Re: Uniform arrays and numerical stuff

[Valentin Kamyshenko <val at kamysh dot materials dot kiev dot ua>]

>>>>> "David" == David Lutterkort <lutter@cise.ufl.edu> writes:

David> Hi,

David> A year ago there were some lively discussions on using Guile
David> for numerical work. I looked at the mail archives but couldn't
David> find answers to a few questions. Maybe somebody could help out:

David> (1) Chris Lee talked about wrappers for BLAS and LAPACK he had
David> written. Are they available somewhere ?

I've just sent to guile-sources@gnu.org a small utility (about 20k in
.tar.gz form), called `g+f', which I use for easy generation of such
wrappers. Unfortunately, I was not able to put it directly to
ftp://pub/guile/contrib/incoming because of the write permission.
(If you like, I can send it directly to you).

It is still very incomplete, but it appeared to be useful for me (I
use it about a year). 
In simple cases you just write smth like:
	int blasfnc(double a[n],int n) => blas:fnc(a)
and it generates a glue code (where blasfnc is the fortran function,
blas:fnc - guile function).

David> (2) Mark Galassi was going to set up a mailing list for
David> numeical stuff + Guile. Is that list still existing/active ?
There was a good discussion there during few months a year ago, but
now it is dead (so far as I see).

--
Best regards,
	Valentin.