ffc team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

On Thu, Jan 28, 2010 at 09:56:18PM +0100, Kristian Oelgaard wrote:
>
>
> On 28 January 2010 21:14, Marie Rognes <meg@xxxxxxxxx> wrote:
> >Kristian Oelgaard wrote:
> >>
> >>
> >>2010/1/27 Medhi Nikbakht <m.nikbakht@xxxxxxxxxx>:
> >>>
> >>>On Wed, 2010-01-27 at 15:50 +0100, Marie Rognes wrote:
> >>>>
> >>>>FFC compilation fails for
> >>>>
> >>>>   ElementRestriction.ufl
> >>>
> >>>The error message says that "ElementRestriction" is not supported by
> >>>FIAT.
> >>>How will the restricted elements be implemented? Wouldn't it be better
> >>>to support them inside FFC?
> >>
> >>There are two types of restrictions supported in UFL (and the old FFC).
> >>1) Restriction onto Measure (like 'dc'), which is what you want.
> >>2) Restriction onto Cell, as for instance restricting a Lagrange element
> >>defined on a 'triangle' to its facets.
> >>
> >>The third restriction that we might want in the future is restriction onto
> >>subdomains possible in UFL but not in FFC.
> >>
> >>To implement 1) I think it is enough to just attach the Measure as a
> >>member to the fiat_element and the simply look for it in appropriate places
> >>when generating the code. That means handle it in FFC.fiatinterface, could
> >>that work?
> >>
> >>To implement 2) I modified the return value of entity_dofs() and
> >>tabulate() accordingly. To do it this way again we will need to make changes
> >>in FIAT_NEW.FiniteElement and FFC.MixedElement. I'm not sure it is a good
> >>idea to make FIAT_NEW.PolynomialSet aware of restrictions, but I don't know
> >>enough about the FIAT internals to back up this claim.
> >>
> >
> >
> >How about adding a class RestrictedElement that contains an element and
> >(based on the restriction)
> >a list of relevant basis function indices, and overrides the methods as
> >FiniteElement/MixedElement?
>
> Yes, that would be one option and we already have MixedElement and RestrictedElement is much the same thing (we should change the name in UFL). Although, one objective of this re-write of FFC was to remove as much as possible w.r.t. FiniteElement.
>
> We could make the RestrictedElement a child class of FIAT.FiniteElement, that should be OK since FFC uses FIAT as the backend for finite elements. Then we just need to modify a few member functions.

I think Marie's suggestion is cleaner with a the RestrictedElement
class (or whatever it will be named) *has* an element. That element
could perhaps be another mixed element. There are not that many
functions to implement (about 8).

--
Anders

Attachment: signature.asc
Description: Digital signature