[LiDIA] compile LiDIA

[Jordi Gutierrez Hermoso]

On 21/04/07, Kuehner Joseph <jrkuehner at t-online.de> wrote:
> I know that LiDIA is no more supported ...

Is there an official statement of this somewhere?

>  Nevertheless I tried to compile LiDia 2.2.0 on a Fedora core 6 system.
> Base and Finite Fields worked fine. Linear Algebra gets errrors. The
> reason seams to be repeated calls to member function "assign(..)" in
> various header files. Generaly speaking it has to be replaced by
> "base_matrix< T >::assign(..)" or something similar.

That's odd... I do wonder where the change came about. I haven't
attempted to compile LiDIA in over two years, however. I wasn't aware
that there were even any modifications since.

It may be a problem with g++ getting better about how it should
resolve dependent names in template lookups. Without looking at the
code, I guess that this -> assign() instead of
base_matrix<T>::assign() might also fix it.

> Doas anybody care about? Or is LiDIA simply out?

I care! LiDIA is still in my list of projects that should not die. I
haven't done number theory in a long time, unfortunately, so I'm not
sure if the free software community really needs the functionality in
LiDIA that can't be found elsewhere (e.g. Pari).

I've been meaning for a long while to package LiDIA into a Debian
package... unfortunately, I keep getting busy with other things (such
as writing a thesis).

- Jordi G. H.