Introduce the "layer-aware" RenderSVGRect implementation (inheriting from RenderSVGModelObject, and thus RenderLayerModelObject). Besides the different inheritance structure with respect to LegacyRenderSVGRect (inheriting from LegacyRenderSVGModelObject, and thus RenderElement -- not RenderLayerModelObject), the implementation is identical to LegacyRenderSVGRect.
Created attachment 448638 [details] Patch, v1
Comment on attachment 448638 [details] Patch, v1 This patch depends on 234954, postponing r? / EWS until it's resolved.
Created attachment 448728 [details] Patch, v2
Comment on attachment 448728 [details] Patch, v2 View in context: https://bugs.webkit.org/attachment.cgi?id=448728&action=review > Source/WebCore/rendering/svg/RenderSVGRect.h:33 > +#include "SVGRectElement.h" Could this be forward reference? > Source/WebCore/svg/SVGRectElement.cpp:94 > + if (0 && document().settings().layerBasedSVGEngineEnabled()) Maybe false is nicer.
(In reply to Rob Buis from comment #4) > Comment on attachment 448728 [details] > Patch, v2 > > View in context: > https://bugs.webkit.org/attachment.cgi?id=448728&action=review > > > Source/WebCore/rendering/svg/RenderSVGRect.h:33 > > +#include "SVGRectElement.h" > > Could this be forward reference? Yes, fixed, good catch. > > > Source/WebCore/svg/SVGRectElement.cpp:94 > > + if (0 && document().settings().layerBasedSVGEngineEnabled()) > > Maybe false is nicer. Both are ugly, but I agree false is less ugly here. Fortunately this is only needed for a short period of time. Thanks Rob again!
Committed r287834 (245886@trunk): <https://commits.webkit.org/245886@trunk>
<rdar://problem/87335756>