Thanks for the quick reply. Could you say just a little more about how to work around this issue? The numpy.pxd file is close to 1000 lines long, and I don't know enough about Cython or NumPy to dive in and do a hot-fix.
In particular, can I get around this without patching numpy.pxd? I've noticed that np.long_t seems to map correctly to the underlying C long type -- would it be simplest to create NumPy arrays with dtype np.long for the moment and adjust my C code to expect long integers? Thanks again, Zak On Sun, Dec 27, 2009 at 10:36 AM, Dag Sverre Seljebotn <[email protected]> wrote: >> Hello everyone, >> >> A very simple novice question: how can I determine the NumPy type that >> Cython will agree to map to, say, the ordinary int type in C? I'm >> wrapping a C function that accepts an int * argument, and I'm looking >> for the appropriate NumPy type for a ctypedef statement. When I make >> the obvious guesses, I see the same error message that someone else >> describes here: > > Use numpy.intc for that purpose. > > However there's no numpy.intc_t in numpy.pxd in Cython, this is a bug, > patch welcome (but it would just be a typedef for "int"). > > I also believe numpy.double maps "more directly" to double float than > numpy.float64. > > Dag Sverre > > _______________________________________________ > Cython-dev mailing list > [email protected] > http://codespeak.net/mailman/listinfo/cython-dev > _______________________________________________ Cython-dev mailing list [email protected] http://codespeak.net/mailman/listinfo/cython-dev
