Hello Sandro, Wednesday, November 7, 2007, 11:17:08 PM, you wrote: SS> Contains 4 versions of your drawing (the same found in SS> DrawingApiTest.swf).
Ok, thanks. The file is very interesting, I'll analyze it now.. SS> Still this doesn't tell me wheter it's the renderer using SS> different behaviour or the "normalizer" implemented in the player SS> (and in the IDE for the proprietary case) My first assumption is that it does *not* have any kind of normalizer, because such a normalizer *normally* never produces paths that would confuse the rasterizer (which is the case for the upper left version). BTW, I think I found an easy solution that will work for SWF-defined, dynamic and malformed (regarding fill side) shapes. I'll try to implement it in a few moments to see if it passes our testcases. Udo _______________________________________________ Gnash-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnash-dev
