2011/5/22 Stefan Behnel <stefan...@behnel.de>: > Dag Sverre Seljebotn, 21.05.2011 09:07: >> >> On 05/21/2011 07:57 AM, Stefan Behnel wrote: >>> >>> In the future, I think we should be more careful with potentially >>> harmful options, and always prefer safety over speed - *especially* when >>> we know that the safe way will improve at some point. >> >> There wasn't a point where anybody wasn't careful about this; > > Sorry if I sounded offensive. It just felt too wrong from today's POV > (which, I hope, is properly reflected by my paragraph above). > > >> it is simply >> something that was inherited from Pyrex. The nonecheck directive came much >> later. > > Well, it wouldn't have been the first time we change the default behaviour > from the way Pyrex originally worked to something we deem more correct. > Looks like we missed that opportunity back then. > > I think Vitja's "uninitialised" branch is our way out of this. >
Now it's called "_control_flow" btw. Now None initialization is removed and everything seems to work. -- vitja. _______________________________________________ cython-devel mailing list cython-devel@python.org http://mail.python.org/mailman/listinfo/cython-devel
