I've been working in a Python script to produce runtime post-processing in my simulation code. <br><br>The idea is quite simple:<br><br>Step 0). Pre-simulation:<br><br>1. Launch paraview and create a pipeline with the desired visualization (viewpoint, lights, slices, streamlines, whatever...). Note that it can be easily done using small models;<br>
<br>2. Save the pipeline as a ParaView State file (it will be used as a template for the runtime postprocessing script --> see the next step)<br><br>Step 1). Running the simulation code:<br><br>1. Launch the simulation and, while the application is runnig (or after finishing), launch the ParaView Python Script passing some arguments, such as, filename for the "template" paraview state file, email (yes, the script can send you screenshots by email).<br>
<br>What does this script do?<br><br>a. Look for results already saved by the application producing a screenshot for each time step found. Note that I still have to have the results saved in some file format. In my tests I used Ensight.<br>
<br>b. After creating the screenshots for the results already produced by the simulation tool, the script starts to "listen" for changes in the application directory. As soon as it detects a new time step, it creates the corresponding screenshot.<br>
<br>Remarks:<br><br>I. The script is able to send screenshots as email attach (it behaves like a spam in some cases :-)<br><br>II. The script is lightweight since it only needs to load the Python stuff once (when launched);<br>
<br>III. It's transparent to the simulation code since it runs as an independent process.<br><br>IV. You don't need to compile or link any library to the existing application.<br><br>V. The script can be used at any time. It's not restrict to run at simulation's runtime<br>
<br>I have not finished the script yet (I'm just a novice in Python), but I was planning to send it to the mail list in the future. <br><br><br><font color="#888888">Renato N. Elias<br>===================================<br>
High Performance Computing Center (NACAD)<br>Federal University of Rio de Janeiro (UFRJ)<br>Rio de Janeiro, Brazil<br>
</font><br><br><div class="gmail_quote">On Wed, Sep 2, 2009 at 3:25 PM, Moreland, Kenneth <span dir="ltr"><<a href="mailto:kmorel@sandia.gov">kmorel@sandia.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div>
<font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">I think Dave means the in-situ sections in the upcoming Vis09 tutorial (not the SC09 tutorial).<br>
<br>
<a href="http://www.paraview.org/Wiki/IEEE_Vis09_ParaView_Tutorial" target="_blank">http://www.paraview.org/Wiki/IEEE_Vis09_ParaView_Tutorial</a><br>
<br>
-Ken<div><div></div><div class="h5"><br>
<br>
<br>
On 9/2/09 11:04 AM, "David E DeMarle" <<a href="http://dave.demarle@kitware.com" target="_blank">dave.demarle@kitware.com</a>> wrote:<br>
<br>
</div></div></span></font><div><div></div><div class="h5"><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">On Wed, Sep 2, 2009 at 9:33 AM, Randy Heiland <<a href="http://heiland@indiana.edu" target="_blank">heiland@indiana.edu</a>> wrote:<br>
</span></font><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">Hello,<br>
<br>
Just wanting to get a better understanding of the latest ParaView, while considering adopting it for a project. I'm assuming ParaView does its UI using C++ calls to Qt, as opposed to using PyQt, is that correct? (if any of the developers would care to comment on the design decisions behind that, I'd welcome that too). And that to <br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"> <br>
Yes C++, not PyQt.<br>
<br>
</span></font><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">create a custom UI, one does so via XML? Any changes on the horizon?<br>
<br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">Sort of. To customize the client one uses a mixture of C++ and XML and designer built widgets. The exact mix depends on what you are trying to change. See the paraview wiki page, specifically the plugins page for examples of how to make changes and the kinds of changes it is easy to make today.<br>
<br>
It is also possible to write entirely new applications. See the OverView and StreamingParaView applications in the source code for examples.<br>
<br>
Yes, changes are on the horizon. Writing a new app is too complicated right now because the design of the client is fairly monolithic and interconnected. Search the mailing list for Utkarsh's posting about branded applications. He has a git repository that is publicly available on git-hub where he is working on that.<br>
<br>
</span></font><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">Are there examples uses of ParaView for real-time vis? Any examples that do computational steering? Not necessarily for parallel apps, just serial.<br>
<br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"> <br>
Several people have done this before, so far no one has done it in a robust enough manner to be widely used. The upcoming tutorial at SuperComputing (again the see wiki page) convers perhaps the best approach so far. It is also worth investigating XDMF's DSM layer which is another good approach. Finally, I personally did it in yet another way. My approach was to add an asynchronous polling loop to the client, in the same way that the streaming paraview app renders for pieces in a loop while the UI stays responsive..<br>
<br>
</span></font><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">thanks, Randy<br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"> <br>
np.<br>
<br>
<br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"><br>
</span></font></div></div><font size="1"><font face="Monaco, Courier New"><span style="font-size: 7.5pt;"><br>
**** Kenneth Moreland<br>
*** Sandia National Laboratories<br>
*********** <br>
*** *** *** email: <a href="http://kmorel@sandia.gov" target="_blank">kmorel@sandia.gov</a><br>
** *** ** phone: (505) 844-8919<br>
*** web: <a href="http://www.cs.unm.edu/%7Ekmorel" target="_blank">http://www.cs.unm.edu/~kmorel</a><br>
</span></font></font><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"><br>
</span></font>
</div>
<br>_______________________________________________<br>
Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
<br>
Visit other Kitware open-source projects at <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
<br>
Please keep messages on-topic and check the ParaView Wiki at: <a href="http://paraview.org/Wiki/ParaView" target="_blank">http://paraview.org/Wiki/ParaView</a><br>
<br>
Follow this link to subscribe/unsubscribe:<br>
<a href="http://www.paraview.org/mailman/listinfo/paraview" target="_blank">http://www.paraview.org/mailman/listinfo/paraview</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br>Renato N. Elias<br>===================================<br>High Performance Computing Center (NACAD)<br>Federal University of Rio de Janeiro (UFRJ)<br>Rio de Janeiro, Brazil<br>
<br>Sent from Rio De Janeiro, RJ, Brazil