/WebCore/ChangeLog:16
> +	   All in all, the existing repainting logic already handles this case
properly.

The only thing I don’t like about this is that it requires global knowledge to
know that this isn’t needed. It’s something you can’t tell by just reading the
function itself. It’s nice to have the test coverage to mitigate this, but it
would be better if there was some good convention so what each function handles
and does not need to handle could be clearer.

A concrete way to put that is that I think a comment might be needed to explain
why this function does *not* have to deal with repainting. Depending on what
our conventions are for this sort of thing in our renderer code.