[Paraview] couple of issues related with co-processing ...

[u.utku.turuncoglu at be.itu.edu.tr]

Hi,

You can download the Catalyst script from following link,

https://www.dropbox.com/s/eajs9dm0of5b4bi/script_data.tar.gz?dl=0

I also included the data and ParaView state that is used to create the
Catalyst script. So, it might help to find the problem.

Thanks,

--ufuk

> I haven't seen issues like that. Can you share the Catalyst script that
> fails?
>
> On Fri, Apr 15, 2016 at 5:02 AM, Ufuk Utku Turuncoglu (BE) <
> u.utku.turuncoglu at be.itu.edu.tr> wrote:
>
>> Hi Andy,
>>
>> As you suggested in your previous mail, i used allinputsgridwriter.py
>> script to write all the available fields as multiblock dataset and i
>> created co-processing scripts using this dataset. Now, i could handle
>> multiple 3d fields but i have still problem in the combination of
>> contour
>> and vector visualization. In this case, the code hangs in same way that
>> i
>> mentioned in my initial mail.
>>
>> So, to find the source of the problem i tried different pipelines and
>> check the co-processing,
>>
>> 1 - A pipeline with only Glyph filter. In this case, i set Gradients as
>> Vectors in Active Attributes section in the properties window. The
>> co-processing is working without any problem in this case.
>>
>> 2 - I used Calculator Filter to compute vector using u*iHat+v*jHat
>> equation along with Glyph to produce vector field visualization. In this
>> case, the pipeline is working without any error under co-processing.
>>
>> 3 - A visualization pipeline with an iso-surface which is colored by
>> different variable. It works also.
>>
>> 4 - The combination of step 2 and 3. In this case, i used same
>> visualization pipeline with 2 but in this case, i add a Contour filter
>> to
>> create an iso-surface for a scalar 3d field (same as step 3). The
>> co-processing code hands with white screen and does not produce the
>> output.
>>
>> So, it seems that the combination of Contour and vector visualization
>> affects the co-processing. Anyway, do you have similar experience in
>> co-processing?
>>
>> Regards,
>>
>> --ufuk
>>
>> On 13/04/16 16:24, Andy Bauer wrote:
>>
>> Hi Ufuk,
>>
>> I would suggest testing out some of the Catalyst examples which are in
>> the
>> source code under the Examples/Catalyst subdirectory. I would also
>> suggest
>> creating Catalyst scripts from data sets generated through Catalyst
>> output
>> instead of the NetCDF files. There are two example scripts for this in
>> Examples/Catalyst/SampleScripts that you can use for this.
>>
>> As for the grid names for the axes, this may be a bug in ParaView. I'll
>> look into that.
>>
>> Best,
>> Andy
>>
>> On Wed, Apr 13, 2016 at 8:44 AM, Ufuk Utku Turuncoglu (BE) <
>> <u.utku.turuncoglu at be.itu.edu.tr>u.utku.turuncoglu at be.itu.edu.tr> wrote:
>>
>>> Hi,
>>>
>>> I am using co-processing (Catalyst) and i have couple of issues about
>>> the
>>> behavior of the co-processing module.
>>>
>>> 1 - If i send multiple 2d variable or single 3d variable to the
>>> co-processor, my custom code is working without any problem and
>>> produces
>>> desired output. On the other hand, the code hangs with multiple 3d
>>> variable
>>> (it basically opens a square white window and waits, see the attached
>>> screenshot). I have no idea about the problem because the terminal does
>>> not
>>> produces any output. I could install the ParaView in debug mode and it
>>> might help?
>>>
>>> 2 - The axis names (edited by Edit Axes Grid section in Properties
>>> window) are not written to the produced png file when the visualization
>>> pipeline is driven by co-processing but it looks fine under ParaView.
>>>
>>> 3 - When i trying to use volume rendering (single 3d field) along with
>>> co-processing, the code gives following error
>>>
>>> ERROR: In
>>> /opt/progs/paraview-5.0.0/src/VTK/Common/ExecutionModel/vtkCompositeDataPipeline.cxx,
>>> line 171
>>> vtkPVDataRepresentationPipeline (0x7fea05244010): Can not execute
>>> simple
>>> alorithm without output ports
>>> ...
>>> ...
>>> ERROR: In
>>> /opt/progs/paraview-5.0.0/src/VTK/Filters/Parallel/vtkPKdTree.cxx, line
>>> 296
>>> vtkPKdTree (0x7fea04f42840): (process 0) NumberOfDatasets = 0, cannot
>>> determine volume bounds.
>>>
>>> I could produce iso-surface with same code without any problem. The
>>> volume rendering is also working under ParaView.
>>>
>>> The last question might be related with the grid definition in the
>>> co-processing because i am using single netcdf dataset to create
>>> visualization scripts (python) under ParaView but in the model code
>>> (co-processing) the dataset is defined as MultiBlockDataSet due to the
>>> parallel nature of the simulation code.
>>>
>>> Regards,
>>>
>>> --ufuk
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> <http://www.kitware.com/opensource/opensource.html>
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ParaView Wiki at:
>>> http://paraview.org/Wiki/ParaView
>>>
>>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview
>>>
>>>
>>
>>
>