ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2021-06-23T19:29:56-04:00https://gitlab.kitware.com/paraview/paraview/-/issues/20745IOSS: For CGNS, ApplyDisplacements does not work2021-06-23T19:29:56-04:00Utkarsh AyachitIOSS: For CGNS, ApplyDisplacements does not workWhen opening a CGNS dataset with Ioss reader, Apply Displacements currently has no effect. Awaiting a test dataset to understand how to implement support for this.
cc: @wascottWhen opening a CGNS dataset with Ioss reader, Apply Displacements currently has no effect. Awaiting a test dataset to understand how to implement support for this.
cc: @wascott5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20739Ensight reader crashes when using 16 procs2021-06-24T19:14:12-04:00Boonthanome NouanesengsyEnsight reader crashes when using 16 procsThe Ensight reader in Paraview will crash if more than a certain number of processors are used. For the test dataset, it seems that the data will load as long as the number of procs is less than 16. But if 16 procs are used, then there i...The Ensight reader in Paraview will crash if more than a certain number of processors are used. For the test dataset, it seems that the data will load as long as the number of procs is less than 16. But if 16 procs are used, then there is a seg fault when loading the dataset. This is with the latest Paraview master.
If more than 16 procs are used and no cell and point variables are selected to load, then the geometry loads fine. Once you choose a cell or point data to load, it will crash.
If you crash Paraview by first loading the geometry, then trying to load a cell data variable, the following error message appears:
```
( 425.565s) [pvserver.15 ] vtkDataSet.cxx:627 ERR| vtkRectilinearGrid (0x307e750): Cell array density_Air with 1 components, has only -500 tuples but there are 0 cells
( 425.566s) [pvserver.14 ] vtkMPICommunicator.cxx:68 WARN| MPI had an error
```
@cory.quammen @utkarsh.ayachit I will email you the test dataset.
To reproduce:
- Connect Paraview to a server with 16 or more procs
- Attempt to load the test dataset
- Paraview will crash
@patchett20025.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20738ParaView Catalyst blueprint: add support for `params` in execute2021-08-11T21:42:07-04:00Utkarsh AyachitParaView Catalyst blueprint: add support for `params` in executeExtend the [ParaView-Catalyst blueprint](https://kitware.github.io/paraview-docs/nightly/cxx/ParaViewCatalystBlueprint.html) as follows:
protocol: 'execute'
---------
**params:** pass parameters that can change per execute invocation.
...Extend the [ParaView-Catalyst blueprint](https://kitware.github.io/paraview-docs/nightly/cxx/ParaViewCatalystBlueprint.html) as follows:
protocol: 'execute'
---------
**params:** pass parameters that can change per execute invocation.
* params: (optional) a 'list' of strings.
Next, add a new function in the `paraview.catalyst` module that Catalyst scripts can use to access these params. Possible names:
* `paraview.catalyst.get_params()`, or more explicitly
* `paraview.catalyst.get_execute_params()`
for details on how to implement this, see paraview/paraview!4869.5.10 (Fall 2021)Corey Wetterer-Nelson (Kitware)Corey Wetterer-Nelson (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/20737Cell Size filter is broken for Wedges2021-06-22T18:07:29-04:00W. Alan ScottCell Size filter is broken for WedgesI have a dataset that is failing using the Integrate Variables filter. We have a trivial test case, with known expected values. Here is how to replicate it, then an analysis.
* Master or 5.9.1 or 5.9.0, Linux or Mac, builtin server.
*...I have a dataset that is failing using the Integrate Variables filter. We have a trivial test case, with known expected values. Here is how to replicate it, then an analysis.
* Master or 5.9.1 or 5.9.0, Linux or Mac, builtin server.
* Open mesh_mb.g. Apply.
Note that this mesh (cuboid) is 1X2X3 in size. There are 4 blocks within this cuboid.
* Calculator. Formula is 1. Apply.
* Integrate Variables filter. Apply.
In the spreadsheet, note that Result integrates to 2. Switch to Cell. Volume is 2. This is a bug.
Note that if you do the same thing with mesh_sb.g, the integrated Result and volume is 6. This is correct.
* Click on the calculator in the pipeline browser.
* Cell Size filter. Apply.
* Change spreadsheet to Cell Size filter (if it isn't already there). Cells. Now, click on volume column. Sort smallest to largest.
The volumes for Wedges will be negative!
That is at least the start of the issue. Note that when corrected, if the integrated volumes aren't 6, we still have bugs.
I will pass the dataset to Utkarsh and Cory.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20734fonts too big for 5.9.1 Help > Example Visualizations2021-08-24T15:51:30-04:00Dan Lipsafonts too big for 5.9.1 Help > Example VisualizationsThis is happening on Windows, resolution 3840x2400, change size of text, apps and other items: 250%
![image](/uploads/27b9740785b9aa21b8931a6f970722fb/image.png)This is happening on Windows, resolution 3840x2400, change size of text, apps and other items: 250%
![image](/uploads/27b9740785b9aa21b8931a6f970722fb/image.png)5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/20729Add search for default_servers.pvsc on macOS in /Applications/ParaView-x.y.z....2021-06-10T15:31:52-04:00Cory Quammencory.quammen@kitware.comAdd search for default_servers.pvsc on macOS in /Applications/ParaView-x.y.z.app/Contents/MacOS/5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/20722Volume mapper bug on negative values with unstructured grids2021-08-10T16:57:25-04:00Yohann Bearzi (Kitware)Volume mapper bug on negative values with unstructured gridsNegative values when volume rendering an unstructured grid doesn't display correctly.
In the following screenshot, all visualizations should be identical, as the only difference between them is a scalar shift. They all have independent ...Negative values when volume rendering an unstructured grid doesn't display correctly.
In the following screenshot, all visualizations should be identical, as the only difference between them is a scalar shift. They all have independent scalar bars. Top left view has only positive values (and displays correctly), and bottom right view has all values bellow the "white color" that are negative.
![image](/uploads/ea44c9d2c711978ef2c6796dd7576234/image.png)
Here's a state file to get the state of this screenshot: [volume_mapper_bug.pvsm](/uploads/3abcb5e06e605f9dfdf5b7829adca016/volume_mapper_bug.pvsm)
I am using Hyper Tree Grids in this example because they are an easy way to produce an unstructured grids without loading a file.
To reproduce locally:
* Sources -> Data Objects -> Hyper Tree Grid (Random)
* Inside the Properties panel, set Split Fraction to 0.15.
* Filters -> Hyper Tree Grids -> HyperTreeGrid To UnstructuredGrid
* Calculator1 : Depth - 1
* Calculator2 : Depth - 2
* Calculator3 : Depth - 3
* Create 4 views
* View top left : volume render output of filter
* View top right : volume render Calculator1
* View bottom left : volume render Calculator2
* View bottom right : volume render Calculator3
For each view, click `Use Separate Color Map`
For each view, click `Rescale To Data Range`
For each color map, cancel red and white, and amplify blue.
Note that volume visualization work correctly on image data (so shifting the scalar fields of the Wavelet Source so there are negative values will work).5.10 (Fall 2021)Yohann Bearzi (Kitware)Yohann Bearzi (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/20720Spack build of paraview master fails at link step for catalyst_replay2021-05-26T13:03:00-04:00Ethan StamSpack build of paraview master fails at link step for catalyst_replay@utkarsh.ayachit @cory.quammen
Last night our spack builds of `paraview@master` failed at the link step for `catalyst_replay`. The paraview spec was `paraview@master%gcc@7.4.0~advanced_debug~cuda~examples+hdf5~ipo+kits+mpi+opengl2+osme...@utkarsh.ayachit @cory.quammen
Last night our spack builds of `paraview@master` failed at the link step for `catalyst_replay`. The paraview spec was `paraview@master%gcc@7.4.0~advanced_debug~cuda~examples+hdf5~ipo+kits+mpi+opengl2+osmesa+plugins~python+python3~qt+shared build_type=RelWithDebInfo cuda_arch=native arch=linux-rhel7-broadwell`
Output of the error:
```console
[100%] Linking CXX executable ../../../../../bin/catalyst_replay
cd /tmp/stam/spack-stage/spack-stage-paraview-master-ffgrcmqjeawokr4qcx6t2l35ojdz5idx/spack-build-ffgrcmq/ThirdParty/catalyst/vtkcatalyst/catalyst/replay && /yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/opt/spack/linux-rhel7-broadwell/gcc-7.4.0/cmake-3.15.3-2hqjqbsmkqxnz5qqb5dgtdbiflbn5ot2/bin/cmake -E cmake_link_script CMakeFiles/catalyst_replay.dir/link.txt --verbose=1
/yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/lib/spack/env/gcc/g++ -O2 -g -DNDEBUG -Wl,-lc -Wl,-lc CMakeFiles/catalyst_replay.dir/catalyst_replay.cpp.o -o ../../../../../bin/catalyst_replay -Wl,-rpath,/tmp/stam/spack-stage/spack-stage-paraview-master-ffgrcmqjeawokr4qcx6t2l35ojdz5idx/spack-build-ffgrcmq/lib64:/usr/projects/hpcsoft/toss3/common/x86_64/anaconda/5.0.1-python-3.6/lib ../../../../../lib64/libcatalyst.so.2 -Wl,-rpath-link,/usr/projects/hpcsoft/toss3/common/x86_64/anaconda/5.0.1-python-3.6/lib
/yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/opt/spack/linux-rhel7-broadwell/gcc-7.4.0/netcdf-c-4.8.0-tdmo2wgwleqmtsqc3dk2amoz5dhsm6px/lib/libnetcdf.so.19: undefined reference to `H5Pset_all_coll_metadata_ops'
/yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/opt/spack/linux-rhel7-broadwell/gcc-7.4.0/netcdf-c-4.8.0-tdmo2wgwleqmtsqc3dk2amoz5dhsm6px/lib/libnetcdf.so.19: undefined reference to `H5Pset_dxpl_mpio'
/yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/opt/spack/linux-rhel7-broadwell/gcc-7.4.0/netcdf-c-4.8.0-tdmo2wgwleqmtsqc3dk2amoz5dhsm6px/lib/libnetcdf.so.19: undefined reference to `H5Pset_coll_metadata_write'
/yellow/usr/projects/paraview/fusion/nightly-snow/spack/spack/opt/spack/linux-rhel7-broadwell/gcc-7.4.0/netcdf-c-4.8.0-tdmo2wgwleqmtsqc3dk2amoz5dhsm6px/lib/libnetcdf.so.19: undefined reference to `H5Pset_fapl_mpio'
collect2: error: ld returned 1 exit status
make[2]: *** [bin/catalyst_replay] Error 1
```5.10 (Fall 2021)Ben BoeckelBen Boeckelhttps://gitlab.kitware.com/paraview/paraview/-/issues/20713VisItBridge: Nek5000 and PFLOTRAN readers fail to update after point/cell arr...2021-08-24T19:33:57-04:00Utkarsh AyachitVisItBridge: Nek5000 and PFLOTRAN readers fail to update after point/cell array selections are changedThis is causing the test `pvcs.VisItBridgeNek`, `pvcrs.VisItBridgeNek` to fail in MPI builds.
Remove these tests from exclusions once fixed.This is causing the test `pvcs.VisItBridgeNek`, `pvcrs.VisItBridgeNek` to fail in MPI builds.
Remove these tests from exclusions once fixed.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20712Remove Auto MPI feature2021-09-16T20:05:59-04:00Mathieu Westphal (Kitware)Remove Auto MPI featureAs stated here, Auto MPI feature is not working very well or at all in certain cases.
It should be removed:
https://gitlab.kitware.com/paraview/paraview/-/issues/18420
FYI @cory.quammen @wascott @francois.mazenAs stated here, Auto MPI feature is not working very well or at all in certain cases.
It should be removed:
https://gitlab.kitware.com/paraview/paraview/-/issues/18420
FYI @cory.quammen @wascott @francois.mazen5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20699Plot Global Variables over Time selection broken2021-11-11T14:28:27-05:00W. Alan ScottPlot Global Variables over Time selection brokenThe Plot Global Variables over Time filter does not allow blocks of global variables to be selected anymore. This is minor for small datsets, but for datasets with many dozens or hundreds of global variables, this is an issue. This wor...The Plot Global Variables over Time filter does not allow blocks of global variables to be selected anymore. This is minor for small datsets, but for datasets with many dozens or hundreds of global variables, this is an issue. This worked up until 5.8.1. Here is how to replicate:
* 5.9.0, Linux, builtin server.
* Load can.exo. All vars on. Apply.
* Plot Global Variables over Time. Apply.
* In the Pipeline browwser, highlight NSTEPS.
* SHIFT-highlight YMON
* Click on one of the checkboxes, attempting to reverse the selection on all of the ones selected.
The selection is lost. This is a bug.5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/20696Ioss: Add support for reading CGNS files2021-09-03T18:12:34-04:00Utkarsh AyachitIoss: Add support for reading CGNS filesAdd support to the ioss reader to read CGNS filesAdd support to the ioss reader to read CGNS files5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20690Group Datasets for PDC or PD2021-09-23T18:11:43-04:00Utkarsh AyachitGroup Datasets for PDC or PDGroup Datasets filter currently only produces a Multiblock. Let's add support for it to produce paritioned dataset and partitioned dataset collection. Would be great to add support to name the components too.Group Datasets filter currently only produces a Multiblock. Let's add support for it to produce paritioned dataset and partitioned dataset collection. Would be great to add support to name the components too.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20688Loading a multiblock with symmetric mode creates an error2021-05-10T13:54:46-04:00Boonthanome NouanesengsyLoading a multiblock with symmetric mode creates an errorI have a python script that loads a multiblock, applies the Integrate Variables filter, then writes out the resulting data in a csv file. When running the script in symmetric mode, the following error pops up:
```
( 0.591s) [pvbatch ...I have a python script that loads a multiblock, applies the Integrate Variables filter, then writes out the resulting data in a csv file. When running the script in symmetric mode, the following error pops up:
```
( 0.591s) [pvbatch ] vtkPVSessionCore.cxx:372 ERR| vtkPVSessionCore (0x7fb82709c1d0): Object type: vtkXMLMultiBlockDataReader, could not find requested method: "SetActiveTimeDataArrayName"
or the method was called with incorrect arguments.
```
There are more error messages after that. Even though these error messages appear, the csv file does show up, so it doesn't actually seem to affect anything. But these error messages should be looked into.
If I run without the symmetric flag, no error messages appear.
Test dataset: [symmetric.tar.gz](/uploads/ec25b3454d51271e3a1a7f9b4dd1ae56/symmetric.tar.gz)
To reproduce:
- untar the test dataset. there will be a python script and vtm file there.
- mpirun -np 1 pvbatch --symmetric integrate.py
- lots of error messages pop up
This is with the latest master branch of Paraview. Be sure to use the latest master when reproducing, as the script does not work with 5.9.0.
@patchett2002 @cory.quammen5.10 (Fall 2021)Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/20673Error remote server with IOSS reader2021-08-24T17:38:58-04:00W. Alan ScottError remote server with IOSS readerWe have an error print when running the IOSS reader in parallel. Here is how to replicate:
* master, Linux, remote server.
* Open can.exo. Apply.
We see millions of these. They appear to be coming from the server side.
( 67.626s) ...We have an error print when running the IOSS reader in parallel. Here is how to replicate:
* master, Linux, remote server.
* Open can.exo. Apply.
We see millions of these. They appear to be coming from the server side.
( 67.626s) [paraview ] vtkOutputWindow.cxx:76 ERR| ERROR: In /snip/VTK/Common/DataModel/vtkDataObjectTree.cxx, line 381
vtkMultiBlockDataSet (0xb53a600): Structure does not match. You must use CopyStructure before calling this method.
( 67.626s) [paraview ] vtkOutputWindow.cxx:76 ERR| ERROR: In /snip/VTK/Common/DataModel/vtkDataObjectTree.cxx, line 381
vtkMultiBlockDataSet (0xb53c660): Structure does not match. You must use CopyStructure before calling this method.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20672Catalyst state with PNG extract, Attribute CompressionLevel does not exist2021-05-11T10:45:05-04:00Ethan StamCatalyst state with PNG extract, Attribute CompressionLevel does not exist@utkarsh.ayachit @cory.quammen
Found this when fiddling with 5.9.1-RC1 and the image extracts. When saving an image extract and changing the compression level to 0, the resulting catalyst state file cannot be run due to the error `Attr...@utkarsh.ayachit @cory.quammen
Found this when fiddling with 5.9.1-RC1 and the image extracts. When saving an image extract and changing the compression level to 0, the resulting catalyst state file cannot be run due to the error `AttributeError: Attribute CompressionLevel does not exist.`
[wavelet_png_extract.py](/uploads/ad5cc2c6c1d48a2e90ae0b3dc150e21e/wavelet_png_extract.py)
To reproduce:
1. Wavelet source -> Show surface -> color by RTData
2. Create PNG image extract, set compression level to 0
3. File -> Save Catalyst State (`wavelet_png_extract.py`)
4. pvbatch wavelet_png_extract.py
```console
$ /Applications/ParaView-5.9.1-RC1.app/Contents/bin/pvbatch ./wavelet_png_extract.py
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
Traceback (most recent call last):
File "/Applications/ParaView-5.9.1-RC1.app/Contents/Python/paraview/servermanager.py", line 479, in __setattr__
setter = getattr(self.__class__, name)
AttributeError: type object 'PNG' has no attribute 'CompressionLevel'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/Applications/ParaView-5.9.1-RC1.app/Contents/Python/paraview/servermanager.py", line 485, in __setattr__
_bc.setattr(self, name, value)
File "/Applications/ParaView-5.9.1-RC1.app/Contents/Python/paraview/_backwardscompatibilityhelper.py", line 335, in setattr
raise AttributeError()
AttributeError
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "./wavelet_png_extract.py", line 136, in <module>
pNG1.Writer.Format.CompressionLevel = '0'
File "/Applications/ParaView-5.9.1-RC1.app/Contents/Python/paraview/servermanager.py", line 492, in __setattr__
raise AttributeError("Attribute %s does not exist. " % name +
AttributeError: Attribute CompressionLevel does not exist. This class does not allow addition of new attributes to avoid mistakes due to typos. Use add_attribute() if you really want to add this attribute.
```5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20671Ioss: ids arrays2021-08-24T17:12:48-04:00Utkarsh AyachitIoss: ids arraysIoss reader names the ids array same for cells and points. Should they be named differently to avoid confusion? "ids" are what that are called by the library and hence may make sense to preserve that notion, but unclear what users would ...Ioss reader names the ids array same for cells and points. Should they be named differently to avoid confusion? "ids" are what that are called by the library and hence may make sense to preserve that notion, but unclear what users would prefer.5.10 (Fall 2021)https://gitlab.kitware.com/paraview/paraview/-/issues/20664Optionally allow automatic displacement in CGNS2021-06-23T18:57:32-04:00W. Alan ScottOptionally allow automatic displacement in CGNSPlease optionally allow automatic displacements in CGNS. Default off. This feature should act the same as the Exodus reader does with displacements. Please look for the vector "displacements". Note that this is related to #20512Please optionally allow automatic displacements in CGNS. Default off. This feature should act the same as the Exodus reader does with displacements. Please look for the vector "displacements". Note that this is related to #205125.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20660vtkMultiBlockFromTimeSeriesFilter (Group Time Steps) doesn't support PDC2021-08-24T17:48:59-04:00Utkarsh AyachitvtkMultiBlockFromTimeSeriesFilter (Group Time Steps) doesn't support PDCvtkMultiBlockFromTimeSeriesFilter (Group Time Steps) doesn't support PDC. Reenable TestGroupDataFromTimeSeries once fixed.vtkMultiBlockFromTimeSeriesFilter (Group Time Steps) doesn't support PDC. Reenable TestGroupDataFromTimeSeries once fixed.5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20656IOSS: exodus block ids are not produced in the output dataset2021-06-22T19:06:26-04:00Utkarsh AyachitIOSS: exodus block ids are not produced in the output datasetExodus block ids are not produced in the output dataset. Let's add those as "objectids" (or similar)Exodus block ids are not produced in the output dataset. Let's add those as "objectids" (or similar)5.10 (Fall 2021)Utkarsh AyachitUtkarsh Ayachit