dolfin team mailing list archive

["Johan Hoffman" <jhoffman@xxxxxxxxxxx>]

Yes, the assembly will rather simple; we're assembling a scalar. So the
functions will look something like that.

/Johan


> Should the interface be something like
>
>     virtual void eval(real& value, const AffineMap& map) const;
>
> or just
>
>     virtual real eval(const AffineMap& map) const;
>
> since we just want to compute one value on each element and then sum
> the values to obtain the value of the functional?
>
> And there is no need for a FiniteElement?
>
> There are two different options also for the assemble() function that
> we need to add to FEM. Either
>
>     static void assemble(Functional& M, real& value, Mesh& mesh);
>
> or
>     static real assemble(Functional& M, Mesh& mesh);
>
> /Anders
>
> On Wed, Nov 02, 2005 at 11:34:19AM +0100, dolfin@xxxxxxxxxx wrote:
>> Commit from hoffman (2005-11-02 11:34 CET)
>> -------------------
>>
>> Added Functional class in DOLFIN. Working on FFC (format/dolfin.py) to
>> automatically generate this class.
>>
>> + dolfin  src/kernel/form/Functional.cpp       1.1
>> + dolfin  src/kernel/form/dolfin/Functional.h  1.1
>>
>> _______________________________________________
>> DOLFIN-dev mailing list
>> DOLFIN-dev@xxxxxxxxxx
>> http://www.fenics.org/cgi-bin/mailman/listinfo/dolfin-dev
>>
>
> --
> Anders Logg
> Research Assistant Professor
> Toyota Technological Institute at Chicago
> http://www.tti-c.org/logg/
>
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/cgi-bin/mailman/listinfo/dolfin-dev
>


-- 
Johan Hoffman, PhD
Assistant Professor
School of Computer Science and Communication
Royal Institute of Technology KTH
SE-100 44 Stockholm
Sweden

Email: jhoffman@xxxxxxxxxxx
URL: www.nada.kth.se/~jhoffman