ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2022-03-22T04:55:08-04:00https://gitlab.kitware.com/paraview/paraview/-/issues/21312SpreadSheetView shifts Ids after sorting, creating duplicates and deletions2022-03-22T04:55:08-04:00David TucholskiSpreadSheetView shifts Ids after sorting, creating duplicates and deletionsSteps to reproduce:
- create new disk source
- in advanced properties set radial resolution to 100, circumferential resulution to 1000
- open in spreadsheet view
- sort by descending pointID
- point with ID 9864 is duplicated, point...Steps to reproduce:
- create new disk source
- in advanced properties set radial resolution to 100, circumferential resulution to 1000
- open in spreadsheet view
- sort by descending pointID
- point with ID 9864 is duplicated, point with id 8840 is missing.
- sort by ascending point ID
- point with ID 91135 is duplicated, point with id 92159 is not in spreadsheet anymore.
(92159-91135=1024 and 9864-8840=1024, if that helps to find the cause)
Tested with versions 5.9.1, 5.10 (fresh install) on my windows laptop, 5.4.1 on a Ubuntu VM hosted on said windows laptop, and 5.8 on a remote windows desktop.https://gitlab.kitware.com/paraview/paraview/-/issues/21311vtkCPPythonPipeline::DetectScriptVersion() often doesn't work for Catalyst v1...2023-11-08T21:18:28-05:00Miren RadiavtkCPPythonPipeline::DetectScriptVersion() often doesn't work for Catalyst v1.0 scripts without explicit headersI recently generated a Catalyst script with ParaView v5.8.1 for which the start of the file looks like:
```python
#--------------------------------------------------------------
# Global timestep output options
timeStepToStartOutputAt=...I recently generated a Catalyst script with ParaView v5.8.1 for which the start of the file looks like:
```python
#--------------------------------------------------------------
# Global timestep output options
timeStepToStartOutputAt=0
forceOutputAtFirstCall=False
# Global screenshot output options
imageFileNamePadding=6
rescale_lookuptable=False
# Whether or not to request specific arrays from the adaptor.
requestSpecificArrays=True
# a root directory under which all Catalyst output goes
rootDirectory=''
# makes a cinema D index table
make_cinema_table=False
#--------------------------------------------------------------
# Code generated from cpstate.py to create the CoProcessor.
# paraview version 5.8.1
#--------------------------------------------------------------
from paraview.simple import *
from paraview import coprocessing
```
It can be deduced from the line
```python
# paraview version 5.8.1
```
that this is a v1.0 script but `vtkCPPythonPipeline::DetectScriptVersion()` returns `0` (i.e. failure) since it only keeps checking additional lines if they are comment lines or empty (see [this part of the definition](https://gitlab.kitware.com/paraview/paraview/-/blob/a20839cf4fbd0d7fc5c727f00cbe5973b3741f22/Clients/PythonCatalyst/vtkCPPythonPipeline.cxx#L137-L185)). In the particular case above, the final `else` block is selected and the following is printed to the log
```
( 1.918s) [pvbatch.0 ]vtkCPPythonPipeline.cxx:182 9| . regex match failed for 'timeStepToStartOutputAt=0'
```
Of course, I could edit this script by hand and add a line such as
```python
# script version: 1.0
```
near the top but this doesn't seem necessary given it would be possible to get a `regex1` match further down the file. Could this function be modified to keep checking after a line with no regex matches?https://gitlab.kitware.com/paraview/paraview/-/issues/21309Unexpected behaviors when building and wrapping a ParaView plugin2022-03-17T03:30:19-04:00Timothee ChabatUnexpected behaviors when building and wrapping a ParaView pluginWhen building a VTK filter in a ParaView plugin, and if there is a compile-related error in a header of the plugin (ie a missing `;` or something like that), I won't have a GCC error but instead this not-so-clear message :
```
vtkWrapH...When building a VTK filter in a ParaView plugin, and if there is a compile-related error in a header of the plugin (ie a missing `;` or something like that), I won't have a GCC error but instead this not-so-clear message :
```
vtkWrapHierarchy-pv5.10: In /home/timothee/Dev/Plugin/MyVTKModule/vtkMyView.h:24: syntax error.
make[2]: *** [Plugin/MyVTKModule/CMakeFiles/MyVTKModule-hierarchy.dir/build.make:79: lib/paraview-5.10/plugins/Plugin/vtk/hierarchy/MyPlugin/MyVTKModule-hierarchy.txt] Error 1
make[1]: *** [CMakeFiles/Makefile2:312: Plugin/MyVTKModule/CMakeFiles/MyVTKModule-hierarchy.dir/all] Error 2
make: *** [Makefile:146: all] Error 2
```
Also if the new filter class `A` is a subclass of `B`, and `B` is a ParaView class of the module `ParaView::C`, and if `ParaView::C` is in the `PRIVATE_DEPENDS` section of the `vtk.module`, the plugin compile just fine (unexpected) but the generated CS wrapping is empty and thus not working (expected since configuration is wrong, but we shouldn't have gone there in the first place).
Tested against ParaView masterhttps://gitlab.kitware.com/paraview/paraview/-/issues/21308Create a filter that takes an AMR dataset and refines/coarsens it to a specif...2022-05-18T07:09:34-04:00Andrew BauerCreate a filter that takes an AMR dataset and refines/coarsens it to a specified levelThe VTK AMR datasets (vtkOverlappingAMR and vtkNonOverlappingAMR) are still somewhat problematic to deal with in general settings. To get around this it would be good to have a filter that produced a composite dataset or vtkImageData (no...The VTK AMR datasets (vtkOverlappingAMR and vtkNonOverlappingAMR) are still somewhat problematic to deal with in general settings. To get around this it would be good to have a filter that produced a composite dataset or vtkImageData (not sure which would be better at this point) for a specified AMR level. This may already be available for vtkOverlappingAMR but doesn't appear to be available for vtkNonOverlappingAMR.Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/21307Create filter to combine composite dataset of vtkImageDatas into a single vtk...2023-03-02T16:18:15-05:00Andrew BauerCreate filter to combine composite dataset of vtkImageDatas into a single vtkImageDataWhile composite datasets have becoming quite a bit easier to work with, they still aren't as functional as datasets that derive from vtkDataSet. If all vtkImageDatas in a composite dataset have the same `Origin` and `Spacing`, it would b...While composite datasets have becoming quite a bit easier to work with, they still aren't as functional as datasets that derive from vtkDataSet. If all vtkImageDatas in a composite dataset have the same `Origin` and `Spacing`, it would be nice to have a filter that would combine them into a single vtkImageData. Many readers produce such composite datasets, or would produce them after they are output from the filter in https://gitlab.kitware.com/paraview/paraview/-/issues/21285https://gitlab.kitware.com/paraview/paraview/-/issues/21306PLYWriter test fails if directory name has period character2023-05-31T16:31:14-04:00Jaswant Panchumarti (Kitware)PLYWriter test fails if directory name has period characterIf the build directory name has `.` character, this test fails.If the build directory name has `.` character, this test fails.https://gitlab.kitware.com/paraview/paraview/-/issues/21303CDIReader in CI2023-11-15T10:21:35-05:00Mathieu Westphal (Kitware)CDIReader in CICDIReader plugin is enabled only on `fedora35_shared_mpi_python_qt` CI.
We may add it in other builds, specially other OS (win, macos). It requires to add [`libcdi`](https://gitlab.dkrz.de/mpim-sw/libcdi) version 2.0.3 as external depen...CDIReader plugin is enabled only on `fedora35_shared_mpi_python_qt` CI.
We may add it in other builds, specially other OS (win, macos). It requires to add [`libcdi`](https://gitlab.dkrz.de/mpim-sw/libcdi) version 2.0.3 as external dependency to those OS.Nicolas VuailleNicolas Vuaillehttps://gitlab.kitware.com/paraview/paraview/-/issues/21302Too many clean filters in ParaView2022-03-15T11:27:17-04:00Mathieu Westphal (Kitware)Too many clean filters in ParaViewThere is too many clean filters in ParaView
- Clean: PD input, merge duplicate points, remove duplicate cells and fix degenerate cells
- CleanToGrid: UG input, merge duplicate points
- CleanCellsToGrid: UG input, needs CleanToGrid befor...There is too many clean filters in ParaView
- Clean: PD input, merge duplicate points, remove duplicate cells and fix degenerate cells
- CleanToGrid: UG input, merge duplicate points
- CleanCellsToGrid: UG input, needs CleanToGrid before, remove duplicate cells and fix degenerate cells
A single filter for any input with unified options would be much betterhttps://gitlab.kitware.com/paraview/paraview/-/issues/21301Warning prints from IOSS reader2024-03-07T20:23:49-05:00W. Alan ScottWarning prints from IOSS readerI have a dataset that is creating warning prints from the IOSS reader. These do not show up in the legacy exodus reader. Needs to be fixed.
* 5.10.0 (and master), Linux, builtin server.
* Load plate_mismatch_components.e. Apply.
Not...I have a dataset that is creating warning prints from the IOSS reader. These do not show up in the legacy exodus reader. Needs to be fixed.
* 5.10.0 (and master), Linux, builtin server.
* Load plate_mismatch_components.e. Apply.
Notice there were lots of warnings. Needs to be fixed or turned off.
(481.821s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '5' != '16'.
(481.859s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '16' != '5'.
(481.872s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '5' != '16'.
(481.872s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '16' != '5'.
(482.507s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '5' != '16'.
(482.542s) [paraview]vtkPVArrayInformation.c:489 WARN| mismatching in number of components for array 'vonm'. '16' != '5'.
plate_mismatch_components.e is UUR.
[plate_mismatch_components.tar.gz](/uploads/bb4dda426aad9debd59a95cd592b05d3/plate_mismatch_components.tar.gz)5.13 (Summer 2024)https://gitlab.kitware.com/paraview/paraview/-/issues/21299Remote rendering for Windows does not allow spaces in plink.exe path2022-03-14T17:32:31-04:00Stan HammonRemote rendering for Windows does not allow spaces in plink.exe pathStandard Windows installation of `plink.exe` (part of PuTTY) is placed in `C:\Program Files`, where the space between "Program" and "Files" causes problems for `ParaView` when making a remote connection.
This problem was observed in `Pa...Standard Windows installation of `plink.exe` (part of PuTTY) is placed in `C:\Program Files`, where the space between "Program" and "Files" causes problems for `ParaView` when making a remote connection.
This problem was observed in `ParaView 5.10.0`
These are the settings used while attempting the remote connection:
![NoQuotes](/uploads/123975940cc6eabe73aaedd3002a0285/NoQuotes.png)
...resulting in this error message:
![NoQuotes-Error](/uploads/32399adc3851319c6371bc1bd3986c6d/NoQuotes-Error.png)
To fix this first error, quotes were added around the `plink.exe` path:
![Quotes](/uploads/8927651cd4045a4b1b6e17ffa68f3c1b/Quotes.png)
...but this leads to another error relating to the command line arguments being used:
![Quotes-Error](/uploads/47952d269da2eca1b2f887f7d80a468c/Quotes-Error.png)
As it stands, this problem can be fixed by using a non-standard installation path for `plink.exe` (without any spaces), but it would also be good to fix this aspect of ParaView so that any installation path will work on Windows.https://gitlab.kitware.com/paraview/paraview/-/issues/21294GlobalIds with 2D image data causes weird rendering using texture2022-03-15T07:55:04-04:00Mathieu Westphal (Kitware)GlobalIds with 2D image data causes weird rendering using textureDisplaying a 2D image data with GlobalIDs causes rendering issue related to texture.
Steps to reproduce:
- Wavelet
- XExtent: 0 0
- Apply
- GenerateGlobalIds, Apply
- Color by GlobalPoint/CellIds
- Unexected rendering and errors:
```
(...Displaying a 2D image data with GlobalIDs causes rendering issue related to texture.
Steps to reproduce:
- Wavelet
- XExtent: 0 0
- Apply
- GenerateGlobalIds, Apply
- Color by GlobalPoint/CellIds
- Unexected rendering and errors:
```
( 170.646s) [paraview ] vtkOpenGLTexture.cxx:206 ERR| vtkOpenGLTexture (0x1f2554a0): No scalar values found for texture input!
```
![image](/uploads/ac82740d042e5d52ef8add4641cd7b49/image.png)
FYI @timothee.chabat @ken-martinhttps://gitlab.kitware.com/paraview/paraview/-/issues/21284Paraview 5.10 immediately crashes when loading NVIDIA IndeX plugin remotely2022-03-15T15:00:14-04:00Ben WibkingParaview 5.10 immediately crashes when loading NVIDIA IndeX plugin remotelyUsing Paraview 5.10 on macOS, I connect to a remote pvserver (on Linux) via a localhost ssh tunnel to the remote cluster node. I then try to load the pvNVIDIAIndeX plugin from the ‘remote plugins’ list, click ‘load selected’, and the cli...Using Paraview 5.10 on macOS, I connect to a remote pvserver (on Linux) via a localhost ssh tunnel to the remote cluster node. I then try to load the pvNVIDIAIndeX plugin from the ‘remote plugins’ list, click ‘load selected’, and the client immediately crashes.
Loading the plugin from the ‘local plugins’ list first, then loading the remote plugin, does not cause a crash.
Discourse thread: https://discourse.paraview.org/t/paraview-5-10-immediately-crashes-when-loading-nvidia-index-plugin/9082https://gitlab.kitware.com/paraview/paraview/-/issues/21281Show All Blocks is missing again on right click menus2024-03-07T20:23:49-05:00W. Alan ScottShow All Blocks is missing again on right click menusRight click menus are sometimes missing the Show All Blocks. I don't totally understand when it is there, and when it is missing.
@utkarsh.ayachit has replicated this bug on his Linux box.
FOR ME, the following works and doesn't work....Right click menus are sometimes missing the Show All Blocks. I don't totally understand when it is there, and when it is missing.
@utkarsh.ayachit has replicated this bug on his Linux box.
FOR ME, the following works and doesn't work. For my customer, Show All Blocks is totally missing.
* Linux, master (v5.10.0-688-g6c833594a1), builtin server.
* can.exo. Apply.
* Right click on the background. Show All Blocks will exist on this menu.
* ssh -Y yourLocalLinuxBox
* run the version of ParaView above in mesa mode. (you may not need to do the ssh above - it just forces my scripts to software render.)
* can.exo. Apply.
* Right click on the background. Show All Blocks will NOT exist on this menu.
This is a bug that is impacting an important customer.5.13 (Summer 2024)Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/21279New particle checkpoint format in AMReX 22.032024-03-07T20:23:49-05:00Nathan X. MarshakNew particle checkpoint format in AMReX 22.03The [AMRex 22.03 release](https://github.com/AMReX-Codes/amrex/releases/tag/22.03) introduces an updated particle checkpoint format, which was merged as part of [AMReX PR #2624](https://github.com/AMReX-Codes/amrex/pull/2624). This chang...The [AMRex 22.03 release](https://github.com/AMReX-Codes/amrex/releases/tag/22.03) introduces an updated particle checkpoint format, which was merged as part of [AMReX PR #2624](https://github.com/AMReX-Codes/amrex/pull/2624). This changes the version string from `"Version_Two_Dot_Zero"` to `"Version_Two_Dot_One"` in [AMReX_ParticleContainerBase.cpp](https://github.com/AMReX-Codes/amrex/blob/7e73c1ed1c142b83afce7bcaec731d2dc334fbe3/Src/Particle/AMReX_ParticleContainerBase.cpp#L107).
With ParaView nightly build `5.10.0-704-g4e23154bc8-MPI-Linux-Python3.9-x86_64`, when I attempt to open an AMReX plotfile with the "AMReX/BoxLib Particles Reader" I get the error message below.
```( 82.318s) [paraview ]vtkAMReXParticlesReader:438 ERR| vtkAMReXParticlesReader (0x7fecfc008600): Bad version string: Version_Two_Dot_One_double
( 82.318s) [paraview ]vtkAMReXParticlesReader:438 ERR| vtkAMReXParticlesReader (0x7fecfc008600): Bad version string: Version_Two_Dot_One_double
( 82.319s) [paraview ] vtkExecutive.cxx:741 ERR| vtkPVCompositeDataPipeline (0x1645c8d0): Algorithm vtkAMReXParticlesReader (0x7fecfc008600) returned failure for request: vtkInformation (0x192d5180)
Debug: Off
Modified Time: 395879
Reference Count: 1
Registered Events: (none)
Request: REQUEST_INFORMATION
ALGORITHM_AFTER_FORWARD: 1
FORWARD_DIRECTION: 0
```
Example AMReX data (with particles) that produces the above output attached here:
[plt00.zip](/uploads/0dd35a5cacf42e063d5c0b1bf5a732e5/plt00.zip)
**Workaround**: I can change the version string by hand in the AMReX output, and that allows me to load my data in ParaView, however I am not using the [new "expanded particle ids" functionality](https://github.com/AMReX-Codes/amrex/pull/2624). I would expect the new functionality to break in ParaView.
ParaView discourse thread: https://discourse.paraview.org/t/particle-checkpoint-format-in-amrex-22-03-not-yet-supported/91045.13 (Summer 2024)https://gitlab.kitware.com/paraview/paraview/-/issues/21278Scatter plots are broken2022-03-03T16:17:23-05:00W. Alan ScottScatter plots are brokenScatter plots are broken with 5.10.0 for Exodus datasets. Note that merge blocks fixes the issue. Here is how to replicate.
* 5.10.0 (and Master), MacOS, builtin server.
* Can.exo. Apply.
* Scatter Plot filter. Apply.
This fails with...Scatter plots are broken with 5.10.0 for Exodus datasets. Note that merge blocks fixes the issue. Here is how to replicate.
* 5.10.0 (and Master), MacOS, builtin server.
* Can.exo. Apply.
* Scatter Plot filter. Apply.
This fails with lots of noisy output.
* after can, run Merge Blocks filter. Then, Scatter Plot filter.
This works.https://gitlab.kitware.com/paraview/paraview/-/issues/21276Gradient Filter HTG Compatibility2022-03-02T11:16:26-05:00Mathieu Westphal (Kitware)Gradient Filter HTG CompatibilityThe gradient filter is not compatible with HTG, it would be nice if it was.
Curently in developement.The gradient filter is not compatible with HTG, it would be nice if it was.
Curently in developement.Charles GueunetCharles Gueunethttps://gitlab.kitware.com/paraview/paraview/-/issues/21274PlotOverLine HTG Compatibility2022-03-02T11:16:30-05:00Mathieu Westphal (Kitware)PlotOverLine HTG CompatibilityThe PlotOverLine filter is not compatible with HTG, it would be nice if it was.
Currently in developpement.The PlotOverLine filter is not compatible with HTG, it would be nice if it was.
Currently in developpement.Charles GueunetCharles Gueunethttps://gitlab.kitware.com/paraview/paraview/-/issues/21273Trace not sensitive to modified Glyph properties2022-03-02T07:29:54-05:00baloeTrace not sensitive to modified Glyph propertiesHere is a small example which will generate an arrow glyph.
```python
pointSource1 = PointSource(registrationName='PointSource1')
glyph1 = Glyph(registrationName='Glyph1', Input=pointSource1, GlyphType='Arrow')
```
One can modify the gl...Here is a small example which will generate an arrow glyph.
```python
pointSource1 = PointSource(registrationName='PointSource1')
glyph1 = Glyph(registrationName='Glyph1', Input=pointSource1, GlyphType='Arrow')
```
One can modify the glyph properties via `GlyphType`, e.g.
```python
glyph1.GlyphType.TipResolution = 32
```
However, these modifications are not caught by the trace.https://gitlab.kitware.com/paraview/paraview/-/issues/21270Skybox and Palette do not have expected behavior2023-10-27T05:31:05-04:00Mathieu Westphal (Kitware)Skybox and Palette do not have expected behaviorUsing a skybox then the palette do not have expected behavior.
Steps to reproduce:
1. Open ParaView
2. Uncheck "Use color palette for background"
3. Set background to skybox
4. Load a skybox
5. Check "Use color palette for background" ...Using a skybox then the palette do not have expected behavior.
Steps to reproduce:
1. Open ParaView
2. Uncheck "Use color palette for background"
3. Set background to skybox
4. Load a skybox
5. Check "Use color palette for background" -> **Skybox stays visible, unexpected**
Proably linked to https://gitlab.kitware.com/paraview/paraview/-/issues/20889 and https://discourse.paraview.org/t/background-vs-path-traced-environment/7402/9https://gitlab.kitware.com/paraview/paraview/-/issues/21269When loading state files, use actual timesteps2024-03-07T20:23:49-05:00W. Alan ScottWhen loading state files, use actual timestepsPlease add the option, default on, so when a state file is loaded, the number of timesteps comes from the dataset being opened, rather than the state file. State files are frequently used for opening a dataset multiple times that is in ...Please add the option, default on, so when a state file is loaded, the number of timesteps comes from the dataset being opened, rather than the state file. State files are frequently used for opening a dataset multiple times that is in the process of being written.5.13 (Summer 2024)