Did you try this against trunk or GWT 2.3/2.4? I made some changes
recently to the way liveness of class literals are calculated that may or may not have an effect relative to this.
If you mean the "Roll seed-function optimization forward again" commit [1] (or others before it), yeah, I was hoping that would fix it, but it did not. [1] https://github.com/stephenh/scalagwt-gwt/commit/8b04f896d509bb71e59d2acb92e48e68a154bd1f)
Do you have the repo code?
Yep: https://github.com/stephenh/scalagwt-sample/commit/6cc1b70ae18c31063f71ce27fc110646c65c73c0 You shouldn't need that whole repo, just the classes in that commit (Foo, FragmentA, FragmentB, etc.). There is also some discussion on the scalagwt list about the bug [2]. Note that I misused the "L(...)" notation in a few places, so read those parts liberally. [2]: https://groups.google.com/d/msg/scalagwt/cVuvYEkJ5sU/XbfbOWUW2p8J http://gwt-code-reviews.appspot.com/1513803/ -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
