[webkit-reviews] review denied: [Bug 54929] REGRESSION: Text selection broken for text with line-height applied : [Attachment 85560] Patch

[bugzilla-daemon at webkit.org]

Ojan Vafai <ojan at chromium.org> has denied Levi Weintraub <leviw at chromium.org>'s
request for review:
Bug 54929: REGRESSION: Text selection broken for text with line-height applied
https://bugs.webkit.org/show_bug.cgi?id=54929

Attachment 85560: Patch
https://bugs.webkit.org/attachment.cgi?id=85560&action=review

------- Additional Comments from Ojan Vafai <ojan at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=85560&action=review

> Source/WebCore/ChangeLog:9
> +	   When an InlineBox has a line-height smaller than its actual height
it results in a negative logicalTop for
> +	   associated InlineTextBoxes. This negative value won't necessarily
result in a point the RenderText 

Is ever having a negative logicalTop a bug? Shouldn't logicalTop always be
non-negative? I don't actually know. Hyatt?

> Source/WebCore/rendering/RenderBlock.cpp:4077
> +	   IntPoint point(pointInLogicalContents.x(), closestBox->isText() ?
closestBox->root()->selectionTop() : closestBox->logicalHeight());

The selectionTop of the rootlinebox is not necessarily inside the closestBox,
is it? As in, what if the closestBox is on the second line?