[Paraview] [Xdmf] unexpected usage of ghost sets for geometry and topology with XDMF reader

[Utkarsh Ayachit]

Chris,

I was able to locate the conversation/patch you are referring to. If
there's a sample dataset that you can share that I can use for
testing, I'd like to add a dashboard test for this so that we don't
break this functionality in the future.

Utkarsh

On Tue, Jul 5, 2011 at 4:31 PM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Chris,
>
> Do you any additional info about the patches that you are referring
> to? Did they make it into Xdmf repo, but just haven't made it into
> ParaView yet?
>
> Utkarsh
>
> On Tue, Jul 5, 2011 at 4:09 PM, Chris Kees <cekees at gmail.com> wrote:
>> Hi,
>>
>> I'm using XDMF for storing results of parallel unstructured finite
>> element computations. I worked with some of the developers for the
>> XDMF reader on getting proper display of the overlapping (ghost)
>> entities a while back, but it doesn't seem like those changes were
>> really merged into the trunk. Paraview 3.10 still doesn't appear to
>> eliminate duplicate points and cells quite the way I would expect it
>> to. Is there anybody still working on that functionality that could
>> help me get some changes permanently made to paraview? I can supply
>> data sets of various complexity (quad,tri,hex,tet, (bi-)quadratic,
>> dynamic, etc) in a dropbox folder and am am willing to change the way
>> we're organizing the information in XDMF (though if my memory serves,
>> we had a patched paraview that worked fine with our current
>> representation).
>>
>> Thanks,
>> Chris
>> _______________________________________________
>> Xdmf mailing list
>> Xdmf at lists.kitware.com
>> http://www.kitware.com/cgi-bin/mailman/listinfo/xdmf
>>
>