fenics team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

On Wed, Dec 14, 2011 at 11:50:04AM +0000, Garth N. Wells wrote:
> On 13 December 2011 18:57, Anders Logg <logg@xxxxxxxxx> wrote:
> > On Tue, Dec 13, 2011 at 10:47:10AM +0100, Martin Sandve Alnæs wrote:
> >> Hi,
> >> UFC has a series/branch 2.0.x. Just to be clear, this
> >> should be maintained to match the 1.0.x branches of
> >> dolfin/ffc/ufl, corresponding to the book, right?
> >
> > Yes, that was the plan. I tried to make it as uniform as possible,
> > with the 2.0.x branch correspondingt to the other 1.0.x branches.
> >
> >> And the next breaking change will be e.g. ufc 2.1.x matching dolfin
> >> 1.1.x.
> >
> > I'm not sure. I have been thinking about UFC 3 as the next big change
> > of UFC. So it might depend on which changes we make. Say that we
> > fix the currently broken quadrature version of tabulate_tensor for
> > interior_facet_integral and maybe modify some functions to take
> > physical coordinates rather than reference coordinates (or the
> > opposite...) then maybe thas is 2.1. But if we make some bigger
> > changes, then we may jump to 3.0.
> >
>
> I think that we need some time for 3.0. It would be good to minimise
> the number of released that have substantial changes, so we need to
> think about a lot of things before 3.0.

Agree.

> I'm happy for the fixes to tabulate_tensor, real vs mapped
> coordinates, etc to become 2.1.

ok.

--
Anders


> Garth
>
>
> >
> > _______________________________________________
> > Mailing list: https://launchpad.net/~fenics
> > Post to     : fenics@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~fenics
> > More help   : https://help.launchpad.net/ListHelp