Robert wrote: >>> B) >>> >>> Given this change (which could happen in -devel), is there any >>> obstacles >>> to getting rid of the detailed type system in PyrexTypes? Basically >>> Cython would only deal with "signed", "unsigned" and "float" (and in >>> time "complex"). >>> >>> Doing >>> >>> cdef short i >>> >>> would basically flag i as "signed"; the "short-ness" would only be >>> present as the cname of the type. > > We use the relative ranking to determine the type of arithmetic > operations. This will become even more important when type inference > is used.
Hmm. Yes...what are those used for for which a size-less system isn't sufficient? I can think of declaration of temps, but nothing else. But that means there's still a real problem even if I fix #303; if we don't know the exact type for external types then e.g. temporaries from expressions of those could loose precision. Would it work to always allocate temps at the widest possible precision? Dag Sverre _______________________________________________ Cython-dev mailing list [email protected] http://codespeak.net/mailman/listinfo/cython-dev
