[SciPy-dev] Inclusion of cython code in scipy

[David Cournapeau]

Travis E. Oliphant wrote:
>
> I would prefer it, however, if the ability to compile Python were done 
> without changing the language.  

That's certainly the best way, and that's why I am a bit wary about 
replacing python code by cython (once it becomes possible to 
compile/accelerate pure python code, we would have to code cython 
modules again; I would much prefer replacing some C by cython, and be 
sure there is always a pure python implementation available, even if 
really slow).

>  It looks like Cython is moving in that 
> direction, but it is not quite there yet. 
>
> My ideal world would be the ability to emit machine instructions 
> directly at various points in the Python code by just writing a 
> restricted set of Python syntax.  Perhaps this could be seamless.
>   

Not that it is in any useful form, but pypy folks are now working on 
supporting floating point operations in their jit:

http://morepypy.blogspot.com/2008/04/float-operations-for-jit.html

cheers,

David