This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: Data Structure used in the ODE part
- From: Gerard Jungman <jungman at lanl dot gov>
- To: ben dot brueggemann at physik dot hu-berlin dot de
- Cc: gsl-discuss at sources dot redhat dot com
- Date: Mon, 10 Dec 2001 14:45:52 -0700
- Subject: Re: Data Structure used in the ODE part
- Organization: LANL T-3
- References: <3C151D18.1559521E@physik.hu-berlin.de>
ben.brueggemann@physik.hu-berlin.de wrote:
>
> Wouldn't it be nice if the data structure of the ordinary differential
> equations would follow the same standard as the rest of gsl ?
> So I would rather like to evolve a gsl_vector or even better a
> gsl_matrix than the used double array. Then I could even use
> the BLAS functions to manipulate the right hand side..
> Thanks
> Ben.
Hi. That was done deliberately, and a similar decision was made
in some other parts of the library. The idea is to make the interfaces
as wide as possible; for instance, if we had only gsl_vector type
interfaces for these functions, then there would people with
the opposite complaint.
If there is enough interest, wrapper functions can be created
with gsl-type interfaces. It's easy to do, but we just didn't
have the time to do that and never got around to it.
Meanwhile, you can use the members for gsl_vector and
gsl_matrix which allow access to the underlying data
representation. Maybe a little annoying, but functionally
the same.
Thanks.
--
G. Jungman