[SciPy-dev] Question including a boundary value problem solver in SciPy

Neil Martinsen-Burrell nmb@wartburg....
Tue Apr 14 09:08:22 CDT 2009

John Salvatier <jsalvati <at> u.washington.edu> writes:

> Sorry for my slow response to this, I had some trouble getting added to the 
> scipy-dev list. I did not realize that SciPy did not depend on Fortran 90. It 
> is too bad that my wrapper cannot be included in the main distribution. Could 
> you tell me who I should talk to about creating a SciKit?Also, I am 
> interested in the wrapper you have made; could you point me to your code or 
> send it to me? Perhaps I can improve my own code by looking at yours. I am 
> also aware of the other boundary value solver, and I think it *would* be a 
> good idea to have them both in a single SciKit.

I think that a BVP scikit would be a wonderful addition.  Along those lines, I
have a question.  I'm looking for a BVP solver that could be wrapped in python
that handles non-separated (particularly periodic) boundary conditions.  I've
looked at the classic Netlib codes, as well as the (excellent F90) BVP_SOLVER
that you've wrapped, but as far as I can tell, they are only for separated BCs.
 Do you know of a freely available periodic BVP code?  (Matlab's bvp4c can
handle non-separated BCs, but it is certainly not free.)


More information about the Scipy-dev mailing list