[petsc-dev] PetscSection

[Matthew Knepley]

On Sat, Nov 10, 2012 at 3:35 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> On Sat, Nov 10, 2012 at 2:28 PM, Matthew Knepley <knepley at gmail.com> wrote:
>>
>> > point that is a member. What about if a given level happens to be empty?
>>
>> DMComplexGetLabelIdIS(). Its not problem if a level is empty.
>
>
> Ah, I misinterpreted what that routine did. Aren't these trivially short
> arrays (max length of 4 for meshes in 3D) that are expected to be identical
> on every process?

Not identical on every process.

> How does one get an IS containing _only_ the labeled points?

DMComplexGetStratumIS()

>>
>> > Can you answer how you can distinguish a quad from a tet in a
>> > non-interpolated mesh?
>>
>> You can't, but that is the point. You are not supposed to distinguish
>> them.
>
>
> Hmm, so these are the same at the category-theory level, but definitely not
> at the mesh topology level. It doesn't seem right to not distinguish them at
> all. How do you write DMComplexInterpolate for mixed-dim strata?

Interpolation is analysis. You need to know what dimension you are in, etc.

   Matt

--
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which
their experiments lead.
-- Norbert Wiener