ParaView issues
https://gitlab.kitware.com/paraview/paraview/-/issues
2021-03-24T05:38:08-04:00
https://gitlab.kitware.com/paraview/paraview/-/issues/19256
Loading the OptiX Ray Tracing backend fails silently in v5.7.0 RC2
2021-03-24T05:38:08-04:00
Nick Leaf
Loading the OptiX Ray Tracing backend fails silently in v5.7.0 RC2
If creating the VisRTX context fails (e.g. if VisRTX is not able to find the OptiX libraries), ray tracing will fail silently. Ray tracing can still be enabled via Python or the checkbox in the client, but the data will just disappear wi...
If creating the VisRTX context fails (e.g. if VisRTX is not able to find the OptiX libraries), ray tracing will fail silently. Ray tracing can still be enabled via Python or the checkbox in the client, but the data will just disappear with no warnings or errors. A warning/error message would let the user know that there is a problem with their environment configuration.
https://gitlab.kitware.com/paraview/paraview/-/issues/19257
View/ Lights - Move to Camera makes ParaView unstable
2024-03-07T20:47:09-05:00
W. Alan Scott
View/ Lights - Move to Camera makes ParaView unstable
Let me start by saying that to a light, "Move to Camera" means move the light to the current camera location. This bug assumes that "Move to Camera" is NOT really "Move Camera".
* Linux, 5.7.0-RC2, builtin server.
* Sources/ Cone. App...
Let me start by saying that to a light, "Move to Camera" means move the light to the current camera location. This bug assumes that "Move to Camera" is NOT really "Move Camera".
* Linux, 5.7.0-RC2, builtin server.
* Sources/ Cone. Apply.
* Not needed, but rotate counter clockwise 90 degrees.
* View/ Light Inspector.
* Add Light.
* Move To Camera.
ParaView is now really weird.
* The mouse has picked up a hand icon, rather than the default pointer.
* You are no longer moving the object with the left mouse button! This can be stopped by using the right mouse button, then the left.
This needs to be fixed.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19259
LineChartView does not support multipiece input
2021-03-24T05:38:00-04:00
Mathieu Westphal (Kitware)
LineChartView does not support multipiece input
LineChartView does not support multipiece input
* Open ParaView
* Open MultiBlockWithMultiPiece.vtm
* Filters -> PlotData, Apply
* Not Data can be shown
* Selection MultiBlockWithMultiPiece.vtm,
* Filter -> Merge Blocks, Apply
* ...
LineChartView does not support multipiece input
* Open ParaView
* Open MultiBlockWithMultiPiece.vtm
* Filters -> PlotData, Apply
* Not Data can be shown
* Selection MultiBlockWithMultiPiece.vtm,
* Filter -> Merge Blocks, Apply
* Filters -> PlotData, Apply
* Works fine
https://gitlab.kitware.com/paraview/paraview/-/issues/19260
Data redistribution filter using DIY
2024-03-07T20:47:09-05:00
Utkarsh Ayachit
Data redistribution filter using DIY
We need to create a new data redistritbution and replace existing usages with the new one.
Filter tasks
===========
* [x] generate global ids (point & cell) * <--- can merge at this point*
* [ ] replace clipping with VTK-m based filter...
We need to create a new data redistritbution and replace existing usages with the new one.
Filter tasks
===========
* [x] generate global ids (point & cell) * <--- can merge at this point*
* [ ] replace clipping with VTK-m based filter (ref: vtk/vtk!5840)
* [ ] tests/comparisons/performance testing (contrast with D3)
* [ ] exchange ghost cells.
ParaView Integration
====================
* [x] add new filter to ParaView / FIlters menu
* [ ] use new filter instead of D3 for parallel rendering use-cases
* [ ] regression tests etc.
5.13 (Summer 2024)
Caitlin Ross
Caitlin Ross
https://gitlab.kitware.com/paraview/paraview/-/issues/19261
Cannot render 1D vtkImageData in RenderView
2021-04-20T17:11:28-04:00
Andrew Bauer
Cannot render 1D vtkImageData in RenderView
Steps to reproduce:
1. Create a `Wavelet`
2. Set the `Whole Extent` to `[-10, 10, -10, -10, -10, -10]`
This results in a `Incorrect dimensionality` error. If you look at the `Wavelet` output in a Spreadsheet view though everything looks...
Steps to reproduce:
1. Create a `Wavelet`
2. Set the `Whole Extent` to `[-10, 10, -10, -10, -10, -10]`
This results in a `Incorrect dimensionality` error. If you look at the `Wavelet` output in a Spreadsheet view though everything looks fine.
https://gitlab.kitware.com/paraview/paraview/-/issues/19262
Strange behavior with the ResetLight button
2021-03-24T05:36:30-04:00
Mathieu Westphal (Kitware)
Strange behavior with the ResetLight button
ResetLight has some strange behavior that depend of the current status of the light.
How to reproduce :
* Open ParaView
* View -> Light Inspector
* Add Light
* Move Light around
* Reset Light
* Works perfectly
* Change to a Headl...
ResetLight has some strange behavior that depend of the current status of the light.
How to reproduce :
* Open ParaView
* View -> Light Inspector
* Add Light
* Move Light around
* Reset Light
* Works perfectly
* Change to a Headlight light
* Reset Light
* Light is not reset to expected position but to another position
@wascott : more lights weirdness
https://gitlab.kitware.com/paraview/paraview/-/issues/19269
Change LOD defaults
2019-11-08T03:48:27-05:00
W. Alan Scott
Change LOD defaults
The LOD defaults are crazy low. Please raise them. The LOD I am talking about are Edit/ Settings/ Rendering/ LOD Threshold. Please discuss with me, Ken, Cory and/or Utkarsh. Maybe raise to 100?
@sebastien.jourdain
The LOD defaults are crazy low. Please raise them. The LOD I am talking about are Edit/ Settings/ Rendering/ LOD Threshold. Please discuss with me, Ken, Cory and/or Utkarsh. Maybe raise to 100?
@sebastien.jourdain
https://gitlab.kitware.com/paraview/paraview/-/issues/19270
Camera Parallel Projection should have option to be sticky
2024-03-07T20:47:09-05:00
W. Alan Scott
Camera Parallel Projection should have option to be sticky
Please add the ability to make Camera Parallel Projection the default when running ParaView. Option defaults off. Probably put it in the settings. This is a sub issue to #19576
Please add the ability to make Camera Parallel Projection the default when running ParaView. Option defaults off. Probably put it in the settings. This is a sub issue to #19576
5.13 (Summer 2024)
Aron Helser
Aron Helser
https://gitlab.kitware.com/paraview/paraview/-/issues/19275
Make the MILI reader work on Windows OS
2024-03-07T20:47:09-05:00
Andrew Bauer
Make the MILI reader work on Windows OS
The MILI reader works on Linux and Mac but now that there's more time it needs to also work for Windows. This is a follow on to
#18689.
The MILI reader works on Linux and Mac but now that there's more time it needs to also work for Windows. This is a follow on to
#18689.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19281
ColorOpacityWidget has strange behavior with negative range and log
2021-03-24T05:48:04-04:00
Mathieu Westphal (Kitware)
ColorOpacityWidget has strange behavior with negative range and log
ColorOpacityWidget has strange behavior with negative range and log.
It change the range so it is compatible with log, that is expected, but then rescaling goes wrong.
Steps to reproduce :
* Open ParaView
* Wavelet, Apply
* Calculat...
ColorOpacityWidget has strange behavior with negative range and log.
It change the range so it is compatible with log, that is expected, but then rescaling goes wrong.
Steps to reproduce :
* Open ParaView
* Wavelet, Apply
* Calculator, RTData - 200, Apply
* Show "Results" array
* Edit Color Map
* The following warning appears, expected
```
Generic Warning: In /home/glow/work/paraview/paraviewFirst/paraview/Qt/ApplicationComponents/pqColorOpacityEditorWidget.cxx, line 709
Ranges not valid for log-space. Changed the range to (0.0126829, 126.829).
```
* Click on "Rescale to data range"
* The warning appears again, expected, but the results in the widget incorrect
![2019-08-28-130802_496x286_scrot](/uploads/178fd7d257dc0d7f426aee7bc1945c19/2019-08-28-130802_496x286_scrot.png)
https://gitlab.kitware.com/paraview/paraview/-/issues/19284
Install the ParaView client QCH file
2020-05-07T15:07:19-04:00
Ben Boeckel
Install the ParaView client QCH file
This seems to be useful. The QCH file should be installed and `paraview-config.cmake` should have a variable holding its path.
See: https://discourse.paraview.org/t/pv-custom-application-version-5-7-where-are-the-pv-plugins/2472/16
Cc:...
This seems to be useful. The QCH file should be installed and `paraview-config.cmake` should have a variable holding its path.
See: https://discourse.paraview.org/t/pv-custom-application-version-5-7-where-are-the-pv-plugins/2472/16
Cc: @utkarsh.ayachit
https://gitlab.kitware.com/paraview/paraview/-/issues/19287
Add a settings to select the default fonts to use in Chart and all related pl...
2020-11-26T08:54:12-05:00
Riccardo Rossi
Add a settings to select the default fonts to use in Chart and all related places
When configuring a charts, it can get pretty tedious to change the font of each element one by one.
In most usecases, only a single font is used in a single chart.
It would be nice to add a settings to set a default fonts for charts. I...
When configuring a charts, it can get pretty tedious to change the font of each element one by one.
In most usecases, only a single font is used in a single chart.
It would be nice to add a settings to set a default fonts for charts. It could be very easy to take the setting into account in
pqFontPropertyWidget.cxx.
https://gitlab.kitware.com/paraview/paraview/-/issues/19290
Add super title for render view
2020-12-24T13:31:22-05:00
Riley Brady
Add super title for render view
Porting this over from the suggestions thread on the [ParaView website](https://paraview.uservoice.com/forums/11350-general/suggestions/36543496-add-super-title-for-render-view) (just seeing the suggestion to move feature requests over h...
Porting this over from the suggestions thread on the [ParaView website](https://paraview.uservoice.com/forums/11350-general/suggestions/36543496-add-super-title-for-render-view) (just seeing the suggestion to move feature requests over here).
**Suggestion:**
For those using render view for small multiples (e.g., subplots of different months of seasons of climate data), it would be nice to be able to add one super title overlying all the individual views upon export/screenshotting.
This would be similar to `pyplot.suptitle` in python.
https://gitlab.kitware.com/paraview/paraview/-/issues/19298
SILO files not reading in time correctly
2024-03-07T20:47:09-05:00
W. Alan Scott
SILO files not reading in time correctly
We have a SILO dataset that is not loading time correctly. This works correctly with VisIt (3.0.1) and with EnSight. Here is how to replicate.
* 5.7.0 RC2, builtin server, Linux
* Load dataset EM_v5b_264. There will be 5 files to thi...
We have a SILO dataset that is not loading time correctly. This works correctly with VisIt (3.0.1) and with EnSight. Here is how to replicate.
* 5.7.0 RC2, builtin server, Linux
* Load dataset EM_v5b_264. There will be 5 files to this dataset, you load all 5. There are also lots of other files in this directory, they may be important also. Note that ParaView is incorrectly not identifying this dataset as Silo. Thus, you have to choose Files of type: All Files (*). Then, chose EM_v5b_264. Chose VisItSiloReader. OK. Don't turn any variables on, we are good to go. Apply.
* Move forward 1 timestep.
The time on this timestep is incorrect. It shows as 1 (both in the time controls, on the information tab and annotate time). It should be time of 503.453.
This is a bug.
Dataset is from an external customer. This customer will get dataset to Kitware. Note that customer can pass dataset to me, and I can then pass it to Kitware.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19299
ColorBar editor position settings has unexpected behavior
2021-03-24T05:47:04-04:00
Mathieu Westphal (Kitware)
ColorBar editor position settings has unexpected behavior
The ColorBar editor as a position settings. However, it does not behave as expected.
Steps to reproduce :
* Open paraview -dr
* Wavelet, apply
* Color by RTData
* Edit Color Map, Edit color legend properties
* Uncheck auto-orient ...
The ColorBar editor as a position settings. However, it does not behave as expected.
Steps to reproduce :
* Open paraview -dr
* Wavelet, apply
* Color by RTData
* Edit Color Map, Edit color legend properties
* Uncheck auto-orient and Select "Horizontal" (just to check it is working for certain settings)
* Set "Window Location" to Upper Right Corner, Apply
* **"Position" settings has not been updated and still shows "0.89" "0.2" when it should be "0.63" "0.91". Also it should be greyed out instead of uneditable**
* Set "Window Location" to Any Location, Apply
* Color bar does not move, expected
* **"Position" settings is still incorrect and still shows "0.89" "0.2" when it should be "0.63" "0.91".**
* Set Position to "0.63" "0.5"
* Apply, color bar is correctly placed
* Save Settings as default
* Create a new view,
* Show wavelet, color by RTData
* **ColorBar is incorectly placed, see screenshot**
* Edit Color Map, Edit color legend properties
* **Position is "0.89" "0.2" when it should be "0.63" "0.5".**
![2019-09-04-070832_1227x879_scrot](/uploads/02398c8d44d01edc62f731256213ddbf/2019-09-04-070832_1227x879_scrot.png)
@rrossi @cory.quammen
https://gitlab.kitware.com/paraview/paraview/-/issues/19300
Set and update on the fly default color map/color legend properties in the cu...
2021-03-24T05:46:46-04:00
Riccardo Rossi
Set and update on the fly default color map/color legend properties in the current view/session
In my understanding, using "Save current color map setting values as defaults for all arrays" in the "Color Map Editor" panel and "Save current settings values as default" in the "Editor Color Legend Properties" only applies to new rende...
In my understanding, using "Save current color map setting values as defaults for all arrays" in the "Color Map Editor" panel and "Save current settings values as default" in the "Editor Color Legend Properties" only applies to new render views, frames created by splitting the current render view or ultimately by closing and reopening Paraview.
In my opinion this is really impractical as it forces the user to set all the default settings before populating the view otherwise all the changes made to the visualization are lost. Also, any additional change to the new default settings would require again a new view and consequently loosing, again, the changes made.
It would be therefore much welcome to have an option to apply immediately the new default setting to all arrays.
https://gitlab.kitware.com/paraview/paraview/-/issues/19302
scaling of ChartViews is wrong when SaveAnimation sets ImageResolution
2024-03-07T20:47:09-05:00
Jean M. Favre
scaling of ChartViews is wrong when SaveAnimation sets ImageResolution
I find it totally frustrating that ParaView cannot save a complete layout, scaling each view appropriately when specifying an ImageResolution that is different than the screen layout. The two images attached say it all. One image was sav...
I find it totally frustrating that ParaView cannot save a complete layout, scaling each view appropriately when specifying an ImageResolution that is different than the screen layout. The two images attached say it all. One image was saved in batch mode, resulting in over-stretched legends, titles and axis, while the same can be done from a ParaView client after careful (and many iterations) setting of the overall screen size.
I use SaveAnimation('/tmp/foo.png', layout, ImageResolution=[1280, 800])
![image.0060](/uploads/a5f3aa1231ecdb65a5353cf69c74a6c6/image.0060.png)
![foo.0060](/uploads/ace0cce28501aa27ffb08a8e93044c7f/foo.0060.png)
any hint to make this full-proof would be greatly appreciated.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19308
Split render view in tiles of same size
2021-03-24T05:45:47-04:00
Riccardo Rossi
Split render view in tiles of same size
It would be very useful to have a command/option to split the render view in tiles of same width and/or height or control the size by input number or percentage of the main window.
It would be very useful to have a command/option to split the render view in tiles of same width and/or height or control the size by input number or percentage of the main window.
https://gitlab.kitware.com/paraview/paraview/-/issues/19313
Plugin auto-load option: "no entry in PV plugin manager"
2021-03-24T05:45:02-04:00
Cornelis Bockemühl
Plugin auto-load option: "no entry in PV plugin manager"
Own plugins can be written and directly integrated into a custom application. Loading them can be automatic with the REQUIRED_PLUGINS list of the paraview_client_add cmake function or inside the program using the vtkPVPluginTracker class...
Own plugins can be written and directly integrated into a custom application. Loading them can be automatic with the REQUIRED_PLUGINS list of the paraview_client_add cmake function or inside the program using the vtkPVPluginTracker class.
In all these cases, the plugins will appear in the list of the plugin manager - as "loaded".
But this does not make too much sense because loading them is not any more required, and unloading makes not much sense (and I would even doubt whether it can be reliable to "unload" a shared library from a running program!?)
Having an an option during plugin auto-load to not show a plugin in the PV plugin manager makes sense thus and avoids filling the list with useless entries.
https://gitlab.kitware.com/paraview/paraview/-/issues/19317
OptiX crashes paraview
2024-03-07T20:47:09-05:00
Mikkel Steenberg
OptiX crashes paraview
I just downloaded the newest version to see whether I could get the Optix path tracer to work on my new laptop with RTX, but Paraview keeps crashing when I turn Optix on.
Build: Paraview-5.7.0-RC3-Windows-Python3.7-msvc2015-64bit
Graphic...
I just downloaded the newest version to see whether I could get the Optix path tracer to work on my new laptop with RTX, but Paraview keeps crashing when I turn Optix on.
Build: Paraview-5.7.0-RC3-Windows-Python3.7-msvc2015-64bit
Graphic card: NVIDIA Quadro RTX 3000 (vers. 431.86)
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19338
ParaView crashing when specifying wrong extents for RAW ImageData
2021-03-24T05:58:44-04:00
Matthieu Dorier
ParaView crashing when specifying wrong extents for RAW ImageData
I'm using ParaView 5.6.2 on MacOSX.
I'm opening a RAW file containing a 3D array of integers (30x30x30 elements). If I specify extents such that the product would be larger than the data size (e.g. if I set [0,30]x[0,30]x[0,30] instead o...
I'm using ParaView 5.6.2 on MacOSX.
I'm opening a RAW file containing a 3D array of integers (30x30x30 elements). If I specify extents such that the product would be larger than the data size (e.g. if I set [0,30]x[0,30]x[0,30] instead of [0,29]x[0,29]x[0,29]), ParaView hangs when I click "apply" and I have to kill it.
https://gitlab.kitware.com/paraview/paraview/-/issues/19339
Assert fail when the legend takes too much space in a line chart view
2021-03-24T05:53:01-04:00
Mathieu Westphal (Kitware)
Assert fail when the legend takes too much space in a line chart view
If the legend takes too much spaces, the chart fails to draws on an assert and ParaView segfaults.
Steps to reproduce :
* Open ParaView
* Open SinceWaves.csv, Apply
* Show in a LineChartView
* Show all arrays
* place legend at the...
If the legend takes too much spaces, the chart fails to draws on an assert and ParaView segfaults.
Steps to reproduce :
* Open ParaView
* Open SinceWaves.csv, Apply
* Show in a LineChartView
* Show all arrays
* place legend at the bottom
* Hide last three arrays
* Increase legend font size to 14
* Show one more array
```
paraview: /home/glow/work/paraview/paraviewFirst/paraview/VTK/Rendering/Context2D/vtkContextClip.cxx:71: virtual void vtkContextClip::SetClip(float, float, float, float): Assertion `width >= 0 && height >= 0' failed.
Thread 1 "paraview" received signal SIGABRT, Aborted.
0x00007ffff7c08755 in raise () from /usr/lib/libc.so.6
(gdb) bt
#0 0x00007ffff7c08755 in raise () from /usr/lib/libc.so.6
#1 0x00007ffff7bf3851 in abort () from /usr/lib/libc.so.6
#2 0x00007ffff7bf3727 in __assert_fail_base.cold () from /usr/lib/libc.so.6
#3 0x00007ffff7c01026 in __assert_fail () from /usr/lib/libc.so.6
#4 0x00007ffff16c8b05 in vtkContextClip::SetClip (this=0x555559abcda0, x=20, y=574, width=242, height=-2) at /home/glow/work/paraview/paraviewFirst/paraview/VTK/Rendering/Context2D/vtkContextClip.cxx:71
```
https://gitlab.kitware.com/paraview/paraview/-/issues/19341
glTF exported streamlines ends up in wrong position
2021-03-24T05:52:49-04:00
Lars Olsson
glTF exported streamlines ends up in wrong position
When exporting streamlines (streamtracer + tubes and glyphs) to glTF in 5.7 RC4, the exported geometries ends up in the wrong position. The export worked correctly in RC3. Note: we're also exporting stl and they end up in the correct pos...
When exporting streamlines (streamtracer + tubes and glyphs) to glTF in 5.7 RC4, the exported geometries ends up in the wrong position. The export worked correctly in RC3. Note: we're also exporting stl and they end up in the correct position....so it seems to be streamline specific.
https://gitlab.kitware.com/paraview/paraview/-/issues/19342
Tetrahedralizer produces invalid/degenerate cells for certain types of input ...
2022-07-26T09:19:20-04:00
Eduardo Olivares
Tetrahedralizer produces invalid/degenerate cells for certain types of input data
cc @utkarsh.ayachit
It seems the tetrahedralizer filter, which is applied internally by ParaView for some volume representations, produces invalid cells depending on the cell type of input dataset. While debugging, we saw duplicated ce...
cc @utkarsh.ayachit
It seems the tetrahedralizer filter, which is applied internally by ParaView for some volume representations, produces invalid cells depending on the cell type of input dataset. While debugging, we saw duplicated cells and overlapping cells sharing edges.
I have attached 3 repro cases:
[repro.tar](/uploads/f4a87e402a898104c012ed4d5c7b935a/repro.tar)
When the volume is built using wedges and tets everything seems fine: "dummyGrid_tetra_and_wedge.vtu"...
but when building the same volume with polyhedrons instead, it results on an ill mesh: "dummyGrid_polyhedron.vtu".
IndeX volume representation was failing to render the resulting mesh and ParaView shows some artifacts for "volume" representation and heavy artifacts when using "Bunyk ray caster" that even becomes unstable and shows this error in the log (ParaView 5.7 RC3).
`"\bbd\ecd3383f\build\superbuild\paraview\src\VTK\Rendering\Volume\vtkUnstructuredGridBunykRayCastFunction.cxx, line 957 vtkUnstructuredGridBunykRayCastFunction (0000029D36277020): Degenerate topology - cell face used more than twice."`
There is a second case where this happens (multiblock_structure.vtm), when using structured data and applying the "MergeBlock" filter with "MergePoints" option enabled. If "MergePoints" is disabled then it renders, but with artifacts at the boundaries between blocks.
https://gitlab.kitware.com/paraview/paraview/-/issues/19345
Automatic selection of objects in pipeline browser when mouse-picked in rende...
2021-03-24T05:52:02-04:00
Riccardo Rossi
Automatic selection of objects in pipeline browser when mouse-picked in render view
Back in the 4.3.1 release, the objects in the pipeline browser where automatically selected when picked in the render view with the mouse.
For example, if you have two tiles in the render view and two objects active separately in each t...
Back in the 4.3.1 release, the objects in the pipeline browser where automatically selected when picked in the render view with the mouse.
For example, if you have two tiles in the render view and two objects active separately in each tile, by clicking one or the other they would be also selected in the pipeline.
In my understanding this feature is gone and now an extra click is required to perform the same operation, one to change the active tile and another one to select the object in the same tile.
If that's the case, would be nice to have it back or at least be able to activate such behavior.
https://gitlab.kitware.com/paraview/paraview/-/issues/19347
vtkPVGridAxes3DRepresentation: avoid calling MarkModified in SetPosition/SetS...
2021-03-24T05:51:10-04:00
Utkarsh Ayachit
vtkPVGridAxes3DRepresentation: avoid calling MarkModified in SetPosition/SetScale
MarkModified should not be called unless absolutely necessary since it causes pipeline updates. vtkPVGridAxes3DRepresentation can be easily re-worked to avoid that.
MarkModified should not be called unless absolutely necessary since it causes pipeline updates. vtkPVGridAxes3DRepresentation can be easily re-worked to avoid that.
https://gitlab.kitware.com/paraview/paraview/-/issues/19348
Add key control to toggle 3D widget visibility in filters that use them
2024-03-07T20:47:09-05:00
Cory Quammen
cory.quammen@kitware.com
Add key control to toggle 3D widget visibility in filters that use them
> when you create a slice [...] and then try to move the object around, sometimes you end up selecting the widget and modifying the slice.
>
> It would be useful, in my opinion, to have a key (shift or alt) to temporarily and on the fly ...
> when you create a slice [...] and then try to move the object around, sometimes you end up selecting the widget and modifying the slice.
>
> It would be useful, in my opinion, to have a key (shift or alt) to temporarily and on the fly disable the widget without having to go back to the property panel and switch the widget off, and that could also help with the scenario you described.
> Maybe add another check box, right below "Show Plane"? Maybe "Interactive Plane"? Then, add a hot key toggle?
From discussion in #19345.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19355
Volume Rendering with multiple dataset does not handle depth correctly
2021-03-24T05:50:43-04:00
Mathieu Westphal (Kitware)
Volume Rendering with multiple dataset does not handle depth correctly
When using Volume Rendering in ParaView with multiple datasets, depths is not managed correctly, even with the "Depth Peeling For Volumes" setting enabled.
Steps to reproduce :
* Open ParaView
* Wavelet, Apply
* Volume
* Wavelet, 0 20,...
When using Volume Rendering in ParaView with multiple datasets, depths is not managed correctly, even with the "Depth Peeling For Volumes" setting enabled.
Steps to reproduce :
* Open ParaView
* Wavelet, Apply
* Volume
* Wavelet, 0 20, 0, 20, Apply
* Volume
* Move the camera around
![Peek_2019-10-02_10-57](/uploads/9fee4055dce90d79f5472b8350d56883/Peek_2019-10-02_10-57.gif)
@ken-martin @michael.migliore
https://gitlab.kitware.com/paraview/paraview/-/issues/19366
RayTracing produce image in image when saving animation in higher resolution
2021-08-16T15:39:41-04:00
Mathieu Westphal (Kitware)
RayTracing produce image in image when saving animation in higher resolution
Saving an animation with higher resolution and ray tracing can result in the following image
![actualTest.0018](/uploads/573f8868e0a8a82270eeb934abb9831f/actualTest.0018.png)
Steps to reproduce :
* Open can.ex2
* Enable Ray Tracing
* E...
Saving an animation with higher resolution and ray tracing can result in the following image
![actualTest.0018](/uploads/573f8868e0a8a82270eeb934abb9831f/actualTest.0018.png)
Steps to reproduce :
* Open can.ex2
* Enable Ray Tracing
* Enable shadows
* Change to OSPRay pathtracer
* 20 samples per pixel
* Save animation, as .png
* Change resolution to 1920 x 1080
* one or two image may look like this. Resave animation if not.
David E. DeMarle
David E. DeMarle
https://gitlab.kitware.com/paraview/paraview/-/issues/19368
disallow filter changes when playing animations
2021-03-24T05:50:09-04:00
Utkarsh Ayachit
disallow filter changes when playing animations
* Load `can.ex2`.
* Add a **Clip** filter, defaults are fine.
* Run animation. Loop it so it repeats indefinitely.
* Now modify the clip filter with the widget in the Render View while the animation is running. Apply.
* Notice that the c...
* Load `can.ex2`.
* Add a **Clip** filter, defaults are fine.
* Run animation. Loop it so it repeats indefinitely.
* Now modify the clip filter with the widget in the Render View while the animation is running. Apply.
* Notice that the clip has not been modified and the animation is running from the cached timesteps.
* Now stop the animation. Change the plane again and click Apply. The geometry is updated.
* Run the animation again. The geometry is updated correctly the first time through.
To support this case, it may be best to disallow changes while animations are playing except camera changes, play, pause and such.
https://gitlab.kitware.com/paraview/paraview/-/issues/19369
Add crinkle slice filter for hyper tree grid
2019-10-10T11:17:25-04:00
Yohann Bearzi (Kitware)
Add crinkle slice filter for hyper tree grid
The crinkle option does not exist for the slice filter of hyper tree grids because no such filter is implemented. Such a filter should be written and added to the features of the slice widget in ParaView.
The crinkle option does not exist for the slice filter of hyper tree grids because no such filter is implemented. Such a filter should be written and added to the features of the slice widget in ParaView.
Yohann Bearzi (Kitware)
Yohann Bearzi (Kitware)
https://gitlab.kitware.com/paraview/paraview/-/issues/19372
Erroneous Material on X3D and VRML output of Lines
2023-06-23T06:09:30-04:00
Nicholas Polys
Erroneous Material on X3D and VRML output of Lines
We LOVE the X3D and VRML export!! :-)
but we found a bug... stil in 5.6.x :
the ISO specs say that lines are not lit/shaded, so diffuseColor is not used.
The RGB Material should be output as emissiveColor
for LineSets and PointSet...
We LOVE the X3D and VRML export!! :-)
but we found a bug... stil in 5.6.x :
the ISO specs say that lines are not lit/shaded, so diffuseColor is not used.
The RGB Material should be output as emissiveColor
for LineSets and PointSets
https://www.web3d.org/standards
https://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/components/rendering.html#LineSet
https://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/components/shape.html#Material
https://gitlab.kitware.com/paraview/paraview/-/issues/19377
legend font size settings not respected when saving animation for split-view ...
2021-04-07T11:04:23-04:00
Paul Navrátil
legend font size settings not respected when saving animation for split-view layout
Hello,
I stumbled on a corner case in the Save Animation feature that I at least wanted to put on the radar. If you change the font size for a legend in a split-view layout and then use Save Animation and click "Save All Views", the leg...
Hello,
I stumbled on a corner case in the Save Animation feature that I at least wanted to put on the radar. If you change the font size for a legend in a split-view layout and then use Save Animation and click "Save All Views", the legend font size resets to default (16 point). If you only save one view, the legend font size settings are kept. It appears to be related to the window adjustments that ParaView makes in order to combine the views for the final render.
I've attached a screen-capture showing the legends (set to Arial 12 for title and Arial 8 for text/annotation) and an output frame from the animation showing the larger font results.
Thanks!
![Screen_Shot_2019-10-09_at_3.23.26_PM](/uploads/7b238b6786bbe23de5845c4b88733794/Screen_Shot_2019-10-09_at_3.23.26_PM.png)
![3d_diptych.0047](/uploads/f92ef95846c6edb5ed0db9b1e5642c1d/3d_diptych.0047.png)
https://gitlab.kitware.com/paraview/paraview/-/issues/19382
Eye Dome Lighting is different with Axes Grid on
2024-03-07T20:47:10-05:00
Boonthanome Nouanesengsy
Eye Dome Lighting is different with Axes Grid on
The Eye Dome Lighting view is shading differently depending on if the Axes Grid is on or off, and behavior is also different if using Paraview locally or with client-server. With Axes Grid on, the shading is lighter than when Axes Grid i...
The Eye Dome Lighting view is shading differently depending on if the Axes Grid is on or off, and behavior is also different if using Paraview locally or with client-server. With Axes Grid on, the shading is lighter than when Axes Grid is off.
When using Paraview locally, the shading is identical with Axes Grid on or off. But if Polar Axes is toggled, then the shading looks different.
When using Paraview in client-server mode, the shading is different when toggling Axes Grid and Polar Axes.
Images of toggling Axes Grid in client-server mode. It would be ideal of the lighting was always like the lighter version, looking like with the Axes Grid turned on.
![client-server-no-axes-small](/uploads/91a2601a1e7794c96c38242149b1da24/client-server-no-axes-small.png)![client-server-axes-small](/uploads/4b72b2893ae57d529b06914e8663f3b5/client-server-axes-small.png)
I have tested the client-server behavior with Paraview 5.5.2, as that's the most current version we support for client-server connections. I've tested the local behavior with the head of the git master from about a week ago.
Unfortunately the data is too large to post, but if anyone would like it I can send it. It's about 150 MB compressed.
How to reproduce:
- load the Eye Dome Lighting plugin
- create a new view of type Eye Dome Lighting
- load the dataset
- in the Properties Tab, change the Point Size to 3
- toggle the Data Axes Grid or Polar Axes checkbox on and off
5.13 (Summer 2024)
Aron Helser
Aron Helser
https://gitlab.kitware.com/paraview/paraview/-/issues/19385
StreamTracerWithCustomSource of ParaView 5.7 with SurfaceStreamlines = 1 goes...
2022-08-22T11:07:19-04:00
Emanuele Leggio
StreamTracerWithCustomSource of ParaView 5.7 with SurfaceStreamlines = 1 goes up to 20 times slower than previous versions of ParaView
I conducted many tests with ParaView 5.7 and I noticed that by activating the SurfaceStreamlines in StreamTracerWithCustomSource filter the processing times have increased a lot.
To prove this claim I considered 3 cases with increasing...
I conducted many tests with ParaView 5.7 and I noticed that by activating the SurfaceStreamlines in StreamTracerWithCustomSource filter the processing times have increased a lot.
To prove this claim I considered 3 cases with increasing numbers of cells, all surfaces are with triangular mesh, seed point come from the surfaces.
| case | seed point | cell |
| :-----: | :-----: | :-----: |
| A | 1260 | 688294 |
| B | 11044 | 5531204 |
| C | 92947 | 40727434 |
I have tested 4 pre-builded versions of ParaView and two SuperBuild:
Pre-build:
* 5.3.0
* 5.4.1
* 5.5.2
* 5.7.0
SuperBuild:
* 5.6.0 (PV5.6SB)
* 5.7.0 (PV5.7SB)
I tested both the case with SurfaceStreamlines = 1 and SurfaceStreamlines = 0.
Each test was repeated 3 times in order to average the times; only with PV5.7SB I did only one run for each test. A little less than 100 runs in total.
To complete the analysis I report the number of points and cells of the final result in addition to the processing times.
Obviously all the tests were conducted on the same machine with the same operating system and in a serial manner without other processes.
RESULTS:
![results1](/uploads/6c1f2899eadfa1ddd2994f55b2a0167d/results1.png)
in the attachment you can find the two simple scripts I used to do this analysis and all the intermediate results logs.
[execute_streamTraceWCS.py](/uploads/b6ea9703fcb78bd4cd8ff13a631fba39/execute_streamTraceWCS.py)
[main_launch.py](/uploads/0a0d86374565e7b8aeb300f9f78b2425/main_launch.py)
[results.tar.gz](/uploads/f3a762a2546bfad325cb50a01d2c3d98/results.tar.gz)
https://gitlab.kitware.com/paraview/paraview/-/issues/19393
HTG Clip filter cannot invert on 'Plane', 'Sphere', 'Cylinder', 'Quadric', 'Box'
2024-03-07T20:47:10-05:00
Yohann Bearzi (Kitware)
HTG Clip filter cannot invert on 'Plane', 'Sphere', 'Cylinder', 'Quadric', 'Box'
The invert flag in `vtkHyperTreeGridAxisClip` filter is only taken into account in the axis aligned plane version. The algorithm needs to be changed to add this functionality for all methods.
The invert flag in `vtkHyperTreeGridAxisClip` filter is only taken into account in the axis aligned plane version. The algorithm needs to be changed to add this functionality for all methods.
5.13 (Summer 2024)
Yohann Bearzi (Kitware)
Yohann Bearzi (Kitware)
https://gitlab.kitware.com/paraview/paraview/-/issues/19398
Lines are bleading through thin objects, Sufrace with Edges, Parallel
2024-03-07T20:47:10-05:00
W. Alan Scott
Lines are bleading through thin objects, Sufrace with Edges, Parallel
We are having lines bleeding through thin objects, such as the seat of a chair, parallel projection, with Surface with Edges. This was first reported here: https://discourse.paraview.org/t/surface-with-edges-representation-in-5-7/2769. ...
We are having lines bleeding through thin objects, such as the seat of a chair, parallel projection, with Surface with Edges. This was first reported here: https://discourse.paraview.org/t/surface-with-edges-representation-in-5-7/2769. Dataset in this bug works well (chair.obj), but here is another way to replicate:
* Linux, 5.7.0, builtin server.
* Load disk_out_ref.exo. Apply.
* Slice filter. Apply.
* Linear Extrusion filter. Vector == 1, 0, 0. Scale Factor == 0.2. Apply.
* Projects/ Camera Parallel Projection.
* Surface with Edges.
Now, by varying Scale Factor, you can see the lines on the back of the object come and go. 0.3 is worse, 0.4 is much better, but we have bleeding along the edges.
Somehow or another, in parallel projection, our coincident line code is WAY too extreme.
I checked, and Linux, 5.6.0, builtin server did NOT have this issue.
OK to slip to 5.9.0.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19401
ospray amr bug
2024-03-07T20:47:11-05:00
Carson Brownlee
ospray amr bug
switching on and off ospray rendering with an amr volume appears to crash in some cases. Reproducible with amr guassian source. @demarle
switching on and off ospray rendering with an amr volume appears to crash in some cases. Reproducible with amr guassian source. @demarle
5.13 (Summer 2024)
David E. DeMarle
David E. DeMarle
https://gitlab.kitware.com/paraview/paraview/-/issues/19405
Hyper Tree Grid documentation
2021-11-11T14:19:58-05:00
Yohann Bearzi (Kitware)
Hyper Tree Grid documentation
The Hyper Tree Grid documentation is extremely poor, and sometimes in French... At some point a large sweep needs to be done to make it easier for users.
The Hyper Tree Grid documentation is extremely poor, and sometimes in French... At some point a large sweep needs to be done to make it easier for users.
Yohann Bearzi (Kitware)
Yohann Bearzi (Kitware)
https://gitlab.kitware.com/paraview/paraview/-/issues/19407
probe isn't working with ctrl p
2021-03-24T06:06:57-04:00
W. Alan Scott
probe isn't working with ctrl p
The <ctrl>p option for probe isnt' working. Here is how to replicate:
* Sources/ FastUniformGrid. Apply. Surface with edges.
* Turn Auto Apply On. (I don't think this step is needed.)
* Probe. Now, <ctrl>-p. Nothing happens.
The <ctrl>p option for probe isnt' working. Here is how to replicate:
* Sources/ FastUniformGrid. Apply. Surface with edges.
* Turn Auto Apply On. (I don't think this step is needed.)
* Probe. Now, <ctrl>-p. Nothing happens.
https://gitlab.kitware.com/paraview/paraview/-/issues/19408
Selecting subset of variables with XDMFReader does not work with series of files
2021-03-24T06:06:38-04:00
Alexander Chadwick
Selecting subset of variables with XDMFReader does not work with series of files
If I try to read in the entire series of XDMF files with the XDMF Reader (i.e., XDMF2 instead of XDMF3), if I select only some of the variables in the dataset, upon clicking apply ParaView ignores the unchecked variables and still appear...
If I try to read in the entire series of XDMF files with the XDMF Reader (i.e., XDMF2 instead of XDMF3), if I select only some of the variables in the dataset, upon clicking apply ParaView ignores the unchecked variables and still appears to read in all of the variables. The issue is minor (having to switch variables I'm coloring by, having to hide extra colorbars when scripting) to severe (massive memory usage with larger datasets). It does not occur if I read in only a single XDMF file at a time or if I use XDFM3T (latter does not have as good parallel performance).
I encounter this issue in the official ParView 5.7.0 binaries for Mac installed via the .pkg distribution on a 2013 Mac Pro running Mojave 10.14.6. The test dataset contains four variables that were written to individual HDF5 archives for each process. The whole dataset is then stitched using the HDF5 virtual dataset feature, with that HDF5 archive being referenced in the XDMF header.
To reproduce:
1. Open the entire iter_*.xdmf set of files (two total)
2. Select XDMF Reader
3. In Cell Arrays, uncheck "ArrayIDontWant1" and "VectorArrayIDontWant1" and click apply
4. Click on coloring menu to change from "Solid Color." Unchecked variables are somehow still visible
Attached: [xdmf_test.zip](/uploads/0cef498a44ce2ec538e6dbe1c281e339/xdmf_test.zip)
https://gitlab.kitware.com/paraview/paraview/-/issues/19410
Spreadsheet should split "Structured Coordinates" in one column per components
2021-03-24T06:06:31-04:00
Nicolas Vuaille
Spreadsheet should split "Structured Coordinates" in one column per components
The artificially-added "Structured Coordinates" array is displayed with its 3 components in a single column instead of 3.
example:
* source / Fast Uniform Grid, apply
* new view, Spreadsheet View
![3compArray](/uploads/5b0f19b4b81c48...
The artificially-added "Structured Coordinates" array is displayed with its 3 components in a single column instead of 3.
example:
* source / Fast Uniform Grid, apply
* new view, Spreadsheet View
![3compArray](/uploads/5b0f19b4b81c48c4cfdcab1eb614812a/3compArray.png)
https://gitlab.kitware.com/paraview/paraview/-/issues/19411
include meshio for more mesh formats?
2021-11-05T05:53:12-04:00
Nico Schlömer
include meshio for more mesh formats?
Some years ago, I've created [meshio](https://github.com/nschloe/meshio), mostly for the purpose of being able to convert various mesh formats to something that ParaView can read, so I can look at it.
It has matured quite a bit since th...
Some years ago, I've created [meshio](https://github.com/nschloe/meshio), mostly for the purpose of being able to convert various mesh formats to something that ParaView can read, so I can look at it.
It has matured quite a bit since then, and recently, somebody even produced [a ParaView plugin for meshio](https://github.com/tianyikillua/paraview-meshio-reader) which allows opening all those formats without having to convert them first.
I'm wondering if this wouldn't be something to include into ParaView proper.
https://gitlab.kitware.com/paraview/paraview/-/issues/19422
Add ability for Windows to run Mesa when hardware is insufficient
2019-10-29T20:30:07-04:00
W. Alan Scott
Add ability for Windows to run Mesa when hardware is insufficient
Please add the ability for Windows versions of ParaView to run software Mesa graphics when the hardware or drivers are insufficient. Here is a good writeup on discourse here:
https://discourse.paraview.org/t/opengl-not-supported-in-w...
Please add the ability for Windows versions of ParaView to run software Mesa graphics when the hardware or drivers are insufficient. Here is a good writeup on discourse here:
https://discourse.paraview.org/t/opengl-not-supported-in-windows-10-and-intel-hd-graphics-card-glew-could-not-be-intialized-missing-gl-version/2701/3
https://gitlab.kitware.com/paraview/paraview/-/issues/19424
Crash on startup inside OpenImageDenoise on EC2
2024-03-07T20:47:11-05:00
David E. DeMarle
Crash on startup inside OpenImageDenoise on EC2
As reported on [discourse](https://discourse.paraview.org/t/floating-point-exception-since-paraview-5-7-0-using-nomachine/2863), ParaView 5.7.0 crashes on startup on EC2 with NoMachine display.
CPU instruction set detection within OpenI...
As reported on [discourse](https://discourse.paraview.org/t/floating-point-exception-since-paraview-5-7-0-using-nomachine/2863), ParaView 5.7.0 crashes on startup on EC2 with NoMachine display.
CPU instruction set detection within OpenImageDenoise is the most likely suspect.
We should attempt a fix for 5.7.1. Failing that it should be fixed 5.8.0, which is coming very soon.
Upstream issue reported here: https://github.com/OpenImageDenoise/oidn/issues/43
5.13 (Summer 2024)
David E. DeMarle
David E. DeMarle
https://gitlab.kitware.com/paraview/paraview/-/issues/19433
Incorrect second order Lagrange Hexahedral order
2021-03-24T06:21:55-04:00
Jørgen Schartum Dokken
Incorrect second order Lagrange Hexahedral order
Hi. In [Release notes on higher order Lagrangian cells](https://blog.kitware.com/modeling-arbitrary-order-lagrange-finite-elements-in-the-visualization-toolkit/), there is an image illustrating the ordering of a Lagrange order 3 hexahedr...
Hi. In [Release notes on higher order Lagrangian cells](https://blog.kitware.com/modeling-arbitrary-order-lagrange-finite-elements-in-the-visualization-toolkit/), there is an image illustrating the ordering of a Lagrange order 3 hexahedron element. The number of the hexahedron element does not correspond to the ordering used for Lagrange order 2 hexahedron elements. I've attached an example where the points are ordered as one would expect by looking at the Lagrange order 3 example:
`<!-- <DataArray type="UInt32" Name="connectivity" format="ascii">0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 </DataArray> -->`
1. Vertices at z=0 counter clockwise
2. Vertices at z=1 counter clockwise
3. Edges at z=0 counter clockwise (starting with the one between vertex 0 and 1)
4. Edges at z=1 counter clockwise (starting with the one between vertex 4 and 5)
5. Edges at z=0.5 counter clockwise
6. Faces, since we are considering CG2 there is only one dof per faces. Describing them with planes: (XZ, Y=0), (XZ, Y=1), (YZ, X=0), (YZ, X=1), (XY, Z=0), (XY, Z=1)
7. One internal node.
This ordering does not yield the correct connectivity visually.
The correct ordering would be:
`<!-- <DataArray type="UInt32" Name="connectivity" format="ascii">0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 </DataArray> -->`
This implies that step 5 is wrong. This should be a lexicographical ordering.
Similarly, step 6 is wrong.
I've attached a mesh with the ordering that's working visually in Paraview 5.7.0-RC4.
[CG2_hex.pvd](/uploads/99ae88ee02ef9b3a8468bdad8c23d040/CG2_hex.pvd)
[CG2_hex.vtu](/uploads/36d869931fd5993de29591844919663d/CG2_hex.vtu)
https://gitlab.kitware.com/paraview/paraview/-/issues/19438
Ensight reader cause segfault using ghost cells
2019-11-26T04:10:14-05:00
Håkon Strandenes
h.strandenes@km-turbulenz.no
Ensight reader cause segfault using ghost cells
I'm unsure if this bug belong in the Paraview or VTK bugtracker, I can re-file it in the VTK bug tracker if needed.
Using the ghost cell feature of the Ensight file format, which is analogue to the similar feature in VTK/Paraview, Para...
I'm unsure if this bug belong in the Paraview or VTK bugtracker, I can re-file it in the VTK bug tracker if needed.
Using the ghost cell feature of the Ensight file format, which is analogue to the similar feature in VTK/Paraview, Paraview segfault/crash. This is happening on recent versions, tested both Paraview 5.5.2 and 5.7.0.
If the 'ghost_flags' block of the Ensight geo-file is removed, Paraview is able to open the case. The attached case consist of three parts, and the corresponding 'ghost_flags' blocks of the three parts start at line 121, 46894 and 93667 in newcase.geo. The 'ens_checker' tool provided together with Ensight to validate the Ensight Gold format give no warnings on the attached case.
Ideally Paraview/VTK should read the content of the 'ghost_flags' and put it in a vtkGhostType-array for proper rendering.
The attached files demonstrate the problem.
[newcase.case](/uploads/4814d4c71dc5e799fe4eeaff1e464b84/newcase.case)
[newcase.geo](/uploads/da450245a9dc255749ceb7ab76921bb3/newcase.geo)
[P.ensdat](/uploads/e44a16fe45c7bfc3c634c52f74933330/P.ensdat)
https://gitlab.kitware.com/paraview/paraview/-/issues/19439
State Files loading options are not used when loading them back through Recen...
2021-03-24T05:59:32-04:00
Joachim Pouderoux
State Files loading options are not used when loading them back through Recent Files menu
When loading a state file, one can choose some file loading options. Unfortunately, those options are not kept in memory and used back when we want to load the state file again using the quick Recent Files menu entry - default options ar...
When loading a state file, one can choose some file loading options. Unfortunately, those options are not kept in memory and used back when we want to load the state file again using the quick Recent Files menu entry - default options are used and thus loading can fail.
https://gitlab.kitware.com/paraview/paraview/-/issues/19444
LOD Resolution has incorrect behavior in ParaView
2021-03-24T05:59:49-04:00
Mathieu Westphal (Kitware)
LOD Resolution has incorrect behavior in ParaView
LOD Resolution does not behave correctly in ParaView 5.7.0 and up (compared to 5.6).
It is probably linked to !3430 !3423 !3358.
Steps to reproduce :
* Open ParaView
* Open [cow.vtp](/uploads/8d2fc25526b27fed7ddef051a9895ef9/cow.vtp)...
LOD Resolution does not behave correctly in ParaView 5.7.0 and up (compared to 5.6).
It is probably linked to !3430 !3423 !3358.
Steps to reproduce :
* Open ParaView
* Open [cow.vtp](/uploads/8d2fc25526b27fed7ddef051a9895ef9/cow.vtp), Apply
* Surface with edges
* Edit -> Settings -> LODTreshold -> 0 -> Apply
* Move camera around, LOD is triggered correctly
* LOD Resolution -> 0
* Move camera around, **LOD is triggered but the resolution has not changed**
* LOD Resolution -> 1
* Move camera around, **LOD is triggered but the resolution has not changed**
* LOD Resolution -> 0
* Triangulate, Apply (this will cause the VTK-m LOD to be used)
* **Move camera around, LOD is triggered but is not as strong as before**
* LOD Resolution -> 0
* **Move camera around, LOD seems to be stronger but definitelly not as strong as expected**
TLDR :
* LOD resolution does not work anymore for the general case
* LOD resolution works with VTK-m LOD, but is generally way less strong than the general case
https://gitlab.kitware.com/paraview/paraview/-/issues/19446
ColorBy could be simpler to call
2020-05-04T22:51:43-04:00
Mathieu Westphal (Kitware)
ColorBy could be simpler to call
ColorBy, the paraview.simple method to easily color meshes is hard to call for simple usecase :
* Open ParaView
* View -> Python Shell
* Source -> Sphere, Apply
* `ColorBy(GetDisplayProperties(), 'Normals')` **does not work**
* `C...
ColorBy, the paraview.simple method to easily color meshes is hard to call for simple usecase :
* Open ParaView
* View -> Python Shell
* Source -> Sphere, Apply
* `ColorBy(GetDisplayProperties(), 'Normals')` **does not work**
* `ColorBy(GetDisplayProperties(), ('POINTS','Normals'))` is the minimal call to color using ColorBy.
https://gitlab.kitware.com/paraview/paraview/-/issues/19449
building ParaView/VTK static with Python enabled still builds mpi4py shared
2022-10-17T12:38:32-04:00
Utkarsh Ayachit
building ParaView/VTK static with Python enabled still builds mpi4py shared
Building ParaView/VTK static with Python enabled still builds mpi4py shared; we will need to build mpi4py statically too to support `zipimport`ing of ParaView/VTK packages.
cc: @ben.boeckel @cory.quammen
Building ParaView/VTK static with Python enabled still builds mpi4py shared; we will need to build mpi4py statically too to support `zipimport`ing of ParaView/VTK packages.
cc: @ben.boeckel @cory.quammen
https://gitlab.kitware.com/paraview/paraview/-/issues/19455
Smarter Python traces
2024-03-07T20:47:11-05:00
W. Alan Scott
Smarter Python traces
I worked with a user today trying to get scripts ready to process hundreds of large exodus runs. This was a time consuming and detailed endeavor. I would like to propose the following:
Lets add code, commented out, to all Python scrip...
I worked with a user today trying to get scripts ready to process hundreds of large exodus runs. This was a time consuming and detailed endeavor. I would like to propose the following:
Lets add code, commented out, to all Python scripts that trivially allow the Python to be controled through parameters. In other words, something like this:
```
# The following code can be uncommented, and used with parameters.
# You would then use this code as follows to batch process lots of directories:
# ../pvbatch thisBatchFile directoryIn1 fileIn1 directoryOut1 fileOut1
# ../pvbatch thisBatchFile directoryIn2 fileIn2 directoryOut2 fileOut2
# ../pvbatch thisBatchFile directoryIn3 fileIn3 directoryOut3 fileOut3
# Uncomment the following lines to use parametres, and comment out the exodusReader0.
# inputDirectoryName = argument[1]
# inputFileName = argument[2]
# outputDirectoryName = argument[3]
# outputFileName = argument[4]
# exodusReader0 = ExodusReader([inputDirectoryName+inputFileName+'.e.1234.0000'])
exodusReader0 = ExodusReader('/home/alan/data/pizzaInput/myfile.e.1234.0000'])
do really important and complex stuff
# SaveScreenshot(outputDirectoryName+outputFileName+'.png')
SaveScreenshot(/home/alan/data/pizzaOutput/myfile.png')
```
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19471
Importing time series after loading a state in 5.4.1
2021-03-24T06:40:26-04:00
Javier Diaz
Importing time series after loading a state in 5.4.1
In the past, I had been using Paraview (version 5.0.1 64-bit in Ubuntu 16.04) to analyse simulation results. In the simplest case I have a (x,y,z) file with simply 3 column vectors separated by a space, with N lines for N particles to pl...
In the past, I had been using Paraview (version 5.0.1 64-bit in Ubuntu 16.04) to analyse simulation results. In the simplest case I have a (x,y,z) file with simply 3 column vectors separated by a space, with N lines for N particles to plot. Then, I use TabletoPoints and then Glyph to have a 3D view of N particles. My data is saved in simple text files with the format “col*.txt” where * is the time step like 001 ,002…100 (the actual numbering is not sequencial).
After saving the state, I was able to load the state again and apply all my filters into different time series files. Therefore, I was able to see the time evolution of the files using one single state file.
Now, I am using Paraview 5.4.1 64-bit in Ubuntu 18.4. With the new 3-options version, I try to do the same by choosing the "Chose File Names " and selecting all the files with the correct format. But I don’t get the time series, instead, only the first time step.
I have attached the minimal example I mention. It has been produced in ParaView 5.4.1 using Ubuntu 18.04.03[minimal-state-in-PV-5.4.1-Ubuntu18.04.03](/uploads/a4ff7eb10f50df90161786fa7218bb48/minimal-state-in-PV-5.4.1-Ubuntu18.04.03)
PS: Issue submission recommend in the Paraview Forum [here](https://discourse.paraview.org/t/importing-time-series-after-loading-a-state-in-5-4-1/2983)
https://gitlab.kitware.com/paraview/paraview/-/issues/19472
Python or xml-only plugins not supported by distributed plugin search mechanism
2021-03-24T06:40:01-04:00
Utkarsh Ayachit
Python or xml-only plugins not supported by distributed plugin search mechanism
Plugins distributed by ParaView currently have to be compiled and cannot be Python-only or XML-only. This is because the lookup for plugins based on name only looks for plugins with shared library suffixes.
Note, such plugins should be ...
Plugins distributed by ParaView currently have to be compiled and cannot be Python-only or XML-only. This is because the lookup for plugins based on name only looks for plugins with shared library suffixes.
Note, such plugins should be supported in static builds too.
https://gitlab.kitware.com/paraview/paraview/-/issues/19475
vtkPVCompositeDataInformation does not have a GetFlatIndex method
2020-05-04T22:49:08-04:00
Mathieu Westphal (Kitware)
vtkPVCompositeDataInformation does not have a GetFlatIndex method
vtkPVCompositeDataInformation is the standard way to get information about a multiblock dataset on the server.
However it does not have a GetFlatIndex method. It would be quite practical to have access to this information.
vtkPVCompositeDataInformation is the standard way to get information about a multiblock dataset on the server.
However it does not have a GetFlatIndex method. It would be quite practical to have access to this information.
https://gitlab.kitware.com/paraview/paraview/-/issues/19476
Clicking on the eye to show a source while their is changes to apply should t...
2021-06-01T08:25:28-04:00
Mathieu Westphal (Kitware)
Clicking on the eye to show a source while their is changes to apply should trigger an Apply
Consider the following steps :
* Open any file, do not apply, let it highlighted
* Click on the eye in the pipeline browser
* Source appears in the render view, but apply is still highlighted, even though the pipeline have been exec...
Consider the following steps :
* Open any file, do not apply, let it highlighted
* Click on the eye in the pipeline browser
* Source appears in the render view, but apply is still highlighted, even though the pipeline have been executed
* Try to add a filter, none can be added has Apply has not been pressed yet
* Press Apply, all is working again
Many (I would say half of the attendees in ParaView basic course) beginners users ends up doing this workflow and get frustrated.
Moreover, sometimes the Properties Panel is voluntarily hidden, so it would make sense to be able to "Apply" from the pipeline browser.
Thoughts @utkarsh.ayachit ?
https://gitlab.kitware.com/paraview/paraview/-/issues/19484
Crystal Eyes stereo rendering needs to be configured at launch and on both th...
2023-10-10T11:09:58-04:00
Mathieu Westphal (Kitware)
Crystal Eyes stereo rendering needs to be configured at launch and on both the client and the server for CAVE
When rendering in a stereo cave, one needs to specify the `--stereo --stereo-type="Crystal Eyes"` option on both the client and server.
It is not supported to :
- Specify only on the client and the the "Same as client" option
- Specify...
When rendering in a stereo cave, one needs to specify the `--stereo --stereo-type="Crystal Eyes"` option on both the client and server.
It is not supported to :
- Specify only on the client and the the "Same as client" option
- Specify on the server and not on the client
- Specify another type of stereo at startup (either client or server) and then change to Crystal Eyes
https://gitlab.kitware.com/paraview/paraview/-/issues/19489
Orientation axis control in python
2021-03-24T06:39:27-04:00
Hongyang Zhou
Orientation axis control in python
Currently only the following can be controlled with pvpython for the orientation axes:
```
r=GetActiveView()
r.OrientationAxesVisibility
r.OrientationAxesInteractivity
r.OrientationAxesLabelColor
r.OrientationAxesOutlineColor
```
It woul...
Currently only the following can be controlled with pvpython for the orientation axes:
```
r=GetActiveView()
r.OrientationAxesVisibility
r.OrientationAxesInteractivity
r.OrientationAxesLabelColor
r.OrientationAxesOutlineColor
```
It would be nice if other properties, like position and size, can be directly set with pvpython. This was originally posted at [Discourse](https://discourse.paraview.org/t/orientation-axis-control/3027).
https://gitlab.kitware.com/paraview/paraview/-/issues/19492
Support versions in Python and XML plugins
2020-04-14T09:37:06-04:00
Ben Boeckel
Support versions in Python and XML plugins
Currently, there's no way to indicate what version a plugin is if it comes from a Python or XML file.
Discourse: https://discourse.paraview.org/t/indicate-version-for-vtkpythonalgorithmbase-based-python-plugin/3055
Cc: @utkarsh.ayachit
Currently, there's no way to indicate what version a plugin is if it comes from a Python or XML file.
Discourse: https://discourse.paraview.org/t/indicate-version-for-vtkpythonalgorithmbase-based-python-plugin/3055
Cc: @utkarsh.ayachit
https://gitlab.kitware.com/paraview/paraview/-/issues/19493
memory leak in catalyst live connection to Chombo simulation eventually cause...
2021-03-24T06:38:44-04:00
Aaron Knoll
memory leak in catalyst live connection to Chombo simulation eventually causes a crash
@demarle So we don't forget, from our SC19 experience. Not sure if this is specifically an issue with GR-CHOMBO data or more broadly in Catalyst. Thank you for the reminder!
@demarle So we don't forget, from our SC19 experience. Not sure if this is specifically an issue with GR-CHOMBO data or more broadly in Catalyst. Thank you for the reminder!
https://gitlab.kitware.com/paraview/paraview/-/issues/19496
Tiff texturing problem
2019-11-29T04:16:45-05:00
Cornelis Bockemühl
Tiff texturing problem
There is an issue with texturing with tiff files like the two in the attachment. My stripped down case (with ParaView 5.7 "release version") is:
- Generate a sphere with the Sphere source
- Apply Texture to plane
- Load one of the two t...
There is an issue with texturing with tiff files like the two in the attachment. My stripped down case (with ParaView 5.7 "release version") is:
- Generate a sphere with the Sphere source
- Apply Texture to plane
- Load one of the two texture files - one with tiff extension and one with only tif
1) TIFF: First of all, the load dialog offers the tiff file only, not the tif. But if you try to apply it, Paraview will simply crash.
2) TIF: If you force using the other file (that is not shown in the dialog because "tif" is not among the supported extensions), by manually typing that file name into the file dialog, the texture is applied just fine.
This is how things are going on a Windows 10 machine. The same thing on a machine with Ubuntu Linux 18.04 does not crash, but instead generates an error message: "Unknown texture file extension" - and of course no texture is shown.
Basically these are two issues:
- the crash
- the fact that only tiff can be loaded from the file dialog, but only tif files can be used as textures
[texture.zip](/uploads/eeb4631cc7cc8037d95813ef93520008/texture.zip)
https://gitlab.kitware.com/paraview/paraview/-/issues/19498
Segmentation Fault in Paraview 5.6.0 when opening a test home-grown HDF5 file
2019-12-02T05:12:42-05:00
de Vicente
Segmentation Fault in Paraview 5.6.0 when opening a test home-grown HDF5 file
When trying to open a simple HDF5 file (see attached file [orszag_0000.h5](/uploads/0b18111bb74342d8d8256701309e3391/orszag_0000.h5), Paraview 5.6.0 crashes with a segmentation fault, as can be seen below, complaining about the group /CO...
When trying to open a simple HDF5 file (see attached file [orszag_0000.h5](/uploads/0b18111bb74342d8d8256701309e3391/orszag_0000.h5), Paraview 5.6.0 crashes with a segmentation fault, as can be seen below, complaining about the group /CORE.
I understand that ParaView cannot read every possible HDF5 file, but it would be nice that instead of crashing it would give a warning that the file is unsupported.
```
angelv@comer ~]$ paraview
ERROR: In /home/angelv/spack/var/spack/stage/paraview-5.6.0-rj74wdfmntjmqhcxelyksaagllsxrw2z/spack-src/VTK/IO/VeraOut/vtkVeraOutReader.cxx, line 253
vtkVeraOutReader (0x557c148f2e10): Can't open group /CORE
[comer:4027338] *** Process received signal ***
[comer:4027338] Signal: Segmentation fault (11)
[comer:4027338] Signal code: Address not mapped (1)
[comer:4027338] Failing at address: (nil)
[comer:4027338] [ 0] /usr/lib/libpthread.so.0(+0x14930)[0x7f10212da930]
[comer:4027338] [ 1] /home/angelv/spack/opt/spack/linux-archrolling-x86_64/gcc-9.1.0/paraview-5.6.0-rj74wdfmntjmqhcxelyksaagllsxrw2z/lib/libvtkIOVeraOut-pv5.6.so.1(_ZN16vtkVeraOutReader9Internals8ReadCoreEv+0xa7)[0x7f101fd71787]
[comer:4027338] [ 2] /home/angelv/spack/opt/spack/linux-archrolling-x86_64/gcc-9.1.0/paraview-5.6.0-rj74wdfmntjmqhcxelyksaagllsxrw2z/lib/libvtkIOVeraOut-pv5.6.so.1(_ZN16vtkVeraOutReader9Internals12LoadMetaDataEv+0x36)[0x7f101fd72126]
[comer:4027338] [ 3] /home/angelv/spack/opt/spack/linux-archrolling-x86_64/gcc-9.1.0/paraview-5.6.0-rj74wdfmntjmqhcxelyksaagllsxrw2z/lib/libvtkIOVeraOut-pv5.6.so.1(_ZN16vtkVeraOutReader18RequestInformationEP14vtkInformationPP20vtkInformationVectorS3_+0x828)[0x7f101fd6b208]
[comer:4027338] [ 4] /home/angelv/spack/opt/spack/linux-archrolling-x86_64/gcc-9.1.0/paraview-5.6.0-rj74wdfmntjmqhcxely
```
https://gitlab.kitware.com/paraview/paraview/-/issues/19499
Add a way to save a multiblock into a parallel XML based format if all the le...
2020-05-04T22:44:22-04:00
Reynaldo Gil García
Add a way to save a multiblock into a parallel XML based format if all the leafs are of the same type.
It is currently not possible to create a distributed XML file (eg. pvtp) using Paraview without running ParaView in parallel and distributing the data.
It would be nice to be able to group dataset of the same type together using GroupDa...
It is currently not possible to create a distributed XML file (eg. pvtp) using Paraview without running ParaView in parallel and distributing the data.
It would be nice to be able to group dataset of the same type together using GroupDatasets filter then save the data as a pvt* (eg pvtp) file that will be produce a distributed dataset.
https://gitlab.kitware.com/paraview/paraview/-/issues/19502
ParaView is not handling FieldData at multiblock root level correctly
2023-10-06T09:00:32-04:00
Mathieu Westphal (Kitware)
ParaView is not handling FieldData at multiblock root level correctly
When manipulating a Multiblock dataset with a field data at root level, ParaView is showing strange behaviors.
Steps to reproduce :
* Open the provided file [rootTimeValue.tgz](/uploads/798098d695e6e43e71b2828dcbf1d06d/rootTimeValue.tg...
When manipulating a Multiblock dataset with a field data at root level, ParaView is showing strange behaviors.
Steps to reproduce :
* Open the provided file [rootTimeValue.tgz](/uploads/798098d695e6e43e71b2828dcbf1d06d/rootTimeValue.tgz) which contains a field data at root level, which should be supported.
* Show information tab, **vtkBlockColors appears as partial but it's not**
* Add a PassArrays filter, uncheck TimeValue, Apply
* **TimeValue is still present**
* Add a RandomAttributes filter, Apply
* Add a DescriptiveStatistics filter, Apply
* **Spreadsheet open on an empty FieldData, instead of opening RowData.**
@utkarsh.ayachit @berkgeveci @jpouderoux
https://gitlab.kitware.com/paraview/paraview/-/issues/19504
Non UTF8 char incorrectly writtten by the python state file saver
2022-09-24T00:10:40-04:00
Mathieu Westphal (Kitware)
Non UTF8 char incorrectly writtten by the python state file saver
When saving a python state containing file with non utf8 char in their name, the resulting state file contains garbage char.
Steps to reproduce :
* Open [tést.pvd](/uploads/7a8b116ba99ad61889689a8f2aa8324a/tést.pvd), Apply
* File -> Sa...
When saving a python state containing file with non utf8 char in their name, the resulting state file contains garbage char.
Steps to reproduce :
* Open [tést.pvd](/uploads/7a8b116ba99ad61889689a8f2aa8324a/tést.pvd), Apply
* File -> Save State -> python -> test.py
* ResetSession
* File -> Load State -> test.py
```
( 72.517s) [paraview ] vtkXMLReader.cxx:299 ERR| vtkPVDReader (0x7f353c019120): Error opening file /home/glow/tmp/tést.pvd
( 72.517s) [paraview ] vtkPVDReader.cxx:134 ERR| vtkPVDReader (0x7f353c019120): Could not read file information
( 72.518s) [paraview ] vtkExecutive.cxx:782 ERR| vtkPVCompositeDataPipeline (0x89d2850): Algorithm vtkPVDReader(0x7f353c019120) returned failure for request: vtkInformation (0x89d2660)
Debug: Off
Modified Time: 1866892
Reference Count: 1
Registered Events: (none)
Request: REQUEST_DATA_OBJECT
ALGORITHM_AFTER_FORWARD: 1
FORWARD_DIRECTION: 0
```
The resulting test.py contains :
```
tstpvd = PVDReader(FileName='/home/glow/tmp/tést.pvd')
```
Note : Strangely, the python trace behave correctly in this case.
https://gitlab.kitware.com/paraview/paraview/-/issues/19506
Moving views around in ParaView is not working perfectly
2021-03-24T06:37:45-04:00
Mathieu Westphal (Kitware)
Moving views around in ParaView is not working perfectly
Moving views around in ParaView sometimes does not work perfectly.
Steps to reproduce :
* Open ParaView
* Split the view in two
* Grab the frame of any view and drop it into the frame of any other view
* The views exchange places a...
Moving views around in ParaView sometimes does not work perfectly.
Steps to reproduce :
* Open ParaView
* Split the view in two
* Grab the frame of any view and drop it into the frame of any other view
* The views exchange places as expected
* Grab the frame of any view and drop it into the frame of any other view
* **nothing happens**
* Grab the frame of any view and drop it into the frame of any other view
* The views exchange places as expected
* ...
https://gitlab.kitware.com/paraview/paraview/-/issues/19509
3DGyphs Representation PipelineConnection mode UseGlyphTable property is conf...
2021-03-24T06:36:28-04:00
Mathieu Westphal (Kitware)
3DGyphs Representation PipelineConnection mode UseGlyphTable property is confusing and behave strangely
When using the 3DGlyphs representation with a PipelineConnection, the UseGlyphTable property appears.
The description of this property is unclear and does not seem to do anything.
The behavior is inconsistent anyway.
Steps to reproduce ...
When using the 3DGlyphs representation with a PipelineConnection, the UseGlyphTable property appears.
The description of this property is unclear and does not seem to do anything.
The behavior is inconsistent anyway.
Steps to reproduce :
* Open ParaView
* Open can.ex2, Apply
* MergeBlocks, Apply
* Connectivity, Apply
* Sphere, Apply
* Cone, Apply
* Cylinder, Apply
* GroupDataSets on Sphere and Cone, Apply
* Active connectivity filter
* 3DGlyph representation
* GlyphType : Pipeline Connection
* Rendering is not correctly updated at this point, see #19507
* Input : Cylinder, Rendering is updated
* **Check UseGlyphTable, nothing happens**
* Uncheck UseGlyphTable
* **Input : GroupDataSets, Rendering is updated, but strange looking.** see fig1
* **Check UseGlyphTable, nothing happens**, rendering is still strange
* Check UseCompositeGlyphTable, Rendering is updated correctly
* Uncheck UseGlyphTable, **rendering is updated and the used glyph is the first block of GroupDataSets, not coherent with the previous step**, see Fig2.
![fig1](/uploads/11a74eb216d2d7fb4b462cc2bf273e4b/fig1.png)*fig1*
![fig2](/uploads/67b36b474d372a1b35b9434b49f52da8/fig2.png)*fig2*
@ken-martin @michael.migliore @allisonvacanti
https://gitlab.kitware.com/paraview/paraview/-/issues/19510
vtkBlockColors is not available when representing a multiblock dataset in Poi...
2021-03-24T06:35:53-04:00
Mathieu Westphal (Kitware)
vtkBlockColors is not available when representing a multiblock dataset in PointGaussian Representation
vtkBlockColors is not available when representing a multiblock dataset in PointGaussian Representation
Steps to reproduce :
* Open can.ex2
* Color by vtkBlockColors
* Switch to PointGaussian
* vtkBlockColors appears with a ? and is...
vtkBlockColors is not available when representing a multiblock dataset in PointGaussian Representation
Steps to reproduce :
* Open can.ex2
* Color by vtkBlockColors
* Switch to PointGaussian
* vtkBlockColors appears with a ? and is not available anymore
https://gitlab.kitware.com/paraview/paraview/-/issues/19514
Support of Polyhedral Cells with XDMF 3
2020-11-19T09:52:00-05:00
Roman Fuchs
Support of Polyhedral Cells with XDMF 3
There are 3 readers for XDMF files in Paraview: XDMF Reader, Xdmf3ReaderS, Xdmf3ReaderT. But there is no information how they differ among each other.
I noticed that:
- XDMF Reader: cannot read polyhedral cells, can read grid-of-grids
...
There are 3 readers for XDMF files in Paraview: XDMF Reader, Xdmf3ReaderS, Xdmf3ReaderT. But there is no information how they differ among each other.
I noticed that:
- XDMF Reader: cannot read polyhedral cells, can read grid-of-grids
- Xdmf3ReaderS: can read polyhedral cells, cannot read grid-of-grids
Reading a polyhedral grid with Xdmf3ReaderS, I also noted that the time tag is not properly imported.
The XDMF website says that the tag "NumberOfElements" is only for clarity; however, Paraview needs that value at least for a few cell types (polyvertex, polyline, polygon) otherwise it crashes.
Clarify/extend your documentation and ensure that all data interfaces run as expected.
https://gitlab.kitware.com/paraview/paraview/-/issues/19515
Building with external VTK and external dependencies
2023-06-22T10:38:43-04:00
Alexander Neumann
Building with external VTK and external dependencies
So I know `USE_EXTERNAL_VTK` is not supported but I got ParaView to nearly build using vcpkg (https://github.com/microsoft/vcpkg/pull/9202) using all external dependencies. The only reason this is not yet successfull is due to some files...
So I know `USE_EXTERNAL_VTK` is not supported but I got ParaView to nearly build using vcpkg (https://github.com/microsoft/vcpkg/pull/9202) using all external dependencies. The only reason this is not yet successfull is due to some files not getting installed by VTK (i.e. internal VTK headers) and some setup problems using Qt5 which are solveable for me and has nothing to do with ParaView.
So I tried to force the build of ParaView fixing the above problems in the release build directory and got it to actually compile. The problem now is that ParaView crashes on startup. By debuging the problem it seems like ParaView wants to create an instance of `vtkPolyDataMapper` which it cannot find in the `NewInstanceFunctions` because it is simply missing.
So I investigated the problem a bit further and I assume it has something to do with the `vtkModuleWrapClientServer.cmake` and the `WrapClientServer.exe` which is generated and probably wraps the vtk/pv modules. So how could I achieve the correct Wrapping using an external VTK build?
So I would like to ask for some advice how I this could by solved and move on from here.
https://gitlab.kitware.com/paraview/paraview/-/issues/19516
paraview.simple ResetSession() is not actually reseting when used in pvpython
2021-03-24T06:45:37-04:00
Mathieu Westphal (Kitware)
paraview.simple ResetSession() is not actually reseting when used in pvpython
```
from paraview.simple import *
Render() -> expected error
s=Sphere()
r=Show()
v=Render()
ResetSession()
Render() -> **no error, view is still shown, unexpected**
Delete(v)
del(v) -> view disappear
Render() -> expected error
```
A wor...
```
from paraview.simple import *
Render() -> expected error
s=Sphere()
r=Show()
v=Render()
ResetSession()
Render() -> **no error, view is still shown, unexpected**
Delete(v)
del(v) -> view disappear
Render() -> expected error
```
A work around is to use
```
Disconnect()
Connect()
```
instead.
https://gitlab.kitware.com/paraview/paraview/-/issues/19518
Manage Links segfault when selecting patent then child element
2021-03-24T06:45:09-04:00
Charles Gueunet
Manage Links segfault when selecting patent then child element
# Bug:
Segfault using the `Manage Links` windows to link two properties.
# Step to reproduce:
* Create a cone
* Apply two clip on this cone (one inverted and one not for example)
* Tools > Manage Links ...
* Add ...
* Select Mode: Prope...
# Bug:
Segfault using the `Manage Links` windows to link two properties.
# Step to reproduce:
* Create a cone
* Apply two clip on this cone (one inverted and one not for example)
* Tools > Manage Links ...
* Add ...
* Select Mode: Property Link
* Select Source > Clip 1 (on le left column)
* Click on the ClipFunction
* Then, unfold the already clicked Clip1
* Select Plane
It seems the right pane does not crash with this manipulation though.
@jpouderoux
https://gitlab.kitware.com/paraview/paraview/-/issues/19519
python object path are incorrect in the python shell
2021-03-24T06:44:55-04:00
Mathieu Westphal (Kitware)
python object path are incorrect in the python shell
When using python, the python shell give module path for python object.
These module path are always incorrect.
```
w=Wavelet()
c=Clip()
Show()
Render()
t=c.ClipType
t
<paraview.servermanager.Plane object at 0x7f4f31803910>
paraview.ser...
When using python, the python shell give module path for python object.
These module path are always incorrect.
```
w=Wavelet()
c=Clip()
Show()
Render()
t=c.ClipType
t
<paraview.servermanager.Plane object at 0x7f4f31803910>
paraview.servermanager.Plane
Traceback (most recent call last):
File "<console>", line 1, in <module>
AttributeError: module 'paraview.servermanager' has no attribute 'Plane'
paraview.servermanager.implicit_functions.Plane
<class 'paraview.servermanager.Plane'>
```
This module path is really hard to guess and must be hardcoded somewhere.
https://gitlab.kitware.com/paraview/paraview/-/issues/19520
Automate setting renderView*.ViewSize
2019-12-10T14:28:55-05:00
W. Alan Scott
Automate setting renderView*.ViewSize
Please automate the setting of renderView*.ViewSize = [*,*]. Users are missing the comment at the top of the file. Please change it to something like:
```
if (ParaViewApp == 'true') // or (interactive == 'true') or somesuch
renderVi...
Please automate the setting of renderView*.ViewSize = [*,*]. Users are missing the comment at the top of the file. Please change it to something like:
```
if (ParaViewApp == 'true') // or (interactive == 'true') or somesuch
renderView*.ViewSize = [*,*]
```
https://gitlab.kitware.com/paraview/paraview/-/issues/19521
Filter: Limited number of output ports
2021-03-24T06:44:32-04:00
Mario Binkowski
Filter: Limited number of output ports
Following the documentation of [util.vtkAlgorithm](https://kitware.github.io/paraview-docs/latest/python/paraview.util.vtkAlgorithm.html) I want to write a filter which provides multiple output ports.
```
n = 10
from vtkmodules.util.vt...
Following the documentation of [util.vtkAlgorithm](https://kitware.github.io/paraview-docs/latest/python/paraview.util.vtkAlgorithm.html) I want to write a filter which provides multiple output ports.
```
n = 10
from vtkmodules.util.vtkAlgorithm import VTKPythonAlgorithmBase
from paraview.util.vtkAlgorithm import smproxy, smproperty, smdomain
@smproxy.filter(label='testMultipleOutput')
@smproperty.input(name='Input')
@smdomain.datatype(dataTypes=['vtkDataSet'], composite_data_supported=False)
class TestMultipleOutput(VTKPythonAlgorithmBase):
def __init__(self):
VTKPythonAlgorithmBase.__init__(self, nOutputPorts=n, outputType='vtkUnstructuredGrid')
def RequestData(self, request, inInfo, outInfo):
print(outInfo.GetNumberOfInformationObjects())
for i in range(n):
print(outInfo.GetInformationObject(i))
return 1
```
The above filter is doing fine. But if n is increased to some value > 10 the for loop crashes when i >= 10.
https://gitlab.kitware.com/paraview/paraview/-/issues/19522
Spreadsheet columns change when selection only is toggled
2021-03-24T06:44:18-04:00
Nicolas Vuaille
Spreadsheet columns change when selection only is toggled
When toggling the `Show only Selected Elements` button, the columns change.
Steps:
* Fast Uniform Grid, apply
* SpreadSheet View
* select at least one row
* click on `Show only Selected Elements`
`Structured Coordinates` column is ...
When toggling the `Show only Selected Elements` button, the columns change.
Steps:
* Fast Uniform Grid, apply
* SpreadSheet View
* select at least one row
* click on `Show only Selected Elements`
`Structured Coordinates` column is no more here, but we have `Points` and `Points Magnitude` ones, that were not present before.
tested on current master (e2464d04d5)
https://gitlab.kitware.com/paraview/paraview/-/issues/19523
Multi-Animation Support
2024-03-07T20:47:11-05:00
John Patchett
Multi-Animation Support
Multiple Animation tracks/Camera Tracks should be supported. Each "track" should be aware of the visibility of pipeline elements and the moving parts ie. Camera, contour/slice value, etc...
Multiple Animation tracks/Camera Tracks should be supported. Each "track" should be aware of the visibility of pipeline elements and the moving parts ie. Camera, contour/slice value, etc...
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19524
Animation of Contour Values Kills Contour Values
2022-11-30T20:25:09-05:00
John Patchett
Animation of Contour Values Kills Contour Values
If there are isosurface values chosen using the properties panel, if an animation is configured to iterate over contour values, all isosurface values chosen using the properties panel are stomped by the animation inspector.
![Animation_...
If there are isosurface values chosen using the properties panel, if an animation is configured to iterate over contour values, all isosurface values chosen using the properties panel are stomped by the animation inspector.
![Animation_Kills_Contour_Values_2](/uploads/7cfc3cb11b1b315b68b799f95e9809e8/Animation_Kills_Contour_Values_2.mov)
@demarle @ethan.stam
https://gitlab.kitware.com/paraview/paraview/-/issues/19527
PROTOBUF_USE_DLLS defined in static build due to CMake module
2020-02-12T12:54:29-05:00
Alexander Neumann
PROTOBUF_USE_DLLS defined in static build due to CMake module
So after successfully integrating paraview into vcpkg using the x64-windows triplet (#19515) and external VTK I want to fix the x64-windows-static triplet where everything is build statically.
The problem I encounter is that `PROTOBUF...
So after successfully integrating paraview into vcpkg using the x64-windows triplet (#19515) and external VTK I want to fix the x64-windows-static triplet where everything is build statically.
The problem I encounter is that `PROTOBUF_USE_DLLS` is defined. Since the only dependency on protobuf is ParaView itself I was wondering where this definition comes from.
After investigating I found out that CMakes FindProtobuf is doing it... so maybe protobuf should be switched over to CONFIG_MODE?
https://gitlab.kitware.com/paraview/paraview/-/issues/19528
Add Getting Started with Catalyst Guide
2024-03-07T20:47:12-05:00
Cory Quammen
cory.quammen@kitware.com
Add Getting Started with Catalyst Guide
Add a Getting Started with Catalyst Guide similar to the Getting Started with ParaView guide that already exists.
Add a Getting Started with Catalyst Guide similar to the Getting Started with ParaView guide that already exists.
5.13 (Summer 2024)
Cory Quammen
cory.quammen@kitware.com
Cory Quammen
cory.quammen@kitware.com
https://gitlab.kitware.com/paraview/paraview/-/issues/19529
Symbol Quad::Quad(void) already defined by VTK
2021-03-24T06:42:41-04:00
Alexander Neumann
Symbol Quad::Quad(void) already defined by VTK
all static build on windows with external VTK
vtkInfovisLayout-8.90d.lib(vtkIncrementalForceLayout.cxx.obj) : error LNK2005: "public: __cdecl Quad::Quad(void)" (??0Quad@@QEAA@XZ) already defined in lightweight_visit_vtk-pv5.7d.lib(vtkUn...
all static build on windows with external VTK
vtkInfovisLayout-8.90d.lib(vtkIncrementalForceLayout.cxx.obj) : error LNK2005: "public: __cdecl Quad::Quad(void)" (??0Quad@@QEAA@XZ) already defined in lightweight_visit_vtk-pv5.7d.lib(vtkUnstructuredGridFacelistFilter.C.obj)
Creating library lib\paraview.lib and object lib\paraview.exp
https://gitlab.kitware.com/paraview/paraview/-/issues/19533
Catalyst ordering of initialization code failure
2024-03-07T20:47:12-05:00
W. Alan Scott
Catalyst ordering of initialization code failure
We have a failure if Catalyst initializes itself in different orders. This should NOT matter. From Jeff:
We had, essentially,
```
processor = vtkCPProcessor::New();
pipeline = vtkCPPythonScriptPipeline::New();
processor->Initialize()...
We have a failure if Catalyst initializes itself in different orders. This should NOT matter. From Jeff:
We had, essentially,
```
processor = vtkCPProcessor::New();
pipeline = vtkCPPythonScriptPipeline::New();
processor->Initialize();
pipeline->Initialize(catalyst_python_file_name.c_str());
```
And I discovered it was dying after "vtkCPPythonScriptPipeline::New();"
I switched the order thus, to match the example:
```
processor = vtkCPProcessor::New();
processor->Initialize();
pipeline = vtkCPPythonScriptPipeline::New();
pipeline->Initialize(catalyst_python_file_name.c_str());
```
And it started working.
Since it worked in 5.6, I suspect some change in vtkCPProcessor and/or vtkCPPythonScriptPipeline which triggered an error in my ordering that did not previously exist, and none of the tests or examples used my (previously working) ordering.
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19544
Add a NamedValuesDomain
2021-03-24T06:41:09-04:00
Mathieu Westphal (Kitware)
Add a NamedValuesDomain
Add a NamedValuesDomain using an information_property to recover the names
See here for reference :
https://discourse.paraview.org/t/is-a-namedvaluesdomain-a-good-addition/1892
Add a NamedValuesDomain using an information_property to recover the names
See here for reference :
https://discourse.paraview.org/t/is-a-namedvaluesdomain-a-good-addition/1892
https://gitlab.kitware.com/paraview/paraview/-/issues/19545
cleanup: merge vtkCompleteArrays and vtkCleanArrays
2021-03-24T06:41:00-04:00
Utkarsh Ayachit
cleanup: merge vtkCompleteArrays and vtkCleanArrays
I suspect we can clean and combine `vtkCompleteArrays` and `vtkCleanArrays`. They seem to do similar things with a slight twist.
I suspect we can clean and combine `vtkCompleteArrays` and `vtkCleanArrays`. They seem to do similar things with a slight twist.
https://gitlab.kitware.com/paraview/paraview/-/issues/19546
cleanup: remove `vtkCompositeDataToUnstructuredGridFilter`
2021-03-24T06:40:52-04:00
Utkarsh Ayachit
cleanup: remove `vtkCompositeDataToUnstructuredGridFilter`
Remove `vtkCompositeDataToUnstructuredGridFilter`. It can be fully replaced by `vtkDataObjectTreeToPointSetFilter`.
Remove `vtkCompositeDataToUnstructuredGridFilter`. It can be fully replaced by `vtkDataObjectTreeToPointSetFilter`.
https://gitlab.kitware.com/paraview/paraview/-/issues/19549
update generic io library
2020-05-07T15:21:34-04:00
Utkarsh Ayachit
update generic io library
there's a new version available. need to confirm that the it still supports opening old versioned files.
cc: @berkgeveci
there's a new version available. need to confirm that the it still supports opening old versioned files.
cc: @berkgeveci
https://gitlab.kitware.com/paraview/paraview/-/issues/19550
add scale legend to the render view
2019-12-22T10:40:56-05:00
Utkarsh Ayachit
add scale legend to the render view
users want to see the a scale legend for the scale of the scene added to one of the corners in the 3D view
cc: @berkgeveci
users want to see the a scale legend for the scale of the scene added to one of the corners in the 3D view
cc: @berkgeveci
https://gitlab.kitware.com/paraview/paraview/-/issues/19552
Creating representation should not call show
2021-03-24T06:53:48-04:00
W. Alan Scott
Creating representation should not call show
When we are creating a representation, we should not Show(). This should be explicitly done in the trace, if desired. Here is a great discourse discussion, with detail from Cory, on what is going on. https://discourse.paraview.org/t/o...
When we are creating a representation, we should not Show(). This should be explicitly done in the trace, if desired. Here is a great discourse discussion, with detail from Cory, on what is going on. https://discourse.paraview.org/t/obtain-information-if-proxy-is-visible/2346/. Note that the user presented a fabulous use case, that I had to slightly change to get to work. Here is what I did:
* 5.7.0, Linux, builtin server.
* Edit user's run.py script as described in discourse. I have attached my edited script here (runAlan.py). Note you will have to edit this, and hard code where your state file is (from the Discourse discussion).
* View/ Python View/ Run Script.
Both Renderviews should not have anything showing, but Renderview 2 will have the fastUniformGrid showing. Further, the print in the run.py will show that we have visibility. This is a bug.
[runAlan.py](/uploads/913f38cbdc978b6e4daecd87cd031ca6/runAlan.py)
https://gitlab.kitware.com/paraview/paraview/-/issues/19553
LineChartView does not automatically update with LiveProgrammableSource
2022-12-02T08:20:42-05:00
Nicolas Vuaille
LineChartView does not automatically update with LiveProgrammableSource
Steps:
* LiveProgrammableSource
* output vtkTable
* Script
``` python
from numpy.random import randn
output.RowData.append(randn(10), 'noise')
```
* * Script (CheckNeedsUpdate)
``` python
self.SetNeedsUpdate(True)
```
* PlotDa...
Steps:
* LiveProgrammableSource
* output vtkTable
* Script
``` python
from numpy.random import randn
output.RowData.append(randn(10), 'noise')
```
* * Script (CheckNeedsUpdate)
``` python
self.SetNeedsUpdate(True)
```
* PlotData
LineChart View is updated only on mouse move hove
see : https://discourse.paraview.org/t/updating-filters-of-live-programmable-sources-on-new-data/2959
tested on master (745783903840e095e681cf8e61bf4519a439e2a4)
https://gitlab.kitware.com/paraview/paraview/-/issues/19557
Crash on .LAS file (ParaView 5.6.2 & 5.7 on MacOS)
2020-01-15T09:51:11-05:00
Alexey
Crash on .LAS file (ParaView 5.6.2 & 5.7 on MacOS)
I checked many LAS files and all of them crashed ParaView. One sample .LAS file attached:
[1050383876.las](/uploads/a5c0e4c45bb5d20e5bf7c0cb61a96c11/1050383876.las)
I checked many LAS files and all of them crashed ParaView. One sample .LAS file attached:
[1050383876.las](/uploads/a5c0e4c45bb5d20e5bf7c0cb61a96c11/1050383876.las)
https://gitlab.kitware.com/paraview/paraview/-/issues/19558
Crash on wrong reader (ParaView 5.6.2 & 5.7 on MacOS)
2020-05-06T10:26:07-04:00
Alexey
Crash on wrong reader (ParaView 5.6.2 & 5.7 on MacOS)
XYZ reader does not handle unhandled file gracefully and crashes :
[tesdata.xyz](/uploads/c6ed301dcd2c18e7c52b9eeb8926fa97/tesdata.xyz)
It would be great to provide some error message instead.
XYZ reader does not handle unhandled file gracefully and crashes :
[tesdata.xyz](/uploads/c6ed301dcd2c18e7c52b9eeb8926fa97/tesdata.xyz)
It would be great to provide some error message instead.
https://gitlab.kitware.com/paraview/paraview/-/issues/19559
Crash on SEG-Y files (ParaView 5.7 on MacOS)
2021-03-24T06:52:49-04:00
Alexey
Crash on SEG-Y files (ParaView 5.7 on MacOS)
This SEG-Y file works fine in ParaView 5.6.2 but not in ParaView 5.7. ParaView 5.6.2 requires ~2GB RAM to open the file but ParaView 5.7 crashes on ~50GB allocated memory (8GB RAM + swap).
The file is split to parts by command `split -b...
This SEG-Y file works fine in ParaView 5.6.2 but not in ParaView 5.7. ParaView 5.6.2 requires ~2GB RAM to open the file but ParaView 5.7 crashes on ~50GB allocated memory (8GB RAM + swap).
The file is split to parts by command `split -b 8000000 sample.sgy` due to upload size limit.
[xab](/uploads/c3fa630cc48784083bf365f73581c11d/xab)
[xaa](/uploads/96194defc763fadfcf29ca0d1c360b6f/xaa)
[xac](/uploads/dc17ac82a700c056f7eca6d46dc5911d/xac)
https://gitlab.kitware.com/paraview/paraview/-/issues/19570
Add Video Tutorial web page, linked in from help.
2024-03-07T20:47:12-05:00
W. Alan Scott
Add Video Tutorial web page, linked in from help.
* Please add a new web page on paraview.org/ Resources/ called Video Tutorials.
* On this page, please link youtube videos that are tutorials of ParaView.
* Please add https://www.youtube.com/watch?v=kGd-dGh9jnI as our first tutorial.
...
* Please add a new web page on paraview.org/ Resources/ called Video Tutorials.
* On this page, please link youtube videos that are tutorials of ParaView.
* Please add https://www.youtube.com/watch?v=kGd-dGh9jnI as our first tutorial.
* On the ParaView Help page, please combine "Professional Support" and "Professional Training" to be "Professional Support and Training"
* Add a line under "Sandia National Labs Tutorials" saying "Video Tutorials", linked to the web page created above.
* Please replace the top of the ParaView web page "WhatHaveYouRenderedInParaView", to something like "Submit your ParaView Training Videos". Lets see if we can get more interest!
@utkarsh.ayachit @cory.quammen @kmorel Thoughts?
5.13 (Summer 2024)
https://gitlab.kitware.com/paraview/paraview/-/issues/19571
If Python3 is turned off, ParaView fails on install
2020-12-10T21:51:32-05:00
W. Alan Scott
If Python3 is turned off, ParaView fails on install
If you turn Python3 off (no clue why yo), the Paraview make won't make pvbatch or pvpython. make install will try to install pvbatch and pvpython - and will then die. Here are the switches I (accidentally) flipped:
I did NOT have '...
If you turn Python3 off (no clue why yo), the Paraview make won't make pvbatch or pvpython. make install will try to install pvbatch and pvpython - and will then die. Here are the switches I (accidentally) flipped:
I did NOT have '-DENABLE_python3:BOOL=ON '
I DID have '-DUSE_SYSTEM_python3:BOOL=ON '
I DID have '-DENABLE_python:BOOL=ON '+\
and '-DENABLE_numpy:BOOL=ON '
W. Alan Scott
W. Alan Scott
https://gitlab.kitware.com/paraview/paraview/-/issues/19576
make default settings override more granular to make it feasible for users to...
2024-03-07T20:47:12-05:00
David E. DeMarle
make default settings override more granular to make it feasible for users to change defaults for individual properties
Users can already override and restore paraview's default settings with their own choices via the disk icon and cycle icons at the top of each section of the Properties tab. However these settings affect all currently non default values ...
Users can already override and restore paraview's default settings with their own choices via the disk icon and cycle icons at the top of each section of the Properties tab. However these settings affect all currently non default values changed on the page. That makes it harder than it should be to change individual settings in isolation.
We can improve the user experience with save and restore controls for every property. We have to be careful when we do so not to clutter the GUI.
See discussion on #19270 and !3822 regarding the camera projection mode.
5.13 (Summer 2024)
Aron Helser
Aron Helser
https://gitlab.kitware.com/paraview/paraview/-/issues/19590
simplify data-information to improve performance
2021-03-24T06:47:26-04:00
Utkarsh Ayachit
simplify data-information to improve performance
Data information is collected fairly frequently in ParaView. Recent analysis has shown that it can be quite time consuming to collect data information. This issue records the need to refactor the data information to improve performance. ...
Data information is collected fairly frequently in ParaView. Recent analysis has shown that it can be quite time consuming to collect data information. This issue records the need to refactor the data information to improve performance. A few ideas:
* split data info into essential and extra frills. avoid collecting extra info unless needed.
* don't collect composite meta-data -- split that into a separate information object; only collect combined information since that's what most of the GUI (and often users) need.
* collect data information in parallel in batch-symmetric and Catalyst modes. Not having distributed data information in those applications is a major inconvenience.
cc: @demarle @acbauer
https://gitlab.kitware.com/paraview/paraview/-/issues/19595
anaglyph stereo eye angle not dynamic, eye strain
2021-03-24T06:47:03-04:00
Andrew Annex
anaglyph stereo eye angle not dynamic, eye strain
Reposting an issue I reported at https://discourse.paraview.org/t/zoom-dependent-eye-angle-for-anaglyph-stereo/3236
In Paraview 5.7.0 on macOS it appears that when using anaglyph stereo, the camera does not dynamically update the eye an...
Reposting an issue I reported at https://discourse.paraview.org/t/zoom-dependent-eye-angle-for-anaglyph-stereo/3236
In Paraview 5.7.0 on macOS it appears that when using anaglyph stereo, the camera does not dynamically update the eye angle setting while zooming in/out of a scene. I am loading digital elevation models with overlain imagery so I tend to zoom into small areas of the scene. Currently without dynamic updates to the eye angle, as I zoom in I experience eye strain and pain as the effective separation of the left and right cameras increase and then the stereo effect fails past a certain zoom level. attached are screen shots that should illustrate this effectively, notice at the more zoomed in view the distance between the views of the left and right cameras effectively increase.
![Screen_Shot_2020-01-18_at_1.50.40_PM](/uploads/8d5407928c07c245f6d13868dc664e85/Screen_Shot_2020-01-18_at_1.50.40_PM.png)
![Screen_Shot_2020-01-18_at_1.50.58_PM](/uploads/25fdb75e7600037ab6e31d090458eee7/Screen_Shot_2020-01-18_at_1.50.58_PM.png)
https://gitlab.kitware.com/paraview/paraview/-/issues/19606
Poor Volume Rendering in VR mode ...
2021-03-24T06:46:44-04:00
Niklas Roeber
Poor Volume Rendering in VR mode ...
The volume rendering of VTK image data is of pretty poor quality if visualization is streamed to VR googles. It appears that sometimes too few slices are used for volume rendering, or that no interpolation takes place between the slabs. ...
The volume rendering of VTK image data is of pretty poor quality if visualization is streamed to VR googles. It appears that sometimes too few slices are used for volume rendering, or that no interpolation takes place between the slabs. This only happens for the GPU based volume rendering, but also not for all volumes. The standard engine data set has this poor rendering, the ParaView internal Wavelet does not. Yet the Wavelet exhibits another issue. If you volume render the Wavelet in VR, please interact with it using "Pick" and move it around and watch from different sides. The volume texture appears to be skewed at places, showing clearly a wrong scaling. If required, I can try a screen capture.
Thanks and Cheers,
Niklas
@ken-martin @michael.migliore
https://gitlab.kitware.com/paraview/paraview/-/issues/19607
Only last XDMF FiniteElementFunction attribute is visible in ParaView
2020-05-07T12:31:49-04:00
Darren Parfitt
Only last XDMF FiniteElementFunction attribute is visible in ParaView
When I have two XDMF attributes using the ItemType="FiniteElementFunction" inside a single grid object ParaView only lists the last one in its available variables to plot. I have uploaded the example file I used. This may be an issue wi...
When I have two XDMF attributes using the ItemType="FiniteElementFunction" inside a single grid object ParaView only lists the last one in its available variables to plot. I have uploaded the example file I used. This may be an issue with ParaView or lower down with the actual VTK/XDMF3Reader.
The example include two cell centred results to show these work. I would have included nodal as well but if you do that you hit another bug (discussed [here](https://discourse.paraview.org/t/xdmf3-reader-finiteelementfunction/3079) ). This may be related because in my case it complains about the nodal result having only 6 values (which it should) when PV is expecting 8 (which is actually the size of the FiniteElement attributes)
I have tried PV 5.5.2, 5.6.1 and 5.7 and they all show the same issue.
[test_finiteelement.xmf](/uploads/654153ebd082900fe29ce82e1f846240/test_finiteelement.xmf)
https://gitlab.kitware.com/paraview/paraview/-/issues/19608
XDMF3 RectilinearGrid with HyperSlabbed coordinate arrays are not working.
2020-07-24T10:48:13-04:00
David E. DeMarle
XDMF3 RectilinearGrid with HyperSlabbed coordinate arrays are not working.
Files like this:
<Grid GridType="Uniform" Name="0">
<Topology Type="2DRectMesh" NumberOfElements="5 5"/>
<Geometry Type="VXVY">
<DataItem ItemType="HyperSlab" Dimensions="5" Type="HyperSlab" ><DataItem Dimensio...
Files like this:
<Grid GridType="Uniform" Name="0">
<Topology Type="2DRectMesh" NumberOfElements="5 5"/>
<Geometry Type="VXVY">
<DataItem ItemType="HyperSlab" Dimensions="5" Type="HyperSlab" ><DataItem Dimensions="3 2" NumberType="Int" Format="XML">0 0 1 1 1 5</DataItem>
<DataItem Format="HDF" Dimensions="4 5" Name="x" NumberType="Float" Precision="8">
test.hdf5:/Locations/x</DataItem>
</DataItem>
<DataItem ItemType="HyperSlab" Dimensions="5" Type="HyperSlab" ><DataItem Dimensions="3 2" NumberType="Int" Format="XML">0 0 1 1 1 5</DataItem>
<DataItem Format="HDF" Dimensions="4 5" Name="y" NumberType="Float" Precision="8">
test.hdf5:/Locations/y</DataItem>
</DataItem>
</Geometry>
</Grid>
Are were found to work in Xdmf2, but not in Xdmf3 such that the X and Y coordinate arrays are nonsensical when read. It should work in either generation of the reader.
Cory Quammen
cory.quammen@kitware.com
Cory Quammen
cory.quammen@kitware.com