[webkit-reviews] review denied: [Bug 43852] [Qt] resizeToContent seems to trigger infinite resize on some pages : [Attachment 100675] patch

[bugzilla-daemon at webkit.org]

Kenneth Rohde Christiansen <kenneth at webkit.org> has denied Luiz Agostini
<luiz at webkit.org>'s request for review:
Bug 43852: [Qt] resizeToContent seems to trigger infinite resize on some pages
https://bugs.webkit.org/show_bug.cgi?id=43852

Attachment 100675: patch
https://bugs.webkit.org/attachment.cgi?id=100675&action=review

------- Additional Comments from Kenneth Rohde Christiansen
<kenneth at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=100675&action=review


r- and this won't apply anymore.

> Source/WebCore/page/DOMWindow.cpp:1095
> -    return static_cast<int>(view->height() / m_frame->pageZoomFactor());
> +    return static_cast<int>(view->logicalHeight() /
m_frame->pageZoomFactor());

I would suggest using view->visibleHeight() here. This will give the same
result for the non-tiled backing store case. As visibleContentRect() sets the
visibleHeight to view->height(). Or at least used to until the mac port
introduced a scaling workaround, using a new m_bounds or similar.

If this breaks for mac due to their workaround, we can ifdef it so that they
will continue using the current code. I don't understand their workaround well
enough to say whether this will be the case or not. But it might be as it seems
that they are lying about the actual size due to clipping.