[webkit-reviews] review canceled: [Bug 81057] Fix rounding and usage of LayoutUnits in RenderBoxModelObject : [Attachment 132126] Patch

[bugzilla-daemon at webkit.org]

Julien Chaffraix <jchaffraix at webkit.org> has canceled Emil A Eklund
<eae at chromium.org>'s request for review:
Bug 81057: Fix rounding and usage of LayoutUnits in RenderBoxModelObject
https://bugs.webkit.org/show_bug.cgi?id=81057

Attachment 132126: Patch
https://bugs.webkit.org/attachment.cgi?id=132126&action=review

------- Additional Comments from Julien Chaffraix <jchaffraix at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=132126&action=review


> Source/WebCore/rendering/RenderBoxModelObject.cpp:1144
> -	   geometry.setPhaseX(geometry.tileSize().width() ?
layoutMod(geometry.tileSize().width() - (xPosition + left),
geometry.tileSize().width()) : LayoutUnit(0));
> +	   geometry.setPhaseX(geometry.tileSize().width() ?
geometry.tileSize().width() - (xPosition + left) % geometry.tileSize().width()
: 0);

Those 2 lines are not equivalent due to % being of higher priority that -. It's
sad that our tests are not catching that. Extra points if you can find a test
to catch that regression.

> Source/WebCore/rendering/RenderBoxModelObject.cpp:1150
> +	   geometry.setPhaseY(geometry.tileSize().height() ?
geometry.tileSize().height() - (yPosition + top) % geometry.tileSize().height()
: 0);

Ditto.