[petsc-dev] PetscSection

[Matthew Knepley]

On Sat, Nov 10, 2012 at 6:11 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> On Sat, Nov 10, 2012 at 5:03 PM, Matthew Knepley <knepley at gmail.com> wrote:
>>
>> Interpolation. I do not have a good, graph theoretic way to phrase it.
>
>
> So if you know the topology of the n-cell, it implies a structure to its
> cone-closure. (You could store that as a cache of single-cell-rooted
> complex.) To interpolate the mesh, you graft in the single-cell, matching
> the intermediate levels (fast query with a vertex index). I've written some
> code for this, though not in full generality (I didn't explicitly store the
> reference single-cells).

I was being too pessimistic. You can assign an intermediate node to any set
of shared vertices, but I am not sure whether you start from the top or bottom,
and the hard part is preallocation.

  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