ufl team mailing list archive

[Martin Sandve Alnæs <martinal@xxxxxxxxx>]

On Fri, Feb 13, 2009 at 1:20 PM, Kristian Oelgaard
<k.b.oelgaard@xxxxxxxxxx> wrote:
> Quoting Martin Sandve Alnæs <martinal@xxxxxxxxx>:
>
>> On Fri, Feb 13, 2009 at 11:53 AM, Kristian Oelgaard
>> <k.b.oelgaard@xxxxxxxxxx> wrote:
>> >
>> >
>> > Hello,
>> >
>> > Any thoughts on how to implement a better rule for determining the
>> polynomial
>> > order of a form?
>> >
>> > as it is now the forms:
>> >
>> > a = v*u*dx
>> > a = f*v*u*dx
>> > a = f*f*f*.....*v*u*dx
>> > a = inner(grad(v), grad(u))*dx
>> >
>> > all result in the same order (2 for linear basis functions)
>>
>> What do we want? Add the degrees of all functions and basis functions
>> that are multiplied? Should be possible to do.
>
> Yes, I think it will be enough if we do it like that, taking into account
> derivatives too.
>
>> > Also, what is the strategy for handling the form
>> >
>> > a = (inner(grad(v), grad(u)) + v*u)*dx ?
>> >
>> > will the order of the form be determined by the highest order, or will it
>> be
>> > possible to split the integrand according to order? The latter could get
>> > complicated for forms like:
>> >
>> > a = f*(inner(grad(v), grad(u)) + v*u))*dx
>>
>> If I got it right, lhs and rhs can handle splitting of arbitrary forms by
>> arity,
>> so it should be possible to do the same for terms with different order.
>
> OK, sounds great. I think this is exactly the kind of functionality I was
> looking for.
>
>> > I do notice the # TODO in formdata.py, I'm just curious.
>> >
>> > Kristian
>>
>> Note that I use the convention
>> FIXME: High priority issue
>> TODO: Low priority issue
>
> OK, good to know.
>
>> Since this obviously can be postphoned until after we get things
>> working in general.
>
> Depends how you define 'working in general', I would say getting a correct
> integration of a = f*f*f*f*v*dx is part of 'working in general' :)
>
> Kristian

'working in general' ~ using ufl and dolfin to solve a few pdes
more complicated than projection/poisson (I had those working
for christmas but sfc is currenty broken).

Integration order should be possible to override by form compiler
options anyway, which should be fine as a first step.

Martin