About a Fortran Scientific Library

That is certainly great. And it has GPLv3! :heart:
But I also struggled to see how to make it work in Linux natively…

I think it’s great people is sharing their knowledge on different numerical Fortran routines. However, I think the point is to think which is the best way to create an standard numerical library that every user can easily install (with the fpm) so that her/his code can easily be compiled and run by other users.

It is true that there’s a lot of good code that can (have to) be (re)used. I think the best way to, at least as starting point, is to try to interface this old well-known code and with modern fortran, and maybe first to define the interface rules we want to set. Once this is done, all the open source routines that has been posted here can be interfaced and added to the fpm.

Does that make sense?

1 Like

One that I noticed on GitHub today is the Library of Numerical Analysis of the Research Computing Center of the Moscow State University . It looks comprehensive. The code comments are in Russian, and procedure names such as ADB1C match the documentation but are not informative on their own. Default reals are used. C and Pascal versions are also available.

Using Google Translate, the catalog is

3 Likes