ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2024-03-25T14:07:36-04:00https://gitlab.kitware.com/paraview/paraview/-/issues/22294Spreadsheet view crashes remote server2024-03-25T14:07:36-04:00W. Alan ScottSpreadsheet view crashes remote serverSpreadsheet view is crashing remote server after slice. Here is how to replicate.
* Master (v5.11.1-1679-g a2b48e7dbb), Linux, remote server (16 ranks)
* Open g1s1. Apply.
* Slice. Apply.
* Split Horizontal. Spreadsheet.
Crash
@co...Spreadsheet view is crashing remote server after slice. Here is how to replicate.
* Master (v5.11.1-1679-g a2b48e7dbb), Linux, remote server (16 ranks)
* Open g1s1. Apply.
* Slice. Apply.
* Split Horizontal. Spreadsheet.
Crash
@cory.quammen @spiros.tsalikis
If you guys can find this for 5.12.0, great. OK to slip to 5.12.1. Must be fixed for 5.12.1.5.12.1 (Spring 2024)Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/21057IOSS Writer: Exodus2023-06-08T10:34:20-04:00Utkarsh AyachitIOSS Writer: ExodusThis issue documents requirements for the new IOSS writer to generate Exodus datasets. Our objective is to ensure that if someone opened an Exodus file and saved it out again as Exodus via this new writer, the input and output files shou...This issue documents requirements for the new IOSS writer to generate Exodus datasets. Our objective is to ensure that if someone opened an Exodus file and saved it out again as Exodus via this new writer, the input and output files should largely be identical. The difficulty arises from the fact that any data operations in ParaView may make that difficult. Here we document some challenges and then potential solutions.
* File-per-rank: initially, we follow the file-per-rank approach. IOSS is not MPI-aware in VTK/ParaView yet and hence using this approach is the easiest.
* Consolidated node-block: with Exodus, there's only 1 node block. If global node ids are present, then it's easy. We can generate a global node-block with ease. However, if they are not present, we can't simply generate them since that may cause point-fields to be clobbered. In that case, we will create a single nodeblock with duplicated points.
* set-blocks: sets rely on global element/node ids together with special array called "element_side". If required arrays are present, they will be saved appropriately as sets. Otherwise, we simply treat them as element blocks.
* blocks with mixed element types: exodus blocks only have 1 element type. However, vtkUnstructuredGrid supports mixed elements. To handle that case, the writer will have to globally determine if its necessary to split an element-block into multiple. Note, since blocknames need to be unique, we'll need to come up with some strategy when splitting a block.5.12 (Winter 2023)Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/19799Simplify the output of the trace recorder2023-01-03T12:42:17-05:00W. Alan ScottSimplify the output of the trace recorderPlease simplify the output of the trace recorder. Currently, it sets all kinds of state that is already being set by ParaView and PVBatch. Stop doing this.
When done, I expect the trace to be as simple as possible, yet create correct ...Please simplify the output of the trace recorder. Currently, it sets all kinds of state that is already being set by ParaView and PVBatch. Stop doing this.
When done, I expect the trace to be as simple as possible, yet create correct traces to be used in Macros, Python Shell/ Run Script and PVBatch.
This should be the default. Note that "give me everything" and "just trace the filters" needs to still work.
OK to slip to 5.10.0.5.12 (Winter 2023)Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/19677Watney Master Issues for 5.9.02022-09-21T13:12:56-04:00W. Alan ScottWatney Master Issues for 5.9.0Here are the list of Watney issues that are targetting 5.9.0. These are generally speaking enhancing ParaView/ Catalyst's ability to do quantitative analysis. This is onging development.
Milestone 5.8.0 Watney bugs are here: https://...Here are the list of Watney issues that are targetting 5.9.0. These are generally speaking enhancing ParaView/ Catalyst's ability to do quantitative analysis. This is onging development.
Milestone 5.8.0 Watney bugs are here: https://gitlab.kitware.com/paraview/paraview/issues/19126.
At Watney's request, I am reordering them by functional group.
* Selection. This is a limiting factor in everything we are doing for getting numbers out of ParaView, and highly coupled to the next item (Internal) because selection is how we identify objects.
- [x] #19676 Extend Extract Cells by Region to include line or splice source - Important
- [x] #19406 Sources/ Ellipse misses one segment - Important
- [ ] #19649 Select cells through in parallel does not trace correctly
* Internal to ParaView. How are objects labeled and referenced? Objects are cells and points, but also objects that ParaVew and VTK define or create such as streamlines. How do I find indices, what do they mean, and how can I use them to apply selection and filters to the desired objects?
- [x] #19128 cgns reader block ids messed up - Required
- [ ] #19170 Document all of the block and id names - Required (should slip post 5.8 release)
- [x] #19661 Append Location Attributes filter crashes in Catalyst with muliblock dataset where leaves are structured - On Sandia's plate.
- [x] #19117 CGNS bad in parallel with D3 filter - Important (probable slip to 5.9)
- [ ] #19234 Clarifications and modifications to Global Ids - Important
* Reading and Writing from/to files. For now, the mechanism is writing csv files through either File/SaveData or Export on the Spreadsheet view, but a much more expansive interface should be developed.
- [x] #19073 Save data products on the client, or server, side - Nice to have
- [x] #19454 add cgns reader globbing for batch
- [x] #20664 optionally add automatic displacements in CGNS
* Scripting issues. Critical for doing anything in Catalyst, and selections are not implemented in scripting so we can’t even get started on this area.
* Other, self explanatory
- [ ] #19161 StreamTracer path length seems suspect - Important
* Build support. This includes Catalyst bugs and features, and SuperBuild support.
- [ ] Superbuild - https://gitlab.kitware.com/paraview/paraview-superbuild/issues/142 Add ability to compile LLVM static. - Important
- [x] #19669 CMAKE_INSTALL_PREFIX not propogated properly in Superbuild.
- [x] #19565 External use of hdf5 not easy to implement, and doesn't work.5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/19454add cgns reader globbing for batch2022-08-17T20:45:08-04:00W. Alan Scottadd cgns reader globbing for batchWhen trying to write a generic python trace that uses cgns, especially for BIG datasets (say over 1000 files), it is hard to make the file names all generic (such as fileDirectory+fileName+'16384.00000', etc.). It would be really, reall...When trying to write a generic python trace that uses cgns, especially for BIG datasets (say over 1000 files), it is hard to make the file names all generic (such as fileDirectory+fileName+'16384.00000', etc.). It would be really, really nice if either:
* We could just load the 0'th file, and have the cgns reader realize that this is a multi-file dataset (as happens with Exodus) or
* Have a commented out block of python code that will do it. Something like for (i = 0; i < 16384; i++ {string += directoryName+fileName+'16384.'Number} Then, CGNSReader(string) (Sorry for the funky pseudocode, I think it shows my thoughts.
NOTE - Lets implement file globbing, as recommended by and explained below by Ken.
Although this will probably be used for Watney, I don't think it's needed now. Catalyst won't use it, and we can always hard code the python manually.5.11 (Spring 2022)Yohann Bearzi (Kitware)Yohann Bearzi (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/21147Add time step option in csv file writer2022-04-28T10:00:37-04:00W. Alan ScottAdd time step option in csv file writerWhen writing a .csv file, we need to optionally add a timestep column. This will allow the .csv file (either as a single file or joined individual files) to be sorted on this timestep column. Just like Add Time, but an integer. Defaul...When writing a .csv file, we need to optionally add a timestep column. This will allow the .csv file (either as a single file or joined individual files) to be sorted on this timestep column. Just like Add Time, but an integer. Default off.5.11 (Spring 2022)Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/21148Allow .csv writer to create a single file with timesteps2022-04-28T09:57:17-04:00W. Alan ScottAllow .csv writer to create a single file with timestepsPlease add the option in the .csv writer to write all timesteps to a single .csv file. Even though the file could grow large, this would make it trivial to load this file into a post processing tool and step through time. Make this an ...Please add the option in the .csv writer to write all timesteps to a single .csv file. Even though the file could grow large, this would make it trivial to load this file into a post processing tool and step through time. Make this an option, default on. You get to this through File/ Save Data.
A good test dataset is g1s1. Be sure to test in parallel.5.11 (Spring 2022)Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/19676Extend Extract Cells by Region to include a line or spline source.2022-04-27T22:36:27-04:00W. Alan ScottExtend Extract Cells by Region to include a line or spline source.Please add to Extract Cells by Region the ability to extract a line or spline source.Please add to Extract Cells by Region the ability to extract a line or spline source.5.11 (Spring 2022)Yohann Bearzi (Kitware)Yohann Bearzi (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/19126Watney master issues for Milestone 5.82021-10-20T21:51:53-04:00W. Alan ScottWatney master issues for Milestone 5.8Here are the list of Watney issues that are required for the milestone. These are generally speaking enhancing ParaView/ Catalyst's ability to do quantitative analysis.
At Watney's request, I am reordering them by functional group.
...Here are the list of Watney issues that are required for the milestone. These are generally speaking enhancing ParaView/ Catalyst's ability to do quantitative analysis.
At Watney's request, I am reordering them by functional group.
* Selection. This is a limiting factor in everything we are doing for getting numbers out of ParaView, and highly coupled to the next item (Internal) because selection is how we identify objects.
- [x] #19400 Selections, and Extract Selection, don't trace correctly
- [x] #19187 Add select along a mesh plane or line for flat structured meshes - Required
- [x] #19068 Add Select by source - Required (OK to slip post 5.8.0)
- [x] #18534 Add the line controls to Spline Source - Required
- [x] #19171 Find Data, Remote server, fails - Required
- [x] #19173 Expand Selections - Required
- [x] #19491 Trace of select cells through is broken - Required
- [x] #19278 Writing .csv file is crashing, remote server - Required
- [x] #19464 Add line source along normal from a node on a surface - Required (OK to slip post 5.8.0)
- [x] #19233 Add circle source - Important
- [x] #19176 Add ability for all 3d widgets to be moved in X, Y or Z direction independently - Important
- [x] #19236 Line source: add multiple ways of generating sample points - Important
- [x] #19388 CGNS, and ExtractSelectionWithSeed, is wireframe - Important
- [x] #19166 Adding direction indications to Spline and PolyLine widgets - Nice to have
* Internal to ParaView. How are objects labeled and referenced? Objects are cells and points, but also objects that ParaVew and VTK define or create such as streamlines. How do I find indices, what do they mean, and how can I use them to apply selection and filters to the desired objects?
- [x] #19115 For data analysis, streamlines must be labeled - Required
- [x] #19228 Please add a Global [node,element] Id generator filter - Required
- [x] #19225 Point Ids and Cell Ids are broken with CGNS - required
- [x] #19226 Point IDs are misnamed with CGNS - required
- [x] #19391 ExtractSubsetWithSeed Crashes on CGNS Dataset - Required
- [x] #19639 Append Location Attributes crashes ParaView - Required
- [x] #19603 Crash with box clip
- [x] #19277 GenerateGlobalId crashes in parallel - Important
* Writing numbers to files. For now, the mechanism is writing csv files through either File/SaveData or Export on the Spreadsheet view, but a much more expansive interface should be developed.
- [x] #19224 Save Data, csv file, from streamtracer is bad. Required. Showstopper for 5.7.0 release.
- [x] #19072 Save Data as csv, add option to save time - Required
- [x] #19169 Add Generate Cell Centers filter - Required
- [x] #19431 Save Data/ csv exporter is broken for row data - Required
- [x] #19441 Crash saving Exodus data after extract selection, remote server - Required
- [x] #19658 Vtm writer failing with structured data in parallel
- [x] #19054 Add spreadsheet functionality to save data (aka select arrays to write) - Important
* Scripting issues. Critical for doing anything in Catalyst, and selections are not implemented in scripting so we can’t even get started on this area.
- [x] #15396 Python trace or Python state doesn't capture selections. - Required
- [x] #16836 Trace recorder misses Find Data selections - Required
- [x] #19172 Python GetActive doesn't return node and element ids - Required
- [x] #19435 Trace of find data finds wrong cell - Required
- [x] #19403 Trace Recorder missing Views - Required
- [x] #19573 Wrong case for names in Catalyst scripts - Required
* Other, self explanatory
- [x] #19138 cgns outline mode blank remote server - Required
- [x] #19146 Streamlines filter uses inconsistent seed ids in parallel - Required
- [x] #19110 Plot Data of multiblock dataset is broken - Important
- [x] #19108 Plot Data Marker Style, add point - Important
* Build support. This includes Catalyst bugs and features, and SuperBuild support.
- [x] #19448 Implement Catalyst Essentials - Required
- [x] #19450 Look at Catalyst Essentials for memory footprint reductions - Required
- [x] #19451 Package ParaView Python files in a single zip file - Required
- [x] #19460 static builds of paraview search for plugin shared libs on startup - Required
- [x] #19486 Python fails with static ParaView - Required
- [x] #19564 Need ability to turn hdf5 off with Catalyst - Required
- [x] #19563 Boost fails build on powerpc - Required
- [x] #19494 Compile failure building mpi2py with OpenMPI 4 - Requiredhttps://gitlab.kitware.com/paraview/paraview/-/issues/19805extract block filter should work by block name as well as index for catalyst2021-08-25T16:09:19-04:00Jeff Mauldinextract block filter should work by block name as well as index for catalystWhen you save an "extract block" filter in a catalyst script, the selected blocks for extraction are saved by an index (which appears to be generated as/during a depth-first traverse of the multiblock datastructure).
Unless the data to ...When you save an "extract block" filter in a catalyst script, the selected blocks for extraction are saved by an index (which appears to be generated as/during a depth-first traverse of the multiblock datastructure).
Unless the data to which the catalyst script is applied has an identical block structure, the extract block filter will fail. User expectations are such that if they specify an extract block filter by clicking the names of the blocks in the gui, the extract block filter will extract the blocks with those names regardless of the block structure which the catalyst script is applied to.
In my own python scripts I have gotten around this by doing the search on the block structure myself in python and finding the required indexes that way before creating the extract block filter. However, I think the appropriate interface for the filter would be to specify included/exclued blocks by block name, not by index. I realize that in some cases blocks might not have names, but we could either allow both methods of specifying block filters, or we could set a default name of a string equal to the block index in cases where no name has been supplied.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19128cgns reader block ids messed up2021-06-22T15:19:59-04:00W. Alan Scottcgns reader block ids messed upThe CGNS reader, and the hover cells on presents different block ids. As with Exodus, they should be consistent, repeatable, linear and contiguous. In other words, for a single block, I expect to find the same number throughout, and ha...The CGNS reader, and the hover cells on presents different block ids. As with Exodus, they should be consistent, repeatable, linear and contiguous. In other words, for a single block, I expect to find the same number throughout, and have it either increment 0,1,2,3,4 or use the numbering scheme from the file (blk-2, blk-1, etc.)
* Linux, master, remote server (16 processes).
* Open the Witney cgns dataset. Turn all blocks off, except have blk-2, Grid on. I turned on all variables, but it won't matter. Apply. You now have one block visible.
* Split View Vertically. Spreadsheet. Turn visibility on if needed. Sort on Block Number. Move back and forth, and you will find block numbers 8 and 9. Sort on block number. Now, you will see block numbers from 11 to 15!
These are bugs.
I'm fairly certain this will be a Watney bug.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19669CMAKE_INSTALL_PREFIX not propogated properly in superbuild2021-05-17T19:42:10-04:00Jeff MauldinCMAKE_INSTALL_PREFIX not propogated properly in superbuildI am trying to build catalyst for linking against SPARC. When I set the CMAKE_INSTALL_PREFIX argument in the superbuild, that command is not propagated down to paraview (or, for that matter, python if built, osmesa, and maybe not llvm)....I am trying to build catalyst for linking against SPARC. When I set the CMAKE_INSTALL_PREFIX argument in the superbuild, that command is not propagated down to paraview (or, for that matter, python if built, osmesa, and maybe not llvm).
cmake runs find, and the build happens, but all the results are placed in the default install directory (parallel to the 'superbuild' directory) in my building directory.
When I look at the CMakeCache.txt file in .../superbuild/paraview/build I see that the CMAKE_INSTALL_PREFIX is NOT set to what I set CMAKE_INSTALL_PREFIX to in the CMakeCache.txt for the overall superbuild.
I believe I can overcome this for paraview by passing cmake arguments down into paraview, but I'm not sure I can fix it for osmesa, python, and llvm. It does look like the llvm libraries get installed into the install lib directory as desired, but no paraview directories, and there is no <install>/lib/cmake subdirectory which is supposed to contain the paraview cmake config file for SPARC to use to find paraview during it's cmake configuration.5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/19873ExtractSubsetWithSeed missing blocks with pvbatch2020-09-30T19:05:35-04:00W. Alan ScottExtractSubsetWithSeed missing blocks with pvbatchJeff wrapped this one up really well in a directory that I can give Kitware. Lets call it ExtractSubsetWithSeed missing blocks with pvbatch.
Here is what Jeff wrote in the README:
The contents of this directory demonstrate an error wi...Jeff wrapped this one up really well in a directory that I can give Kitware. Lets call it ExtractSubsetWithSeed missing blocks with pvbatch.
Here is what Jeff wrote in the README:
The contents of this directory demonstrate an error with the
ExtractSubsetWithSeed filter in pvbatch/catalyst. In the gui operation,
ExtractSubsetWithSeed successfully follows a path from block to block.
In the pvbatch operation, we only get data from one block. This error occurs
even on single process operation. To see this bug in operation (or for
debugging):
In the gui, open the cgns data set, including cell data. Make an
ExtractSubsetWithSeed filter with seed point [1.0, 0.08685, 0.08685] and
direction 'JK'. Render by surface, color by block and examine.
On the command line, execute the following line (make sure you have access to
pvbatch for paraview 5.8):
pvbatch --symmetric cgns2catalyst.py sparc-volume-1process.cgns -ps do_essws_script_from_gui.py
or
pvbatch cgns2catalyst.py sparc-volume-1process.cgns -ps do_essws_script_from_gui.py
And look at the output image.
(Snip second bug that has not been replicated)
Sending the tarball to Kitware.5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19565External use of hdf5 not easy to implement, and doesn't work.2020-09-30T17:50:45-04:00W. Alan ScottExternal use of hdf5 not easy to implement, and doesn't work.External use of hdf5 not easy to implement, and doesn't work. Here is from an e-mail with @jamauld. This is a followon to bug #19564:
Next, I tried turning VTK_MODULE_ENABLE_VTK_hdf5 back on, but then using
VTK_MODULE_USE_EXTERNAL_...External use of hdf5 not easy to implement, and doesn't work. Here is from an e-mail with @jamauld. This is a followon to bug #19564:
Next, I tried turning VTK_MODULE_ENABLE_VTK_hdf5 back on, but then using
VTK_MODULE_USE_EXTERNAL_VTK_hdf5:BOOL=ON
This successfully found SPARC's hdf5 library and used that for configure.
But then my build failed. When linking the pvserver executable (and a couple others, like maybe pvdataserver), I got undefined symbol errors. The undefined symbols were like H5Create (or H5PCreate or something), and they were undefined in the library
libhdf5_hl.a
when I messed around with the nm command, I discovered that the symbols were undefined in the libhdf5_hl.a library, but they were defined in the libhdf5.a library.
When I looked at the link.txt file for pvserver, I saw that both the libhdf5.a and libhdf5_hl.a were in the link line, but libhdf5.a came first. I added libhdf5.a again after libhdf5_hl.a. Then pvserver linked. Clearly if libhdf5 was shared the order wouldn't have mattered. I don't want to add all symbols for libhdf5.a because that might bloat the binary.
What we need here is a successful compile ordering of the static hdf5 libraries in the link.txt file. Also, it would probably help to be able to specify using system hdf5 libraries up at the superbuild cmake level as well.
Getting either #19564 or this bug fixed is a crisis. We can take either, depending on which is easier, but would prefer getting #19564 fixed first. Setting milestone and labels appropriately.5.9 (Fall 2020)https://gitlab.kitware.com/paraview/paraview/-/issues/19073Save data products on the client, or server, side2020-09-28T20:36:38-04:00W. Alan ScottSave data products on the client, or server, sidePlease add the ability to save data products on the client, or the server side. Default as we have now. This is important for .csv files, where a user is trying to do some data analysis, and wants her data to be local. Alternatively, ...Please add the ability to save data products on the client, or the server side. Default as we have now. This is important for .csv files, where a user is trying to do some data analysis, and wants her data to be local. Alternatively, we have cases where screen shots and movies need to be used on the remote side, and have to be manually copied.
Not sure how important this is at this time. Requested by Watney.5.9 (Fall 2020)https://gitlab.kitware.com/paraview/paraview/-/issues/19403Trace recorder is missing views2020-09-16T17:58:28-04:00W. Alan ScottTrace recorder is missing viewsThe trace recorder is missing creating a view or layout. Here is how to replicate:
* Linux, 5.7.0, builtin server.
* Tools/ Start Trace.
* Sources/ Wavelet. Apply.
* Plot over line filter. Apply.
* File/ Save Screenshot. Save All Vie...The trace recorder is missing creating a view or layout. Here is how to replicate:
* Linux, 5.7.0, builtin server.
* Tools/ Start Trace.
* Sources/ Wavelet. Apply.
* Plot over line filter. Apply.
* File/ Save Screenshot. Save All Views - checked. OK.
* Tools/ Stop Trace. Save this python script.
Now, run this python script using pvbatch. It will fail, as follows:
( 3.216s) [main thread ]vtkSMSaveScreenshotProx:408 ERR| vtkSMSaveScreenshotProxy (0x5612fb0): Cannot WriteImage without a view or layout.
Watney bug.
Marking as required, but ok to move to Inportant if there is a manual way to edit the scripts correctly.5.8 (Winter 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19658Vtm writer failing with structured data in parallel2020-08-31T12:51:00-04:00W. Alan ScottVtm writer failing with structured data in parallelThe vtk/vtm writer appears to be failing with structured data in parallel. Here is how to replicate:
* Linux, 5.8.0-RC2, multiple rank - remote server (I'm on 8 ranks).
* Open Watney's HiFiRE CGNS dataset, sparc-volume.cgns.144.*. Only...The vtk/vtm writer appears to be failing with structured data in parallel. Here is how to replicate:
* Linux, 5.8.0-RC2, multiple rank - remote server (I'm on 8 ranks).
* Open Watney's HiFiRE CGNS dataset, sparc-volume.cgns.144.*. Only load Blk-3, Grid (only for load time reduction). I turned all vars on. Apply.
* File/ Save Data. Vtk Multi Block Files (vtm). I'm calling it deleteMeC. OK, Take defaults, OK.
* Reset session.
* Try to open this dataset. I am trying to open the file deleteMeC.vtm. This should open the directory.
Note that this error does not occur if the writer and reader is builtin server. It DOES occur if the writer is parallel, and the reader is builtin.
Error message, which is totally insufficient, is:
```
ERROR: In /projects/viz/paraview/src/ParaView/5.8.0-RC2/ParaView5/VTK/IO/XML/vtkXMLMultiBlockDataReader.cxx, line 300
vtkXMLMultiBlockDataReader (0xa997bb0): Syntax error in file.
```5.8 (Winter 2020)https://gitlab.kitware.com/paraview/paraview/-/issues/19639Append Location Attributes crashes paraview with this exodus-dumped slice2020-08-25T21:30:30-04:00Jeff MauldinAppend Location Attributes crashes paraview with this exodus-dumped sliceI created this exodus-format slice by writing exodus out from SPARC. I can load the slice in paraview and color-by-variable just fine. But when I try to apply the filter "Append Location Attributes" paraview crashes. The slice is from G...I created this exodus-format slice by writing exodus out from SPARC. I can load the slice in paraview and color-by-variable just fine. But when I try to apply the filter "Append Location Attributes" paraview crashes. The slice is from Greg's UUR dataset. I was running SPARC with 36 processes, so there are 36 files.[data1.zip](/uploads/f107196186dde61b428b8c095f88c8ba/data1.zip)
This is against paraview 5.8 RC1. I am running on Windows. Greg Weirs has been getting a crash when running catalyst (with SPARC) with a pipeline that includes an Append Location Attributes, so that could be related. He does _not_ get this crash when running paraview himself on cgns volue data dumped from SPARC.5.8 (Winter 2020)Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/19117CGNS bad in parallel with the D3 filter2020-06-29T21:22:23-04:00W. Alan ScottCGNS bad in parallel with the D3 filterCGNS data is not being handled properly with the D3 filter in parallel. Here is how to replicate:
* Master (v5.6.1-1676-gc5bc267), remote server (I am using 16 ranks), Linux.
* Open sparc.cgns.4. Ask me for this dataset if needed. De...CGNS data is not being handled properly with the D3 filter in parallel. Here is how to replicate:
* Master (v5.6.1-1676-gc5bc267), remote server (I am using 16 ranks), Linux.
* Open sparc.cgns.4. Ask me for this dataset if needed. Default variables and blocks. Ignore FlowSolutionPointers on. Apply.
* D3. Apply. (This should be surface.)
Notice there are holes in the inside of the curve. D3 appears to be drawing triangles wrong.
This is not holding anything up, but seems important since it is CGNS.
This is a Watney bug.5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19406Sources/ Ellipse misses one segment2020-06-16T12:21:04-04:00W. Alan ScottSources/ Ellipse misses one segmentSources/ Ellipse misses one segment. Here is how to replicate:
* Linux, master, builtin server.
* Sources/ Ellipse. Resolution 4. Start Angle 0. Sector Angle 90. Apply. I will call this ellipse-0.
* Sources/ Ellipse. Resolution 4...Sources/ Ellipse misses one segment. Here is how to replicate:
* Linux, master, builtin server.
* Sources/ Ellipse. Resolution 4. Start Angle 0. Sector Angle 90. Apply. I will call this ellipse-0.
* Sources/ Ellipse. Resolution 4. Start Angle 90. Sector Angle 90. Apply. I will call this ellipse-90.
These two partial ellipses should touch, but don't.
* Change ellipse-0 and ellipse-90 to be non-Closed.
This is what I would expect. Now, they touch correctly. Close and non-Close should look exactly the same, only one has a closing line.
This is a Watney bug.5.9 (Fall 2020)Vicente Boleavicente.bolea@kitware.comVicente Boleavicente.bolea@kitware.com