[Webkit-unassigned] [Bug 88908] [chromium] webkit-backface-visibility doesn't work with <video>

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=88908





--- Comment #16 from Christopher Cameron <ccameron at chromium.org>  2012-06-13 18:51:44 PST ---
(In reply to comment #15)
> (From update of attachment 147459 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=147459&action=review
> 
> > Source/WebCore/platform/graphics/chromium/LayerChromium.h:178
> > +    void setIsContentsLayer(bool isContentsLayer) { m_isContentsLayer = isContentsLayer; }
> > +    bool isContentsLayer() const { return m_isContentsLayer; }
> 
> At first glance I'm not a big fan of adding these to LayerChromium and beyond - it is a quirk of GraphicsLayerChromium that the doubleSided property applies to both a layer and its immediate child in some cases, GraphicsLayerChromium should take care of just setting the bits (like doubleSided) on the appropriate LayerChromiums.  The compositor itself doesn't need to know about the concept of "contents layer" 

The concern I have with that approach is that in order to get the correct results we must mark the parent as having preserve-3d as well (because the culling is done with respect to the transformation, which is applied to the parent).  This seems to break correctness because other children may be counting on preserve-3d not being set.

> it's way too confusing given the other naming collisions.

Another alternative would be to have a name the variable/method to express it does, "I will use my parent's backface culling properties" (m_useParentCullingLogic) instead of expressing when it is used "I'm a contents layer" (m_isContentsLayer).

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.