David Cournapeau wrote: > On Wed, Mar 25, 2009 at 12:00 AM, Dag Sverre Seljebotn > <[email protected]> wrote: >> Gabriel Gellner wrote: >>>> [X] No, that would bloat the language and/or the compiler/project too >>>> much, I'd prefer to see this in a standalone tool/plugin >>>> >>>> >>> Just my opinion, but f2py rocks at this, so it seems like a distraction to >>> me. >>> >> That's certainly a valid argument against this. > > One potentially big advantage of cython is the ability to generate C > code compatible with python 3. I don't know how much work would be > required for f2py to do this, but if that's a non trivial task, then > it definitely make sense for cython to do this.
I don't know about f2py w/ respect to Py3 either. But in general, there's certain overlap in implementation between Cython and f2py (basically f2py does directly much of what Cython provides), which is a waste of developer resources anyway (well, if Cython is gearing up for more direct Fortran support, one way or another, over summer). I can certainly see Cython replacing or becoming an additional backend; so that f2py long-term becomes a tool which parses Fortran and generates Cython and Fortran sources instead of C and Fortran sources. We should keep in touch on GSoC projects related to this, I think a joint Cython/f2py project could be fruitful. (Is there anybody but you I should ping over at the SciPy camp about this?) -- Dag Sverre _______________________________________________ Cython-dev mailing list [email protected] http://codespeak.net/mailman/listinfo/cython-dev
