ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2023-03-06T10:36:25-05:00https://gitlab.kitware.com/paraview/paraview/-/issues/21669Make `--venv` a pvpython/pvbatch/paraview CLI argument2023-03-06T10:36:25-05:00Utkarsh AyachitMake `--venv` a pvpython/pvbatch/paraview CLI argumentCurrently, python scripts can import [`paraview.web.venv`](https://gitlab.kitware.com/paraview/paraview/-/blob/master/Web/Python/paraview/web/venv.py) to setup module paths etc. to load packages from a virualenv. The current approach req...Currently, python scripts can import [`paraview.web.venv`](https://gitlab.kitware.com/paraview/paraview/-/blob/master/Web/Python/paraview/web/venv.py) to setup module paths etc. to load packages from a virualenv. The current approach requires that `import paraview.web.venv` is done in the Python script that pvpython/pvbatch executes, which makes it impossible for the Python script itself to be a part of the virtual env. For example, I have a nice Python package with Viz pipelines that I pip install in a venv. I simply want to run that package with pvpython as follows:
```sh
pvpython -m my_nice_package
```
It's not possible currently. Hence, I propose we move the venv handling to pvpython/pvbatch itself. Thus, make the following work:
```sh
pvpython --venv /.../venv -m my_nice_package
```
It makes sense to add this cmd-line arg to all executables that support Python including the server executables.Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/21800Add a quick toggle for precision on the information tab2023-03-07T07:39:16-05:00W. Alan ScottAdd a quick toggle for precision on the information tabPlease add a quick toggle for precision displayed on the information tab.
Idea came from discourse post here: https://discourse.paraview.org/t/data-arrays-ranges-in-the-information-tab-may-be-at-too-high-precision/11561Please add a quick toggle for precision displayed on the information tab.
Idea came from discourse post here: https://discourse.paraview.org/t/data-arrays-ranges-in-the-information-tab-may-be-at-too-high-precision/115615.13 (Summer 2024)https://gitlab.kitware.com/paraview/paraview/-/issues/21806File Open Set reader as defaults fails with CGNS files2023-03-07T22:43:22-05:00W. Alan ScottFile Open Set reader as defaults fails with CGNS filesThe Set Reader as Default is failing with the CGNS file reader. Replicate as follows:
* 5.11.0, builtin server, Linux.
* File/ Open/ SomeCGNSFile.cgns. Open Data Dialog opens. Select IOSS reader. Set Reader as Default. Dialog will cl...The Set Reader as Default is failing with the CGNS file reader. Replicate as follows:
* 5.11.0, builtin server, Linux.
* File/ Open/ SomeCGNSFile.cgns. Open Data Dialog opens. Select IOSS reader. Set Reader as Default. Dialog will close. Note - I used sparc.cgns.4.[0-3].
* Delete cgns file.
* File/ Open/ SomeCGNSFile.cgns.
The Open Data Dialog will open again. This is a bug.5.13 (Summer 2024)https://gitlab.kitware.com/paraview/paraview/-/issues/21797Python venv should not be inside paraview.web2023-03-08T10:15:33-05:00Loic TetrelPython venv should not be inside paraview.webFollowing this post: https://discourse.paraview.org/t/quick-guide-to-using-pytorch-in-paraview/9037
I was trying to load my custom python venv, unfortunately this functionality lives inside paraview web.
IMO I think the venv option shou...Following this post: https://discourse.paraview.org/t/quick-guide-to-using-pytorch-in-paraview/9037
I was trying to load my custom python venv, unfortunately this functionality lives inside paraview web.
IMO I think the venv option should be activated by default since it is very common for python devs.
If we don't want to have it by default, it does not make sense to have it inside paraview web.https://gitlab.kitware.com/paraview/paraview/-/issues/21798set properties value in qt project2023-03-08T10:19:58-05:00KDG532set properties value in qt projectHi, I'm working on a project with QT based on ParaView. I don't know how to achieve this function in my code:
**Set specific properties(Transforming, transparency, etc.) in the properties panel after loading some VTK objects.**
I load...Hi, I'm working on a project with QT based on ParaView. I don't know how to achieve this function in my code:
**Set specific properties(Transforming, transparency, etc.) in the properties panel after loading some VTK objects.**
I load the VTK objects with this code:
`QVector<pqPipelineSource *> Qvtkpointer = pqLoadDataReaction::loadFilesForSupportedTypes(files);`https://gitlab.kitware.com/paraview/paraview/-/issues/21796Extreme render artifacts for Surface With Edges on side sets2023-03-08T10:21:19-05:00Kenneth MorelandExtreme render artifacts for Surface With Edges on side setsI am seeing some extreme rendering artifacts when rendering side sets and setting the representation to `Surface With Edges`. I on a Mac with an Intel processor using the binary downloaded from Kitware. To replicate the problem, do the f...I am seeing some extreme rendering artifacts when rendering side sets and setting the representation to `Surface With Edges`. I on a Mac with an Intel processor using the binary downloaded from Kitware. To replicate the problem, do the following:
1. Open `disk_out_ref.ex2`
1. Turn off block `block_1`.
2. Turn on sets `surface_1` and `surface_2`
3. `Apply`
2. Change the representation to `Surface With Edges`
3. Rotate the camera
At this point, I am seeing extreme rendering effects. Here is a video I captured on my machine.
![render-artifacts](/uploads/37e48f23c8539657b6d6000c71279565/render-artifacts.m4v)https://gitlab.kitware.com/paraview/paraview/-/issues/21801Chombo reader cannot read displacement field to deal with non-cartesian geome...2023-03-08T11:50:58-05:00Alankar DuttaChombo reader cannot read displacement field to deal with non-cartesian geometriesThe Chombo AMR reader in `ParaView` cannot read the displacement field to properly parse grids with non-Cartesian data. The same files on being read by `VisIt` get the displacement field detected and hence geometric warping of the cells ...The Chombo AMR reader in `ParaView` cannot read the displacement field to properly parse grids with non-Cartesian data. The same files on being read by `VisIt` get the displacement field detected and hence geometric warping of the cells accounted for. See the thread on discourse for more details: https://discourse.paraview.org/t/displacement-of-mesh-operator-in-paraview/11590/11https://gitlab.kitware.com/paraview/paraview/-/issues/21028Use SMP accelerated contour in ParaView2023-03-08T11:51:11-05:00Mathieu Westphal (Kitware)Use SMP accelerated contour in ParaViewDefault Contour filter should use SMP accelerated contour in ParaView when possible.
Current status:
* AMR: uses vtkAMRDualContour (non-SMP)
* HTG: uses vtkHyperTreeGridContour (non-SMP)
* UG: uses vtkContour3DLinearGrid is ComputeScala...Default Contour filter should use SMP accelerated contour in ParaView when possible.
Current status:
* AMR: uses vtkAMRDualContour (non-SMP)
* HTG: uses vtkHyperTreeGridContour (non-SMP)
* UG: uses vtkContour3DLinearGrid is ComputeScalars is off (SMP)
* Image: uses vtkSynchronizedTemplates2D/3D if not UniformGrid (non-SMP)
* Rectilinear: uses vtkRectilinearSynchronizedTemplates (non-SMP)
* Other: uses vtkGridSynchronizedTemplates3D (non-SMP)
Should be used
* AMR: use vtkAMRDualContour (non-SMP)
* HTG: use vtkHyperTreeGridContour (non-SMP)
* UG: use vtkContour3DLinearGrid *all the time* (SMP)
* Image: use vtkFlyingEdges2D/3D (SMP)
* Other: uses vtkContourFilter (non-SMP, uses vtk*SynchronizedTemplates*)
The only needed changes are for UG and ImageData.Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/21808CGNS Reader does not open distributed datasets as expected2023-03-09T03:42:24-05:00Mathieu Westphal (Kitware)CGNS Reader does not open distributed datasets as expectedThe CGNS Reader should open distributed dataset distributed on multiple nodes, but only the surfaces are distributed, not the volume:
Steps to reproduce:
- mpirun -np 2 ./bin/server
- run ParaView
- connect to server
- open channelBump_...The CGNS Reader should open distributed dataset distributed on multiple nodes, but only the surfaces are distributed, not the volume:
Steps to reproduce:
- mpirun -np 2 ./bin/server
- run ParaView
- connect to server
- open channelBump_solution.cgns for ParaView testing data, load surface and volume
- surfaces are distributed, volumes are not distributedhttps://gitlab.kitware.com/paraview/paraview/-/issues/21810empty plots with pvbatch because of wrong order of operations in python trace2023-03-09T05:09:57-05:00Arianna Formentiempty plots with pvbatch because of wrong order of operations in python traceHi all!
I'm trying to save an animation from terminal using `pvbatch`.
To do so, I follow this workflow:
1. generate a state file with the desired plotting configuration (say, `state.pvsm`)
2. reset session
3. start a python trace
4. ...Hi all!
I'm trying to save an animation from terminal using `pvbatch`.
To do so, I follow this workflow:
1. generate a state file with the desired plotting configuration (say, `state.pvsm`)
2. reset session
3. start a python trace
4. load `state.pvsm`
5. save animation
6. abort
7. save trace (say, `trace.py`)
8. run `pvbatch` with `trace.py` as input
I'm using ParaView 5.11 (ParaView-5.11.0-MPI-Linux-Python3.9-x86_64).
The issue I have found is that `pvbatch` generates empty plots (e.g. white .png files) if I use `trace.py` as is.
With @ax3l's help, I was able to solve the problem by switching two blocks of code.
The original`trace.py` file which does not work reads:
```
from paraview.simple import *
paraview.simple._DisableFirstRenderCameraReset()
animationScene1 = GetAnimationScene()
timeKeeper1 = GetTimeKeeper()
renderView1 = FindViewOrCreate('RenderView1', viewtype='RenderView')
LoadState('state.pvsm')
# continues...
```
The modified `trace.py` script which seems to work reads:
```
from paraview.simple import *
paraview.simple._DisableFirstRenderCameraReset()
LoadState('state.pvsm')
animationScene1 = GetAnimationScene()
timeKeeper1 = GetTimeKeeper()
renderView1 = FindViewOrCreate('RenderView1', viewtype='RenderView')
# continues...
```
(To be precise, I also had to remove two lines like this one:\
`layout1.SetSize(1919, 909)`\
from `trace.py` because they raise an attribute error.)
It seems like the state should be loaded before generating the render view.
Does this make sense? Do you think it could be fixed?
Pinging @berkgeveci and @ax3l since I know you may talk about this soon enough.https://gitlab.kitware.com/paraview/paraview/-/issues/21807Plot Over Line Crash in recent versions (working fine in version 5.4.1)2023-03-09T11:33:24-05:00Joshua ObradorsPlot Over Line Crash in recent versions (working fine in version 5.4.1)I'm using ParaView 5.11.0 on Windows and noticed Plot Over Line leads to a crash under certain circumstances e.g. when the line intersects parts of the model that contain different properties (in my files POL crashes if done from top to ...I'm using ParaView 5.11.0 on Windows and noticed Plot Over Line leads to a crash under certain circumstances e.g. when the line intersects parts of the model that contain different properties (in my files POL crashes if done from top to bottom of the column but will not crash if done from Y=1200 to top).
I have verified that this does not happen in ParaView 5.4.1.
Any help on that would be much appreciated.
[Test_POL_015.plt](/uploads/0c6aef36996b541c0b4fd4560e43905d/Test_POL_015.plt)[Test_POL_015.xmf](/uploads/64bd736f6c277249ecbad7cf45bfaf1d/Test_POL_015.xmf)Yohann Bearzi (Kitware)Yohann Bearzi (Kitware)https://gitlab.kitware.com/paraview/paraview/-/issues/21803Clean up Prism plugin log scale axis grid2023-03-10T15:57:19-05:00W. Alan ScottClean up Prism plugin log scale axis gridThe Prism plugin displays objects in pressure/temperature/density phase space. These spaces can be log scaled. Current log scaling is correct, but confusing. The tick locations should not be linearly placed, but placed along a log spac...The Prism plugin displays objects in pressure/temperature/density phase space. These spaces can be log scaled. Current log scaling is correct, but confusing. The tick locations should not be linearly placed, but placed along a log space. See the Color Legend, log scaled, for correct behavior.
@spiros.tsalikis5.13 (Summer 2024)Spiros TsalikisSpiros Tsalikishttps://gitlab.kitware.com/paraview/paraview/-/issues/17304Scale Glyphs when changing variables2023-03-13T06:06:48-04:00W. Alan ScottScale Glyphs when changing variablesPlease have the Glyph filter update properly when users change variables.
* Mac, 5.3.0, local server
* can.exo. All vars on. Apply.
* Move forward one time step.
* Glyph. Watch Scale Factor. Change Vectors to DISPL.
The Scale Fac...Please have the Glyph filter update properly when users change variables.
* Mac, 5.3.0, local server
* can.exo. All vars on. Apply.
* Move forward one time step.
* Glyph. Watch Scale Factor. Change Vectors to DISPL.
The Scale Factor should update to this new variable.5.6 (Fall 2018)Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/21817ParaView fails to find ExprTk due to changed type in the version variable2023-03-13T08:24:12-04:00yurivictParaView fails to find ExprTk due to changed type in the version variableSince exprtk-0.0.2 the type of the 'version' variable changed. FindExprTk.cmake parses the header and fails to find ExprTk version >=2.7
Patch that resolves this problem:
```
--- VTK/CMake/FindExprTk.cmake.orig 2023-03-11 22:27:31 U...Since exprtk-0.0.2 the type of the 'version' variable changed. FindExprTk.cmake parses the header and fails to find ExprTk version >=2.7
Patch that resolves this problem:
```
--- VTK/CMake/FindExprTk.cmake.orig 2023-03-11 22:27:31 UTC
+++ VTK/CMake/FindExprTk.cmake
@@ -21,8 +21,8 @@ mark_as_advanced(ExprTk_INCLUDE_DIR)
if (ExprTk_INCLUDE_DIR)
file(STRINGS "${ExprTk_INCLUDE_DIR}/exprtk.hpp" _exprtk_version_header
- REGEX "static const char\\* version")
- string(REGEX MATCH "static const char\\* version = \"([0-9.]+)\"" _exprtk_version_match "${_exprtk_version_header}")
+ REGEX "static char_cptr *version")
+ string(REGEX MATCH "static char_cptr *version = \"([0-9.]+)\"" _exprtk_version_match "${_exprtk_version_header}")
set(ExprTk_VERSION "${CMAKE_MATCH_1}")
unset(_exprtk_version_header)
unset(_exprtk_version_match)
```
But it would be better if you would compile and run a program that prints this value in cmake, instead of parsing the header file.https://gitlab.kitware.com/paraview/paraview/-/issues/21818rescale glyph in python script using current values2023-03-14T04:11:02-04:00ahmad alminnawirescale glyph in python script using current valuesHi,
I am using the python tracing feature to write a script that I will use for multiple models
I have an arrow glyph and I am trying to script the scale factor in python. I want to reset the glyph scale factor with respect to each spec...Hi,
I am using the python tracing feature to write a script that I will use for multiple models
I have an arrow glyph and I am trying to script the scale factor in python. I want to reset the glyph scale factor with respect to each specific model.
using the user interface it is simple and straightforward forward using a button but in the script, it saves a value that is specific to one model. I can assign a scalar to it using "glyph1.ScaleFactor = ..." but I want to automate it so that it would reset using current data values (just like pressing the button on the user interface)
How can I write the script to reset that factor depending on each case?
I can not figure out how to do so can you please help?https://gitlab.kitware.com/paraview/paraview/-/issues/21819Selecting points on a line rendered as tubes crashes ParaView2023-03-14T09:40:06-04:00Kenneth MorelandSelecting points on a line rendered as tubes crashes ParaViewWhen you try to select points on line elements that are being rendered as tubes with enough thickness, ParaView crashes (observed in 5.11, Mac). To replicate, do the following.
1. Create as `Line` source. `Apply`.
2. Change the `Line Wi...When you try to select points on line elements that are being rendered as tubes with enough thickness, ParaView crashes (observed in 5.11, Mac). To replicate, do the following.
1. Create as `Line` source. `Apply`.
2. Change the `Line Width` to `5`.
3. Turn on `Render Lines as `Tubes`.
4. Click `Select Points On` and select some region of the line. -> Crashhttps://gitlab.kitware.com/paraview/paraview/-/issues/21820ParaView Simple import into Python fails on Linux2023-03-15T07:04:58-04:00CoryParaView Simple import into Python fails on LinuxWhen trying to import ```paraview.simple``` into my Python script, I get the following error:
```
Traceback (most recent call last):
File "/home/cory/git-repos/MachLine/dev/helper_scripts/paraview_functions.py", line 3, in <module>
...When trying to import ```paraview.simple``` into my Python script, I get the following error:
```
Traceback (most recent call last):
File "/home/cory/git-repos/MachLine/dev/helper_scripts/paraview_functions.py", line 3, in <module>
import paraview.simple as pvs
File "/home/cory/Paraview/lib/python3.9/site-packages/paraview/simple.py", line 41, in <module>
from paraview import servermanager
File "/home/cory/Paraview/lib/python3.9/site-packages/paraview/servermanager.py", line 60, in <module>
from paraview.modules.vtkRemotingApplication import *
ImportError: /lib/x86_64-linux-gnu/libp11-kit.so.0: undefined symbol: ffi_type_pointer, version LIBFFI_BASE_7.0
```
I am running Linux Mint 20.2 Cinnamon 64-bit.https://gitlab.kitware.com/paraview/paraview/-/issues/21822output message frozen2023-03-15T07:44:10-04:00shi an tingoutput message frozenhttps://gitlab.kitware.com/paraview/paraview/-/issues/21816[Feature Request] Ship ParaView with a mangled VTK2023-03-15T13:17:55-04:00Julien Fausty[Feature Request] Ship ParaView with a mangled VTK## Problem Statement
When including ParaView into applications that already compile and link to VTK, collisions between ParaView's VTK version and the application's can occur at runtime leading to a crash if the VTK versions are incompa...## Problem Statement
When including ParaView into applications that already compile and link to VTK, collisions between ParaView's VTK version and the application's can occur at runtime leading to a crash if the VTK versions are incompatible.
As an example, this can happen when using ParaView's `Catalyst` implementation in a code that already links to an external system VTK.
## Proposed Solution
Given that it is much more likely for another application that already links VTK to develop a dependency on ParaView than it is for ParaView to develop a dependency on another application using VTK, it seems only natural that it should be ParaView's responsibility to mangle its internal VTK libraries using the relatively new `VTK_ABI_NAMESPACE` feature developed in https://gitlab.kitware.com/vtk/vtk/-/merge_requests/8993 (discourse here: https://discourse.vtk.org/t/new-feature-in-vtk-vtk-abi-namespace/9421).
This functionality wraps all of VTK in a compile time defined namespace. The symbols are then "sealed" against collision.
Thoughts?https://gitlab.kitware.com/paraview/paraview/-/issues/21791Bounds from CGNS zone are based on all coordinates array, not on actual cells2023-03-16T03:27:19-04:00Pierre CrepierBounds from CGNS zone are based on all coordinates array, not on actual cellsHello,
First, my apologies for that strange title, I am not sure how I can put what I am seeing into words.
So here is the complete story, in less compact format:
We use the CGNS format as output of our CFD solver.
The files contain tw...Hello,
First, my apologies for that strange title, I am not sure how I can put what I am seeing into words.
So here is the complete story, in less compact format:
We use the CGNS format as output of our CFD solver.
The files contain two zones: Volume and Surfaces
These two zones contain arrays defining all vertices present in the zone.
This means that the Volume zone contains all vertices, and the Surfaces zone contains all boundary vertices.
Both zones can contain multiple sub-zone: various grids for the Volume zone, and the boundary cells for the surfaces.
To extract a specific boundary, I use the extract block filter, and choose the blocks I want.
The issue I am seeing now with version 5.11 is that if I extract only one zone, the bounds still correspond to all vertices, and not to the vertices shown on the screen.
This was not happening in 5.9
To reproduce:
1. Read the file, using the CGNS reader
2. extract the first block
3. look at the bounds: they are not what one expect it to be
So I have a feeling that the CGNS reader was changed and now only relies on the complete array of vertices to define the bounds instead of relying on the cells that are actually declared in the specific block.
I hope that makes the issue clearer. (and I attached an example file file too)
[cube.cgns](/uploads/fef0c968792bc24bf26243a26db568fd/cube.cgns)