Dimitris Papavasiliou writes:
> It would be more logical to do it the other way round I think (that is
> implement just gsl_vector versions and let the user create a view of his
> arrays if needed) but it's the same thing I guess. It also means that
> my wrappers won't be of much use (and I just finished the templates for
> double and float today, oh well). Let me know if you're interested in
> any of them or the DCTs (although they'll need some work for the initial
> permutation).
If you want to put them on the web somewhere I can link to them from the
GSL webpage