dolfin team mailing list archive

["Garth N. Wells" <g.n.wells@xxxxxxxxxx>]

Anders Logg wrote:
> I went for a slightly different (but better solution):
>
> Since PETScPreconditioner and uBlasPreconditioner don't have much in
> common, it's not obvious that we should make a typedef Preconditioner
> for one of them. They assume different data structures and if someone
> wants to make their own preconditioner, they probably need to know the
> data structure: PETSc or uBlas.
>
> So I just put the enum list straight into the common file
> Preconditioner.h that all can use. 

Great. This is what I had in mind.

This simplifies the usage a bit:
>     KrylovSolver solver(amg);
>
> No need for prefixing with some namespace.

This is nicer than before.

> Will do the same thing with FooKrylovSolver::Type --> KrylovMethod.

OK.

Garth


> /Anders
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev