Hello zou, Tuesday, November 6, 2007, 9:35:29 AM, you wrote: zl> Try the following example, you'll see the differences between Gnash zl> and PP. Conceptually, we need the zl> 'normalization' step to finish the last two lines(commented in the zl> example code) if user forget to do that. And this zl> step is probably more complex in general cases and we don't have any zl> specs to define the "expected" behaviour.
Yes, that's the big problem. zl> That's why I said drawingAPI users should be responsible for their zl> shapes. Note that the drawing API is mostly used for dynamic vectors. What the "user" draws might depend greatly on some dynamic data which in some situations might produce "malformed" shapes. This means that you force users to do automatic normalization of dynamic shapes, so this does not solve the problem - it just moves it around. Anyway, since the proprietary player does not require such "normalization" we can't do it either. Normalization in the renderer: nope, duplicated code Normalization in the drawing API: yes, sir! zl> // lineTo(200, 200); // with or without these two lines doesn't make zl> a difference with pp. zl> // endFill(); This could be a different bug. I guess the drawing API just does not implicitely endFill() at the end of the script, which it should obviously do. Udo _______________________________________________ Gnash-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnash-dev
