ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2020-11-27T13:09:34-05:00https://gitlab.kitware.com/paraview/paraview/-/issues/19374remove deprecated Catalyst Script Generator GUI and other dead code paths2020-11-27T13:09:34-05:00David E. DeMarleremove deprecated Catalyst Script Generator GUI and other dead code pathsThe Export Inspector isn't new anymore, time to remove its predecessor and clean up some code.
- [x] remove Temporal Script Plugin
- [x] remove Catalyst Script Generator Wizard and Catalyst Writers
- [x] transition tests that use the ol...The Export Inspector isn't new anymore, time to remove its predecessor and clean up some code.
- [x] remove Temporal Script Plugin
- [x] remove Catalyst Script Generator Wizard and Catalyst Writers
- [x] transition tests that use the old GUI
https://gitlab.kitware.com/paraview/paraview/merge_requests/3627
- [x] update the documentation to describe the new GUI instead of the old plugin
https://gitlab.kitware.com/paraview/paraview-guide-ce/commit/d01f71c0d4c477bcfc358e1c46587dcfa48a49a4
https://gitlab.kitware.com/paraview/paraview-guide-ce/merge_requests/94
https://gitlab.kitware.com/paraview/paraview-tutorial/merge_requests/10
- [ ] remove Catalyst Writers XML definitions, while keeping backward compatibility
- [ ] remove original INVERTIBLE_LUT value rendering approach
https://gitlab.kitware.com/vtk/vtk/merge_requests/6062
https://gitlab.kitware.com/paraview/paraview/merge_requests/3640
- [ ] remove old python code paths in cpstate.py, cinema and others, streamline what remains. For example outward facing pose camera.
https://gitlab.kitware.com/paraview/paraview/merge_requests/36355.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20097vtkWindowToImageFilter crashes in catalyst2020-11-27T13:08:19-05:00Seshu YamajalavtkWindowToImageFilter crashes in catalystI am trying to use vtkWindowToImageFilter in a ProgrammableFilter in a Catalyst pipeline. The following ProgrammableFilter script results in a seg. fault:
from paraview.simple import GetActiveView ...I am trying to use vtkWindowToImageFilter in a ProgrammableFilter in a Catalyst pipeline. The following ProgrammableFilter script results in a seg. fault:
from paraview.simple import GetActiveView
import vtk
ss = vtk.vtkWindowToImageFilter()
ss.SetInputBufferTypeToZBuffer()
ss.SetInput(GetActiveView().SMProxy.GetRenderWindow())
ss.Modified()
ss.Update()
output = ss.GetOutput()
I have attached a stack trace.[dump.txt](/uploads/00fc442ef6eefdcc7dd03c52ad001cc1/dump.txt)https://gitlab.kitware.com/paraview/paraview/-/issues/20125Catalyst: add easy mechanism to save data products for all input channels2020-11-27T13:06:25-05:00Utkarsh AyachitCatalyst: add easy mechanism to save data products for all input channelsAdd easy mechanism to save data products for all input channels, similar to gridwriter.py and allinputsgridwriter.py.Add easy mechanism to save data products for all input channels, similar to gridwriter.py and allinputsgridwriter.py.5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20353compilation of ParaView fails with error: no match for »operator+=« (operand...2020-11-27T10:48:57-05:00joergbuchwaldcompilation of ParaView fails with error: no match for »operator+=« (operand types are »vtkVector3d« and »const vtkVector3d«)Hi!
Using `cmake -GNinja -DPARAVIEW_USE_PYTHON=ON -DPARAVIEW_USE_MPI=ON -DVTK_SMP_IMPLEMENTATION_TYPE=TBB -DCMAKE_BUILD_TYPE=Release .. && ninja` the compilation of the current master fails on my machine running arch with gcc (GCC) 10.2....Hi!
Using `cmake -GNinja -DPARAVIEW_USE_PYTHON=ON -DPARAVIEW_USE_MPI=ON -DVTK_SMP_IMPLEMENTATION_TYPE=TBB -DCMAKE_BUILD_TYPE=Release .. && ninja` the compilation of the current master fails on my machine running arch with gcc (GCC) 10.2.0 poducing the following output
```
[18623/20287] Building CXX object VTKExtensions/FiltersGeneral/CMakeFiles/VTKExtensionsFiltersGeneral.dir/vtkSurfaceVectors.cxx.o
FAILED: VTKExtensions/FiltersGeneral/CMakeFiles/VTKExtensionsFiltersGeneral.dir/vtkSurfaceVectors.cxx.o
/usr/bin/c++ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX -DVTKExtensionsFiltersGeneral_EXPORTS -DVTK_MODULE_ENABLE_VTK_FiltersParallelFlowPaths=1 -DVTK_MODULE_ENABLE_VTK_FiltersParallelMPI=1 -DVTK_MODULE_ENABLE_VTK_ParallelMPI=1 -D_MPICC_H -IVTKExtensions/FiltersGeneral -I../VTKExtensions/FiltersGeneral -IVTK/Common/Core -I../VTK/Common/Core -IVTK/Common/DataModel -I../VTK/Common/DataModel -IVTK/Common/Math -I../VTK/Common/Math -IVTK/Common/Transforms -I../VTK/Common/Transforms -IVTK/Common/ExecutionModel -I../VTK/Common/ExecutionModel -IVTK/Filters/General -I../VTK/Filters/General -IVTK/Common/Misc -I../VTK/Common/Misc -IVTK/Filters/Core -I../VTK/Filters/Core -IVTK/Filters/Parallel -I../VTK/Filters/Parallel -IVTK/Filters/Extraction -I../VTK/Filters/Extraction -IVTK/Filters/Geometry -I../VTK/Filters/Geometry -IVTK/Filters/Hybrid -I../VTK/Filters/Hybrid -IVTK/Filters/Modeling -I../VTK/Filters/Modeling -IVTK/Filters/Sources -I../VTK/Filters/Sources -IVTK/Filters/Texture -I../VTK/Filters/Texture -IVTKExtensions/AMR -I../VTKExtensions/AMR -IVTK/Common/System -I../VTK/Common/System -IVTK/Filters/AMR -I../VTK/Filters/AMR -IVTKExtensions/Core -I../VTKExtensions/Core -IVTK/IO/XML -I../VTK/IO/XML -IVTK/IO/XMLParser -I../VTK/IO/XMLParser -IVTKExtensions/FiltersRendering -I../VTKExtensions/FiltersRendering -IVTKExtensions/Misc -I../VTKExtensions/Misc -IVTK/Filters/Generic -I../VTK/Filters/Generic -IVTK/Filters/HyperTree -I../VTK/Filters/HyperTree -IVTK/Imaging/Core -I../VTK/Imaging/Core -IVTK/Imaging/Sources -I../VTK/Imaging/Sources -IVTK/Parallel/Core -I../VTK/Parallel/Core -IVTK/Filters/ParallelFlowPaths -I../VTK/Filters/ParallelFlowPaths -IVTK/Filters/FlowPaths -I../VTK/Filters/FlowPaths -IVTK/Common/ComputationalGeometry -I../VTK/Common/ComputationalGeometry -IVTK/Filters/ParallelMPI -I../VTK/Filters/ParallelMPI -IVTK/Parallel/MPI -I../VTK/Parallel/MPI -isystem VTK/Utilities/KWIML -isystem ../VTK/Utilities/KWIML -isystem VTK/Utilities/KWSys -isystem ../VTK/Utilities/KWSys -isystem VTK/Utilities/MPI -isystem ../VTK/Utilities/MPI -O3 -DNDEBUG -fPIC -fvisibility=hidden -fvisibility-inlines-hidden -pthread -std=c++11 -MD -MT VTKExtensions/FiltersGeneral/CMakeFiles/VTKExtensionsFiltersGeneral.dir/vtkSurfaceVectors.cxx.o -MF VTKExtensions/FiltersGeneral/CMakeFiles/VTKExtensionsFiltersGeneral.dir/vtkSurfaceVectors.cxx.o.d -o VTKExtensions/FiltersGeneral/CMakeFiles/VTKExtensionsFiltersGeneral.dir/vtkSurfaceVectors.cxx.o -c ../VTKExtensions/FiltersGeneral/vtkSurfaceVectors.cxx
../VTKExtensions/FiltersGeneral/vtkSurfaceVectors.cxx: In Lambda-Funktion:
../VTKExtensions/FiltersGeneral/vtkSurfaceVectors.cxx:139:16: Fehler: no match for »operator+=« (operand types are »vtkVector3d« and »const vtkVector3d«)
139 | normal += cross.Dot(normal) > 0 ? cross : -cross;
| ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../VTKExtensions/FiltersGeneral/vtkSurfaceVectors.cxx: In Elementfunktion »virtual int vtkSurfaceVectors::RequestData(vtkInformation*, vtkInformationVector**, vtkInformationVector*)«:
../VTKExtensions/FiltersGeneral/vtkSurfaceVectors.cxx:159:18: Fehler: no match for »operator-=« (operand types are »vtkVector3d« and »vtkVector3d«)
159 | inVector -= k * normal;
| ~~~~~~~~~^~~~~~~~~~~~~
[18626/20287] Building CXX object VTKExtensions/FiltersGeneral/CM...ExtensionsFiltersGeneral.dir/vtkRectilinearGridConnectivity.cxx.o
ninja: build stopped: subcommand failed.
```https://gitlab.kitware.com/paraview/paraview/-/issues/20346Paraview failed to correctly show the distribution of velocity2020-11-26T22:15:38-05:00Hongwei TangParaview failed to correctly show the distribution of velocityHi, I recently got a quite strange problem when visualizing the `OpenFOAM` simulation results using `ParaView`. I followed this [tutorial](http://www.tfd.chalmers.se/~hani/kurser/OS_CFD_2019/lectureNotes/ImplementParabolicVelocityFixedVa...Hi, I recently got a quite strange problem when visualizing the `OpenFOAM` simulation results using `ParaView`. I followed this [tutorial](http://www.tfd.chalmers.se/~hani/kurser/OS_CFD_2019/lectureNotes/ImplementParabolicVelocityFixedValueBoundaryCondition.pdf) to set up a parabolic inlet velocity profile and used `simpleFoam` to run a simulation for flow past a wall mounted cube.
When I use `ParaView` to visualize the velocity distribution on inlet, the results are wrong.
![Screenshot_from_2020-11-24_11-16-31](/uploads/0be0a8a7ec505e852e1570128826c760/Screenshot_from_2020-11-24_11-16-31.png)
To further check the results, I use `sample` utility provided by `OpenFOAM` to extract velocity on the plane of symmetry (same position as shown in above figure), and the result is
![inlet_velocity](/uploads/65109224d123fe23b70b0c8b004951dc/inlet_velocity.png)
I also have utilized `Tecplot` and `Fluent` to check the results and it looks right.
![Screenshot_from_2020-11-24_11-35-23](/uploads/f91a8121a674bfc94c5a9d192b161602/Screenshot_from_2020-11-24_11-35-23.png)
I am curious whether that's bug of `ParaView`. The information about the version of software is listed as below:
`ParaView`: `5.7.0` and `5.8.0` (Both show wrong results)
`OpenFOAM`: `4.x` (`simpleFoam` solver)
mesh was generated by `gmsh 4.0`
I upload the case files (include results and mesh files) [here](https://pan.nuaa.edu.cn/share/0aa5805cd8863685329878bac0) and hope it can help.https://gitlab.kitware.com/paraview/paraview/-/issues/20053vth files not working correctly with TimeArray2020-11-25T15:19:28-05:00Boonthanome Nouanesengsyvth files not working correctly with TimeArrayWe have a set of vth files written by the Pagosa adaptor. The files have field data containing TimeValue, Cycle Number, and Simulation Time. When using the nightly build of Paraview, which has the feature of being able to choose the Time...We have a set of vth files written by the Pagosa adaptor. The files have field data containing TimeValue, Cycle Number, and Simulation Time. When using the nightly build of Paraview, which has the feature of being able to choose the Time Array, choosing Simulation Time for the Time Array creates problems, and breaks things.
Note that we also have a set of vtm files, but there doesn't seem to be any issues with the vtm files, just vth files.
Example dataset: [insitu.zip](/uploads/656d9c987a42e89c848cfaa05a94680a/insitu.zip)
How to reproduce:
1. Use a nightly build of Paraview
2. Choose the .vth files as a file series
3. The Time Array is automatically set to TimeValue, hit Apply to load
4. Note that there are 7 timesteps loaded
5. Change Time Array to Simulation Time, hit Apply
6. If you check the "Time" dropbox, the values haven't been updated
7. Step forward through a timestep
8. Change the Time Array back to TimeValue, hit Apply
9. Now Paraview says there are only two timesteps loaded
Pat has investigated this and gave an explanation of what is going on:
> XMLMultiBlockReader is a XMLCompositeDataSetReader is a XMLReader
> XMLUniformGridAMRReader is a XMLCompositeDataSetReader is a XMLReader
>
> When I change from TimeValue to CycleNumber with the .vtm, Modified() is set to true and RequestInformation is called (on MultiBlock, then Composite, and finally XMLReader). That causes the ActiveTimeArray to be set to CycleNumber and causes each of the seven .vtm files to be read. The data value corresponding to CycleNumber is pulled out of the FieldData and stored so that TIME_STEPS can be set. Then it goes back to the first file and loads the data via RequestData.
>
> When we change TimeValue to CycleNumber with the .vth, Modified() is set to true, but some extra stuff is called first. This includes RequestDataObject(), ReadVTKFile() and ReadPrimaryElement(). This last one calls XMLReader::ReadPrimaryElement which calls XMLReader::ReadXMLInformation which is where the CycleNumber is fetched.
>
> Then the regular RequestInformation sequence is called on UniformGridAMRReader, CompositeReader and XMLReader. The problem is that when XMLReader::ReadXMLInformation() was called the first time, it changed Modified() to false. So when it is called this time, and should read all the .vth and pull
> out the CycleNumber, that part does not execute, because it thinks there is no change.
>
> I was able to test this by removing the Modified() reset and was able to get it to work, but of course other stuff would be wrong.
>
> I think there is a problem in vtkXMLUniformGridAMRReader in that it should not be reading the FieldData and allowing the Modified flag to be reset prematurely.
@patchett2002 @cory.quammen5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20272Progessive rendering in OSPRay raycaster has no effect2020-11-24T13:43:45-05:00Christos TsolakisProgessive rendering in OSPRay raycaster has no effectIn current nightly builds (tested with `ParaView-5.8.1-1423-gbbd6ecf2f2-MPI-Linux-Python3.7-64bit`) increasing the number of passes in `OSPRay raycaster ` has no effect.
Steps to reproduce:
1. Enable streaming from `Edit/Settings` and re...In current nightly builds (tested with `ParaView-5.8.1-1423-gbbd6ecf2f2-MPI-Linux-Python3.7-64bit`) increasing the number of passes in `OSPRay raycaster ` has no effect.
Steps to reproduce:
1. Enable streaming from `Edit/Settings` and restart paraview.
1. Create a sphere with some theta and phi around 30.
1. enable raytracing
1. set progressive passes to 1 and save a screenshot
1. set progressive passes to 10 and save a screenshot
1. the two screenshots are bitwise the same but they shouldn't.
release version passes the above test. see also https://gitlab.kitware.com/vtk/vtk/-/issues/180295.9 (Fall 2020)https://gitlab.kitware.com/paraview/paraview/-/issues/20338v5.9-RC1 crash when raytracing molecular data2020-11-24T10:22:39-05:00Jean M. Favrev5.9-RC1 crash when raytracing molecular dataI downloaded and installed ParaView-5.9.0-RC1-MPI-Linux-Python3.8-64bit from www.paraview.org
paraview /local/apps/VTK/VTKBuild/ExternalData/Testing/Data/nanowireTB23K298.xyz
switch raytracing on
OSPRay ERROR: OSPData: all numItems mu...I downloaded and installed ParaView-5.9.0-RC1-MPI-Linux-Python3.8-64bit from www.paraview.org
paraview /local/apps/VTK/VTKBuild/ExternalData/Testing/Data/nanowireTB23K298.xyz
switch raytracing on
OSPRay ERROR: OSPData: all numItems must be positive
OSPRay ERROR: null handle provided to void ospCommit(OSPObject)
OSPRay ERROR: OSPData: shared buffer is NULL
OSPRay ERROR: null handle provided to void ospCommit(OSPObject)
Loguru caught a signal: SIGSEGV5.9 (Fall 2020)David E. DeMarleDavid E. DeMarlehttps://gitlab.kitware.com/paraview/paraview/-/issues/20269OSPRay memory leaks2020-11-23T20:10:54-05:00Michael MiglioreOSPRay memory leaks# Leak memory observed on Paraview (master branch)
When loading VTS files and rendering them with ray-tracing activated using Surface representation, memory leaks are observable.
## Step for reproducing the issue
- Create a .vts file...# Leak memory observed on Paraview (master branch)
When loading VTS files and rendering them with ray-tracing activated using Surface representation, memory leaks are observable.
## Step for reproducing the issue
- Create a .vts file (this can be done by creating a 'Wavelet' filter and then applying a 'Warp By Scalar' filter).
- To make the memory leak easily noticeable create a big wavelet (Whole Extent: x:[-1000 1000] y[-1000 1000] z[0 0] for example)
- Load it and activate OSPRay (it can ever be the pathtracer or the raytracer)
- File -> Save Animation ... then Save it with the default parameters
- Hitting the "Play Animation" button inside Paraview also seems to produce the memory leak
## Additional remarks
- The rendering of the wavelet when loading the .vts file is also much slower than rendering the wavelet/warp
- Not observed on PV 5.8.1https://gitlab.kitware.com/paraview/paraview/-/issues/19961MergeBlocks Filter Coincident Point Merging Decision2020-11-23T16:54:49-05:00John PatchettMergeBlocks Filter Coincident Point Merging DecisionWe need a scheme for managing multiblock datasets and merging blocks. We have data sets that have physically discontinuous meshes that might have coincident points that should never be merged. We have a scalar field that has values that ...We need a scheme for managing multiblock datasets and merging blocks. We have data sets that have physically discontinuous meshes that might have coincident points that should never be merged. We have a scalar field that has values that would determine whether any given point could be merged and we could create a bitmask of the same. We want to be able to merge points to manage parallel distributed created datasets and run things like ghost cell generator for distributed analysis. Although the meshes are discontinuous they are represented in the same unstructured grid dataset.5.9 (Fall 2020)Yohann Bearzi (Kitware)Yohann Bearzi (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/20307Release ParaView 5.9.0-RC12020-11-20T15:50:39-05:00Cory Quammencory.quammen@kitware.comRelease ParaView 5.9.0-RC1# Preparatory steps
- Update ParaView guides
- User manual
- [x] Rename to ParaViewGuide-5.9.0.pdf
- [x] Upload to www.paraview.org/files/v5.9
- Catalyst Guide
- [x] Rename to ParaViewCatalystGuide-5.9.0.pdf
...# Preparatory steps
- Update ParaView guides
- User manual
- [x] Rename to ParaViewGuide-5.9.0.pdf
- [x] Upload to www.paraview.org/files/v5.9
- Catalyst Guide
- [x] Rename to ParaViewCatalystGuide-5.9.0.pdf
- [x] Upload to www.paraview.org/files/v5.9
- Getting Started Guide
- [x] Rename to ParaViewGettingStarted-5.9.0.pdf
- [x] Upload to www.paraview.org/files/v5.9
- Assemble release notes into `Documentation/release/ParaView-5.9.0`.
- [x] Get positive review and merge.
# Update ParaView
- [x] Update `master` branch for **paraview**
```
git fetch origin
git checkout master
git merge --ff-only origin/master
git submodule update --recursive --init
```
- [x] Update `version.txt` and tag the commit
```
git checkout -b update-to-v5.9.0-RC1
echo 5.9.0-RC1 > version.txt
git commit -m 'Update version number to 5.9.0-RC1' version.txt
git tag -a -m 'ParaView 5.9.0-RC1' v5.9.0-RC1 HEAD
```
- Integrate changes to `master` branch
- [x] Create a merge request targeting `master` (do *not* add `Backport: release`)
- [x] `Do: merge`
- Update VTK's `paraview/release` branch
- [x] Change directory to VTK source
- [x] `git push origin <paraview-vtk-submodule-hash>:paraview/release`
- [x] Update kwrobot with the new `paraview/release` branch rules
- Integrate changes to `release` branch
- [x] Change directory to ParaView source. Stay on the `update-to-v5.9.0-RC1` branch.
- [x] `git config -f .gitmodules submodule.VTK.branch paraview/release`
- [x] `git commit -m "release: follow VTK's paraview/release branch" .gitmodules`
- [x] Merge new `release` branch into `master` using `-s ours`
- `git checkout master`
- `git merge --no-ff -s ours -m "Merge branch 'release'" update-to-v5.9.0-RC1`
- [x] `git push origin master update-to-v5.9.0-RC1:release`
- [x] Update kwrobot with the new `release` branch rules
- Create tarballs
- [x] ParaView (`Utilities/Maintenance/create_tarballs.bash --txz --tgz --zip -v v5.9.0-RC1`)
- Upload tarballs to `paraview.org`
- [x] `rsync -rptv $tarballs paraview.release:ParaView_Release/v5.9/`
# Update ParaView-Superbuild
- [x] Update `release` branch for **paraview/paraview-superbuild**
```
git fetch origin
git checkout release
git merge --ff-only origin/release
git submodule update
git checkout -b update-to-v5.9.0-RC1
```
- Update `CMakeLists.txt`
- [x] Set ParaView source selections in `CMakeLists.txt` and force explicit
version in `CMakeLists.txt`:
```
# Force source selection setting here.
set(paraview_SOURCE_SELECTION "5.9.0-RC1" CACHE STRING "Force version to 5.9.0-RC1" FORCE)
set(paraview_FROM_SOURCE_DIR OFF CACHE BOOL "Force source dir off" FORCE)
```
- Update versions
- [x] Guide selections in `versions.cmake`
- [x] Docker: update default tag strings (in `Scripts/docker/ubuntu/development/Dockerfile`)
- [x] ARG PARAVIEW_TAG=v5.9.0-RC1
- [x] ARG SUPERBUILD_TAG=v5.9.0-RC1
- [x] Commit changes and push to GitLab
```
git add versions.cmake CMakeLists.txt Scripts/docker/ubuntu/development/Dockerfile
git commit -m "Update the default version to 5.9.0-RC1"
git gitlab-push
```
- Integrate changes to `master` branch
- [x] Create a merge request targeting `master`, title beginning with WIP (do *not* add `Backport: release` to description)
- [x] Build binaries (`Do: test`)
- [x] Download the binaries that have been generated in the dashboard results. They will be deleted within 24 hours.
- [x] Remove explicit version forcing added in CMakeLists.txt, amend the commit, and force push
```
git add CMakeLists.txt
git commit --amend --no-edit
git gitlab-push -f
```
- Finalize merge request
- [x] Remove WIP from merge request title
- [x] Get positive review
- [x] `Do: merge`
- [x] `git tag -a -m 'ParaView superbuild 5.9.0-RC1' v5.9.0-RC1 HEAD`
- Integrate changes to `release` branch
- [x] `git push origin update-to-v5.9.0-RC1:release`
# Sign macOS binaries
- [x] Upload to signing server, run script, download resulting .pkg and .dmg files
- [ ] Install from .pkg and verify that it is signed with `codesign -dvvv /Applications/ParaView-5.9.0-RC1.app/`
- [ ] Install from .dmg and verify that it is signed with `codesign -dvvv /Applications/ParaView-5.9.0-RC1.app/`
# Validating binaries
For each binary, open the Python shell and run the following:
```python
import numpy
s = Show(Sphere())
ColorBy(s, ('POINTS', 'Normals', 'X'))
Show(Text(Text="$A^2$"))
```
Check that
- Getting started guide opens
- Examples load and match thumbnails in dialog
- Python. Open the Python shell and run
- Plugins are present and load properly
- OSPRay raycasting and pathtracing runs
- OptiX pathtracing runs
- IndeX runs
- AutoMPI
Binary checklist
- [x] macOS
- [x] Linux
- [x] Linux osmesa
- [x] Windows MPI (.exe)
- [x] Windows MPI (.zip)
- [x] Windows no-MPI (.exe)
- [x] Windows no-MPI (.zip)
# Upload binaries
- [x] Upload binaries to `paraview.org` (`rsync -rptv $binaries paraview.release:ParaView_Release/v5.9/`)
- [x] Ask @utkarsh.ayachit to regenerate `https://www.paraview.org/files/listing.txt` and `md5sum.txt` on the website
```
buildListing.sh
updateMD5sum.sh v5.9
```
- [x] Test download links on https://www.paraview.org/download
# Push tags
- [x] In the `paraview` repository, run `git push origin v5.9.0-RC1`.
- [x] In the `paraview-superbuild` repository, run `git push origin v5.9.0-RC1`.
# Post-release
- [x] Post an announcement in the Announcements category on
[discourse.paraview.org](https://discourse.paraview.org/).
/cc @ben.boeckel
/cc @cory.quammen
/cc @utkarsh.ayachit5.9 (Fall 2020)https://gitlab.kitware.com/paraview/paraview/-/issues/20314v5.9-RC1 file is constantly re-opened and read when playing camera keyframes ...2020-11-20T08:43:47-05:00Jean M. Favrev5.9-RC1 file is constantly re-opened and read when playing camera keyframes animationThe issue raised https://discourse.paraview.org/t/xmlimagedatareader-keeps-re-executing-with-a-static-dataset-when-playing-keyframes-animation/5692 appears in v5.9-RC1. A reproducer is included on the discourse pageThe issue raised https://discourse.paraview.org/t/xmlimagedatareader-keeps-re-executing-with-a-static-dataset-when-playing-keyframes-animation/5692 appears in v5.9-RC1. A reproducer is included on the discourse page5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20334three different format.h in VTK causing segfaults in ParaView2020-11-20T08:29:56-05:00Utkarsh Ayachitthree different format.h in VTK causing segfaults in ParaView```
find . | grep format.h
...
./VTK/ThirdParty/diy2/vtkdiy2/include/vtkdiy2/fmt/format.h
./VTK/ThirdParty/ioss/vtkioss/fmt/format.h
./VTK/ThirdParty/vtkm/vtkvtkm/vtk-m/vtkm/thirdparty/diy/vtkmdiy/include/vtkmdiy/thirdparty/fmt/format.h
...```
find . | grep format.h
...
./VTK/ThirdParty/diy2/vtkdiy2/include/vtkdiy2/fmt/format.h
./VTK/ThirdParty/ioss/vtkioss/fmt/format.h
./VTK/ThirdParty/vtkm/vtkvtkm/vtk-m/vtkm/thirdparty/diy/vtkmdiy/include/vtkmdiy/thirdparty/fmt/format.h
...
```
Looks like they are not mangled appropriately or something:
Here's a callstack from a segfault when opening can.ex2 with IOSS reader:
```
....ParaView/VTK/ThirdParty/vtkm/vtkvtkm/vtk-m/vtkm/thirdparty/diy/vtkmdiy/include/vtkmdiy/thirdparty/fmt/format.h:2068
....ParaView/VTK/ThirdParty/ioss/vtkioss/fmt/format.h:3158
```5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19891StreamTracer breaks at random positions2020-11-19T07:38:14-05:00Jo HeyStreamTracer breaks at random positionsI recurrently face the issue that several tracers are lost without any particular reasons in the StreamTracer filter (no termination value is given). One can see the bug quite well below
![streamtracer_bug](/uploads/cafdea4a4fa515973d1...I recurrently face the issue that several tracers are lost without any particular reasons in the StreamTracer filter (no termination value is given). One can see the bug quite well below
![streamtracer_bug](/uploads/cafdea4a4fa515973d1cbd0b480c40ef/streamtracer_bug.png)
where a continuous line injection shows “holes” forming by the loss of a trajectory. It seems that the choice of the integrator between RK2/RK4 and RK4-5 modifies which tracer is missed. However, changing tracing parameters (steps, terminal speeds, max error, cell/point locator) do not have an effect.
Restarting the tracer exactly at the point it were lost do not solve the problem (no streamline is outputed).
I attached the vtu velocity file here (mesh is produced by snappyHexMesh of openfoam) :
[Periodic_BCC_Packing_lowprec.vtu](https://drive.google.com/file/d/1qPvREufg7OL5MeuD9nFjAaeSbFXyQ-Jz/view?usp=sharing)
And a set of position where the streamtracer breaks (in FORWARD direction)
[point_lost_BCC.txt](/uploads/ab5ec0869a89cc618f09ca35f004df9c/point_lost_BCC.txt)
Regards,
Jorishttps://gitlab.kitware.com/paraview/paraview/-/issues/20305Mac/Python 3 issue with -m option2020-11-18T16:16:15-05:00Sebastien JourdainMac/Python 3 issue with -m optionRunning the following command should give you the following
```
$ pvbatch -m paraview.apps -l
Available applications:
divvy
flow
lite
visualizer
```
On Mac with ParaView-5.8.1-1539-ge438006c96.app we get
```
./pvbatch -m parav...Running the following command should give you the following
```
$ pvbatch -m paraview.apps -l
Available applications:
divvy
flow
lite
visualizer
```
On Mac with ParaView-5.8.1-1539-ge438006c96.app we get
```
./pvbatch -m paraview.apps -l
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
Available applications:
```Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/20318Rotations seems to be inverted in pqCameraToolbar2020-11-18T13:28:28-05:00MelanieCarriereRotations seems to be inverted in pqCameraToolbarThe `CW` and the `CCW`rotations seems to be inverted https://gitlab.kitware.com/paraview/paraview/-/blob/master/Qt/ApplicationComponents/pqCameraToolbar.cxx#L52The `CW` and the `CCW`rotations seems to be inverted https://gitlab.kitware.com/paraview/paraview/-/blob/master/Qt/ApplicationComponents/pqCameraToolbar.cxx#L52Mathieu Westphal (Kitware)Mathieu Westphal (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/20280Crystal Eyes stereo with cave broken in ParaView > 5.7.02020-11-18T11:19:47-05:00Paul LafoixCrystal Eyes stereo with cave broken in ParaView > 5.7.0In Windows and Linux, the stereo with crystal eyes is broken when using a cave display in versions > 5.7.0.
This works on a ParaView 5.7.0.
Steps to reproduce :
./paraview --stereo --stereo-type="Crystal Eyes"
./pvserver --stereo --st...In Windows and Linux, the stereo with crystal eyes is broken when using a cave display in versions > 5.7.0.
This works on a ParaView 5.7.0.
Steps to reproduce :
./paraview --stereo --stereo-type="Crystal Eyes"
./pvserver --stereo --stereo-type="Crystal Eyes" PARAVIEW_SOURCES/Examples/VR/zSpace.pvx
The file zSpace.pvx is in the sources of paraview and enable the cave display in the server. Replace PARAVIEW_SOURCES with the path to your paraview sources.5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/19965OSPRay Background/Env Map colors too bright when in C/S mode2020-11-18T10:58:08-05:00Marston OSPRay Background/Env Map colors too bright when in C/S modeIssue with new OSPRay features in the dev build. Currently testing with build 5.8.0-611-gf6f72baa39 but this bug has been around for a while.
**Bug:**
When running in client server mode with OSPRay Pathtracer on, the background images ...Issue with new OSPRay features in the dev build. Currently testing with build 5.8.0-611-gf6f72baa39 but this bug has been around for a while.
**Bug:**
When running in client server mode with OSPRay Pathtracer on, the background images and colors (for both Backplates and Env Maps) are way brighter than they should be. I believe from previous conversations with the Paraview team this may be to both the client and server loading the images on top of each other?
This does not happen when running in regular client mode, so it is a C/S bug.
**How to replicate:**
Fortunately, this is very easy to test. Run in C/S mode, and just turn on the Pathtracing. The standard grey background will change color. I’ve got some screenshots below showing this.
Regular Color:
![image](/uploads/792c390c6f1bf49375d9ce89412e8a67/image.png)
Brightened Colors:
![image](/uploads/175f003dc45b2802aba2794ad5f93c64/image.png)5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/16850Accessing "Settings" in Python2020-11-18T05:54:57-05:00Utkarsh AyachitAccessing "Settings" in PythonAs for v5.1.2, there is no easy way to access application settings through Python. We need to add a convenient API to access and change application settings through Python.As for v5.1.2, there is no easy way to access application settings through Python. We need to add a convenient API to access and change application settings through Python.Christian WitzlerChristian Witzlerhttps://gitlab.kitware.com/paraview/paraview/-/issues/20308__file__ is not defined when ParaView python state file is loaded through ``F...2020-11-17T22:29:50-05:00Nima Tofighi__file__ is not defined when ParaView python state file is loaded through ``File->Load State...''When loading a python state file through ``File->Load State...'' the __file__ variable is not defined. This makes it impossible to use relative paths inside python state files when loaded in this manner. Loading the state file through Py...When loading a python state file through ``File->Load State...'' the __file__ variable is not defined. This makes it impossible to use relative paths inside python state files when loaded in this manner. Loading the state file through Python Shell via Run Script works as expected and __file__ points to the location of the state file.
Example file:
[placeholder.py](/uploads/67fb103c0b3a10dea251dc39b3f79b61/placeholder.py)
Error message:
Traceback (most recent call last):
File "<string>", line 14, in <module>
NameError: name '__file__' is not defined5.9 (Fall 2020)Utkarsh AyachitUtkarsh Ayachit