[Webkit-unassigned] [Bug 120476] [CSS Masking] Implement luminance masking

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=120476





--- Comment #4 from Andrei Parvu <parvu at adobe.com>  2013-09-06 07:24:50 PST ---
(In reply to comment #3)
> (In reply to comment #2)
> > (From update of attachment 210071 [details] [details])
> > View in context: https://bugs.webkit.org/attachment.cgi?id=210071&action=review
> > 
> > > Source/WebCore/platform/graphics/Image.h:191
> > > +    bool isLuminanceMask() const { return m_isLuminanceMask; }
> > > +    void setIsLuminanceMask() { m_isLuminanceMask = true; }
> > 
> > This seems like a strange layer to add this concept at. All images have such a flag? The point of this class is an abstraction for an image. We shouldn’t be building in magic images into the platform layer. The platform layer can provide a new drawing function, but the actual logic for triggering it needs to be up at the filter/SVG level of the call.
> 
> I don't quite understand what you mean by filter/SVG level of the call. As I see it, another way to deal with this would be to add another parameter to the drawTiledImage / drawPattern functions, which would be defaulted to false. Is this what you wanted to suggest?

Looking through the code, it seems that we could either move the luminanceMask flag from the Image class to the GraphicsContext class, and then use it from there in the drawPattern functions, or, as I said above, add another parameter to the drawPattern function.  Darin, Dirk, any suggestions regarding one of these two approaches?

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