Currently drawImageFromRect re-implements pretty much everything drawImage does already for it. It should just use drawImage directly instead.
Created attachment 89608 [details] Patch
Comment on attachment 89608 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=89608&action=review > Source/WebCore/bindings/js/JSCanvasRenderingContext2DCustom.cpp:299 > - ustringToString(exec->argument(9).toString(exec))); > - return jsUndefined(); > + ustringToString(exec->argument(9).toString(exec)), ec); > + setDOMException(exec, ec); > + > + return jsUndefined(); This change looks unrelated. If you did intend to include it, then the changelog should describe it, and the testcase should exercise it.
Attachment 89608 [details] did not build on chromium: Build output: http://queues.webkit.org/results/8397690
Created attachment 89616 [details] Patch
> > Source/WebCore/bindings/js/JSCanvasRenderingContext2DCustom.cpp:299 > > - ustringToString(exec->argument(9).toString(exec))); > > - return jsUndefined(); > > + ustringToString(exec->argument(9).toString(exec)), ec); > > + setDOMException(exec, ec); > > + > > + return jsUndefined(); > > This change looks unrelated. If you did intend to include it, then the changelog should describe it, and the testcase should exercise it. Good call. I initially thought it would be best for drawImageFromRect to behave just as drawImage does WRT dom exceptions and such. Though, since it's simply a legacy routine used by only a few sites and dashboard widgets, I think it's best to leave its behavior as-is. Updated patch does so.
Attachment 89608 [details] did not build on chromium: Build output: http://queues.webkit.org/results/8400761
Committed r83888: <http://trac.webkit.org/changeset/83888>