View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0003589||ThirdParty||[All Projects] Bug||public||2020-11-09 18:57||2020-11-11 12:00|
|Platform||Unix||OS||Other||OS Version||(please specify)|
|Fixed in Version||8|
|Summary||0003589: Threshold not working on lagrangian data|
When trying to threshold the data from the lagrangian solve it doesn't work. I have to the glyph the data then threshold to get it to work which messes up the data. I don't have this issue when using the v2006. It is also not possible to threshold the data once I have foamToVTk the data.
|Steps To Reproduce||Run the cyclone tutorial can and threshold the data.|
|Additional Information||I did this on windows.|
|Tags||No tags attached.|
It is not clear if you are referring to a problem in OpenFOAM or ParaView. The ParaView threshold functionality does not relate directly to OpenFOAM so it maybe that you are seeing a problem with the ParaView version you are using.
> Run the cyclone tutorial can and threshold the data.
This is not sufficient information to reproduce the problem, could you provide more detail?
I like the changes that were made to the lagrangian solvers so I moved my MPPIC case, that I setup in v2006, to the reactingParticelFoam solver so I can model the heat transfer from the hot gas to a limestone cloud.
When I got my case running and got some of the data lagrangian data I first tried to use paraFoam but that would not load the .openFOAM reader. I then used foamToVTK to get the data into VTK format and then loaded it into ParaView 5.8. applied a threshold filter but it wouldn't actual threshold the data as I want to see only the particles less with an age less than one second. I could do this with the MPPIC lagrangian cloud data that was changed to VTK with foamToVTK from v2006.
I tested to see if it was some setup issue by running the \tutorials\lagrangian\denseParticleFoam\cyclone case to 1.3 seconds and then changing it to VTK with foamToVTK. I would load that into ParaView 5.8 and the threshold filter will not work unless I apply a glyph filter to the data first and then apply the threshold.
I have not tried to see if it works in ubuntu. I will see if I get time tomorrow.
I am also not sure if this is a ParaView or OpenFOAM issue. I know that the two versions save the lagrangian results in different formats and that KitWare only supports the .com version with a builtin reader. But if it doesn't work after changing it to VTK surely it must be OpenFOAM then?
I hope this clears things up.
||There are some serious bugs in ParaView-5.8 so we are using ParaView-5.6.3.|
||I ran the cyclone in OpenFOAM-dev to 1.3s, loaded the Lagrangian data for time 1.3s into ParaView-5.6.3 using the paraFoam reader and thresholded on age from 0 to 0.1s and it works fine.|
||Okay, thanks. I will look into it a bit further on my side to try and get paraFoam to load the reader correctly.|
I looked into a bit further and it seemed to be an issue with the foamToVTk function of the lagrangian data.
I got paraFoam working correctly but removing and purging and the dev and parafoam and reinstalling. Then closing the ubuntu shell and Xming and reopening it. Now paraFoam is working correctly and able to load .openFOAM files and the threshold works. Now I tested the foamToVTK on the complete case of the cyclone tutorial by reconstructing the data and running foamToVTK (just foamToVTK no extra arguments). I open the vtk file after in paraFoam and it can not threshold the data. I then reloaded the case using the PVreader to load the data and exported from paraFoam to a VTM file. This work correctly in 5.8 and parafoam. This to me points to an issue with the foamToVTK.
I cannot reproduce this. The age data (and everything else) is present in the VTK output, and Paraview 5.6.3 and 5.8.1 have no problem loading it and applying a threshold. I do not have Paraview 5.8.0 installed.
If you want us to investigate further, please upload the VTK file in question as well as a Paraview state file showing how you are viewing it (File > Save State).
However, I suspect that this is either an issue of usage or a bug in the version of Paraview that you are using. In either of these cases we will not be able to help. Paraview's bug tracker can be found here: https://gitlab.kitware.com/paraview.
Please see the attached vtk data.
cloud_6000.vtk (1,768,138 bytes)
save_state_threshold.pvsm (263,367 bytes)
||Apologies for the above. I can now see the issue. The problem is that the lagrangian VTK files only have the points listed. They have no VERTICES section, which is needed for them to display by default and for various filters to work. We will add this.|
Resolved in dev and in version 8 by the following commits:
|2020-11-09 18:57||Kyle||New Issue|
|2020-11-09 19:45||henry||Note Added: 0011670|
|2020-11-09 20:25||Kyle||Note Added: 0011671|
|2020-11-09 21:44||henry||Note Added: 0011672|
|2020-11-09 21:58||henry||Note Added: 0011673|
|2020-11-10 04:38||Kyle||Note Added: 0011674|
|2020-11-10 16:45||Kyle||Note Added: 0011675|
|2020-11-11 09:09||will||Note Added: 0011676|
|2020-11-11 10:29||Kyle||File Added: cloud_6000.vtk|
|2020-11-11 10:29||Kyle||File Added: save_state_threshold.pvsm|
|2020-11-11 10:29||Kyle||Note Added: 0011677|
|2020-11-11 10:33||will||Note Added: 0011678|
|2020-11-11 12:00||will||Assigned To||=> will|
|2020-11-11 12:00||will||Status||new => resolved|
|2020-11-11 12:00||will||Resolution||open => fixed|
|2020-11-11 12:00||will||Fixed in Version||=> 8|
|2020-11-11 12:00||will||Note Added: 0011679|