On, Sat Jun 07, 2008, Brian Fisher wrote: > On Sat, Jun 7, 2008 at 1:19 PM, Marcus von Appen <mva@xxxxxxxxxxxx> wrote: > > > After this manual unlock, the PixelArray is unlikely to work in any > > case. You might expect weird behaviour in some cases, random > > segmentation faults and other crashes on trying to access self.pixels > > after the unlock. See my other mail about the why and how to avoid such > > things. > > > > would it be good to try and turn that into an exception? like if the > PixelArray would check the lock count or something? Surfaces do not feature a tracking of lock counts at the moment. And checking for an existing lock might mess up user intendend locking behaviour, I think. Usually one should stick to the docs about locks caused by objects async methods :-). Regards Marcus
Attachment:
pgpG6F1jNMGvd.pgp
Description: PGP signature