You are (understandably) misunderstanding. You need to look at the hack in siplib.c where createType() is called. The sipTypeDef pointer is passed to createType() and the returned sipWrapperType pointer replaces the original pointer in the table.
Oh, oh, yes, that's it. Sorry for bugging you -- I've in fact seen this before, but it slipped my mind.
It's my bug -- I'm grabbing the sipWrapperType ptr out of that table before the module init function is called, when the ptr is invalid. Doh!
Is there a good way of injecting code into the module init function? Right now, I'm using global singleton objects in C++ to force my out-of-band initializers to run (hence the race with the module init).
If not, I do have a workaround.
Thanks, Dave
_______________________________________________ PyKDE mailing list [EMAIL PROTECTED] http://mats.imk.fraunhofer.de/mailman/listinfo/pykde
