<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Fri, May 26, 2017 at 12:27 PM, Fabian.Jakub <span dir="ltr"><<a href="mailto:Fabian.Jakub@physik.uni-muenchen.de" target="_blank">Fabian.Jakub@physik.uni-muenchen.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear Petsc Team,<br>
<br>
I am playing around with DMPlex, using it to generate the Mesh for the<br>
ICON weather model(<a href="http://doi.org/10.1002/2015MS000431" rel="noreferrer" target="_blank">http://doi.org/10.1002/<wbr>2015MS000431</a>), which employs a<br>
triangle mesh horizontally and columns, vertically.<br>
<br>
This results in a grid, looking like prisms, where top and bottom faces<br>
are triangles and side faces are rectangles.<br>
<br>
I was delighted to see that I could export the triangle DMPlex (2d Mesh)<br>
to hdf5 and use petsc_gen_xdmf.py to then visualize the mesh in<br>
visit/paraview.<br>
This is especially nice when exporting petscsections/vectors directly to<br>
VTK.<br></blockquote><div><br></div><div>Great.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
I then tried the same approach for the prism grid in 3D.<br>
I attached the code for one single cell, as well as the output in hdf5.<br>
<br>
However, trying to convert the hdf5 output, it fails with:<br>
<br>
make prism.xmf<br>
<br>
$PETSC_DIR/bin/petsc_gen_xdmf.<wbr>py prism.h5<br>
Traceback (most recent call last):<br>
File<br>
"/software/meteo/xenial/x86_<wbr>64/petsc/master/debug_gcc/..//<wbr>bin/petsc_gen_xdmf.py",<br>
line 241, in <module><br>
generateXdmf(f)<br>
File<br>
"/software/meteo/xenial/x86_<wbr>64/petsc/master/debug_gcc/..//<wbr>bin/petsc_gen_xdmf.py",<br>
line 235, in generateXdmf<br>
Xdmf(xdmfFilename).write(<wbr>hdfFilename, topoPath, numCells,<br>
numCorners, cellDim, geomPath, numVertices, spaceDim, time, vfields,<br>
cfields)<br>
File<br>
"/software/meteo/xenial/x86_<wbr>64/petsc/master/debug_gcc/..//<wbr>bin/petsc_gen_xdmf.py",<br>
line 193, in write<br>
self.writeSpaceGridHeader(fp, numCells, numCorners, cellDim, spaceDim)<br>
File<br>
"/software/meteo/xenial/x86_<wbr>64/petsc/master/debug_gcc/..//<wbr>bin/petsc_gen_xdmf.py",<br>
line 75, in writeSpaceGridHeader<br>
''' % (self.cellMap[cellDim][<wbr>numCorners], numCells, "XYZ" if<br>
spaceDim > 2 else "XY"))<br>
KeyError: 6<br>
<br>
<br>
Also, if I try to export a vector directly to vtk, visit and paraview<br>
fail to open it.<br>
<br>
My question is:<br>
Is this a general limitation of these output formats, that I can not mix<br>
faces with 3 and 4 vertices or is it a limitation of the<br>
petsc_gen_xdmf.py or the VTK Viewer.<br></blockquote><div><br></div><div>petsc_gen_xdmf. Take a look here</div><div><br></div><div> <a href="https://bitbucket.org/petsc/petsc/src/1731673c3fe570066779d46b51a4aee7a45775ed/bin/petsc_gen_xdmf.py?at=master&fileviewer=file-view-default#petsc_gen_xdmf.py-9">https://bitbucket.org/petsc/petsc/src/1731673c3fe570066779d46b51a4aee7a45775ed/bin/petsc_gen_xdmf.py?at=master&fileviewer=file-view-default#petsc_gen_xdmf.py-9</a></div><div><br></div><div>This is what fails. You need to add something like</div><div><br></div><div> 6: "Wedge"</div><div><br></div><div>in the dictionary. See <a href="http://www.xdmf.org/index.php/XDMF_Model_and_Format">http://www.xdmf.org/index.php/XDMF_Model_and_Format</a></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
I'd also welcome any thoughts on the prism mesh in general.<br>
Is it that uncommon to use and do you foresee other complications with it?<br></blockquote><div><br></div><div>You need an element that works with prisms, but it seems you already have one. I know</div><div>there is good work from here: <a href="https://arxiv.org/abs/1411.2940">https://arxiv.org/abs/1411.2940</a></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
I fear I cannot change the discretization of the host model but maybe it<br>
makes sense to use a different grid for my radiative transfer code?<br></blockquote><div><br></div><div>I do not really do RT, but would be happy to try and think about it.</div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Many thanks,<br>
<br>
<br>
Fabian<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div><div><br></div><div><a href="http://www.caam.rice.edu/~mk51/" target="_blank">http://www.caam.rice.edu/~mk51/</a><br></div></div></div>
</div></div>