Hi Tom,

Looking over the code, yes, I believe it could be wrapped pretty straightforwardly. The main loop seems to be in powmes.f90, in the routine powmes, and it goes:

input_parms do_the_job read_part output_results

What I think we could do is override the input_parms function, supply a slight modification to read_part that simply points the supplied numpy arrays into the allocatable arrays (memory doubling would ultimately be a concern, but maybe we can get around that ...) and then pulls the results back into the main body.

To do this, we could either use a C wrapper, using F/C linkage (Cython would actually work nicely for this) or we could use something like f2py. I'm happy to give either assistance, or have a go at this, but I don't think I'll get to it before the week after next. (Although there is a possibility I could track it down next week.)

What do you think? I'm actually pretty intrigued by this... it would be very nice ot have a simple power spectrum generator for the N-body component.

-Matt

On Fri, Nov 19, 2010 at 6:46 AM, Tom Abel tabel@slac.stanford.edu wrote:

Hi All,

what do you think of wrapping http://www.projet-horizon.fr/www.projet-horizon.fr/article345.html into yt ?

Would this be difficult ? I think it potentially would have lots of users, don't you think?

Best, Tom

Tom Abel Associate Professor of Physics Kavli Institute for Particle Astrophysics and Cosmology, Stanford University tabel@stanford.edu

---

Yt-dev mailing list Yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org