VTK issueshttps://gitlab.kitware.com/vtk/vtk/-/issues2023-09-13T15:52:22-04:00https://gitlab.kitware.com/vtk/vtk/-/issues/19077Feature edges is wrong for all nonlinear cells.2023-09-13T15:52:22-04:00Florian MaurinFeature edges is wrong for all nonlinear cells.Go to the unstructured grid source.
Select for instance a single quadratic hexagon (1x1x1 grid)
Switch the representation to feature edge.
We are supposed to get a single cube ( like it is the case for all linear cells). We have instead...Go to the unstructured grid source.
Select for instance a single quadratic hexagon (1x1x1 grid)
Switch the representation to feature edge.
We are supposed to get a single cube ( like it is the case for all linear cells). We have instead the following ( note that things get even funnier when you play with the subdivision level).
![image](/uploads/cac35084777c517d3716c9146c1778e6/image.png)
In the source, replace the quadratic hexagon by other nonlinear cells, the same kind of errors will be produced.
I don't know if this is the same problem, but there was already something already reported for the feature edges:
https://gitlab.kitware.com/paraview/paraview/-/issues/21614https://gitlab.kitware.com/vtk/vtk/-/issues/19076increase xdmf2 MAX_COLLECTABLE_NUMBER_OF_GRIDS2023-09-15T09:41:46-04:00Mike Treeincrease xdmf2 MAX_COLLECTABLE_NUMBER_OF_GRIDSWe have a use-case wherein we load in more than 1000 grids in an xdmf2 file into ParaView and find that the parallelization of the data processing along the pipeline is then ruined. Increasing the MAX_COLLECTABLE_NUMBER_OF_GRIDS solves t...We have a use-case wherein we load in more than 1000 grids in an xdmf2 file into ParaView and find that the parallelization of the data processing along the pipeline is then ruined. Increasing the MAX_COLLECTABLE_NUMBER_OF_GRIDS solves the issuehttps://gitlab.kitware.com/vtk/vtk/-/issues/19068Move CommonImplicitArrays in CommonCore multiply defined symbols found2024-01-18T16:27:32-05:00Andrew MacleanMove CommonImplicitArrays in CommonCore multiply defined symbols foundThis MR !10461 is Ok however I am seeing this error in relation to !10253:
Here is the wrror in Windows VS Studio: [MultipleSymbols.txt](/uploads/4b881f8ae4660b7098439f7db9dbe904/MultipleSymbols.txt)This MR !10461 is Ok however I am seeing this error in relation to !10253:
Here is the wrror in Windows VS Studio: [MultipleSymbols.txt](/uploads/4b881f8ae4660b7098439f7db9dbe904/MultipleSymbols.txt)https://gitlab.kitware.com/vtk/vtk/-/issues/19066VTK 9.3.0rc1 build failure on Fedora Rawhide i6862023-09-04T04:35:42-04:00Orion PoplawskiVTK 9.3.0rc1 build failure on Fedora Rawhide i686This only occurs on the i686 32-bit build:
```
[ 64%] Building CXX object Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build/Filters/DSP/Testing/Cxx && /usr/b...This only occurs on the i686 32-bit build:
```
[ 64%] Building CXX object Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build/Filters/DSP/Testing/Cxx && /usr/bin/g++ -Dkiss_fft_scalar=double -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Filters/DSP -I/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/DataModel -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/DataModel -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Math -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Math -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/kissfft/vtkkissfft -I/builddir/build/BUILD/VTK-9.3.0.rc1/ThirdParty/kissfft/vtkkissfft -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Transforms -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Transforms -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/ExecutionModel -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/ExecutionModel -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Filters/General -I/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/General -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Misc -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Misc -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Filters/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Testing/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/Testing/Core -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/Utilities/KWIML -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/Utilities/KWIML -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/Utilities/KWSys -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/Utilities/KWSys -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/kissfft -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/ThirdParty/kissfft -O2 -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wno-complain-wrong-lang -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m32 -march=i686 -mtune=generic -msse2 -mfpmath=sse -mstackrealign -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -D_UNICODE -DHAVE_UINTPTR_T -g -std=c++11 -fPIE -fvisibility=hidden -fvisibility-inlines-hidden -MD -MT Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o -MF CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o.d -o CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o -c /builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx: In function 'int TestBandFiltering(int, char**)':
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:101:27: error: no matching function for call to 'CheckArray(vtkDataArray*&, const std::array<double, 18>&)'
101 | exitCode += ::CheckArray(arr, EXPECTED_VALUE1);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: candidate: 'template<long unsigned int N> int {anonymous}::CheckArray(vtkDataArray*, std::array<double, N>)'
31 | int CheckArray(vtkDataArray* array, std::array<double, N> expected)
| ^~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: template argument deduction/substitution failed:
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:101:27: note: mismatched types 'long unsigned int' and 'unsigned int'
101 | exitCode += ::CheckArray(arr, EXPECTED_VALUE1);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:117:27: error: no matching function for call to 'CheckArray(vtkDataArray*&, const std::array<double, 18>&)'
117 | exitCode += ::CheckArray(arr, EXPECTED_VALUE1);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: candidate: 'template<long unsigned int N> int {anonymous}::CheckArray(vtkDataArray*, std::array<double, N>)'
31 | int CheckArray(vtkDataArray* array, std::array<double, N> expected)
| ^~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: template argument deduction/substitution failed:
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:117:27: note: mismatched types 'long unsigned int' and 'unsigned int'
117 | exitCode += ::CheckArray(arr, EXPECTED_VALUE1);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:124:27: error: no matching function for call to 'CheckArray(vtkDataArray*&, const std::array<double, 54>&)'
124 | exitCode += ::CheckArray(arr, EXPECTED_VALUE2);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: candidate: 'template<long unsigned int N> int {anonymous}::CheckArray(vtkDataArray*, std::array<double, N>)'
31 | int CheckArray(vtkDataArray* array, std::array<double, N> expected)
| ^~~~~~~~~~
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:31:5: note: template argument deduction/substitution failed:
/builddir/build/BUILD/VTK-9.3.0.rc1/Filters/DSP/Testing/Cxx/TestBandFiltering.cxx:124:27: note: mismatched types 'long unsigned int' and 'unsigned int'
124 | exitCode += ::CheckArray(arr, EXPECTED_VALUE2);
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
gmake[2]: *** [Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/build.make:93: Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/TestBandFiltering.cxx.o] Error 1
gmake[2]: Leaving directory '/builddir/build/BUILD/VTK-9.3.0.rc1/build'
gmake[1]: *** [CMakeFiles/Makefile2:28587: Filters/DSP/Testing/Cxx/CMakeFiles/vtkFiltersDSPCxxTests.dir/all] Error 2
```
I don't know if VTK is still supporting 32-bit builds or not.https://gitlab.kitware.com/vtk/vtk/-/issues/19065vtkSwtComponent.java:68 uses unreferenced symbol from parent class2023-08-31T05:35:42-04:00JensgwvtkSwtComponent.java:68 uses unreferenced symbol from parent classvtkSwtComponent uses the renderWindow from its parent class vtkAbstractComponent. However, the called method in line 68 renderWindow.SetFrameBlitModeToBlitToCurrent() cannot be resolved. Reason being the renderWindow in vtkAbstractCompon...vtkSwtComponent uses the renderWindow from its parent class vtkAbstractComponent. However, the called method in line 68 renderWindow.SetFrameBlitModeToBlitToCurrent() cannot be resolved. Reason being the renderWindow in vtkAbstractComponent is initialized as vtkRenderWindow. However, only the subclass of vtkRenderWindow vtkOpenGLRenderWindow has this method. So either the abstraction to vtkRenderWindow in vtkAbstractComponent or the use of renderWindow.SetFrameBlitModeToBlitToCurrent() in vtkSwtComponent is wrong. Please fix this.https://gitlab.kitware.com/vtk/vtk/-/issues/19063Attempting to use an MPI routine before initializing MPICH2023-08-30T07:36:10-04:00Thierry ThomasAttempting to use an MPI routine before initializing MPICHThe message in the title is displayed when launching ParaView built with MPICH.
The problem has been reproduced with a smaller program: TaskParallelism, in the examples.
This error arise when the program is linked with libvtkRemotingVi...The message in the title is displayed when launching ParaView built with MPICH.
The problem has been reproduced with a smaller program: TaskParallelism, in the examples.
This error arise when the program is linked with libvtkRemotingViewsPython-pv5.11.so, but there is no error without this library.
Note: I know that this code belongs to ParaView and not VTK, and I firstly reported this issue at https://gitlab.kitware.com/paraview/paraview/-/issues/22256 (more details are available there), but since this code comes from VTK, I have been asked to report it here.https://gitlab.kitware.com/vtk/vtk/-/issues/19062VTK Python Examples Segfault on Exit (Ubuntu WSL)2023-08-30T10:14:30-04:00Chris MortonVTK Python Examples Segfault on Exit (Ubuntu WSL)Windows 11 WSL2 Ubuntu 20.04 Python 3.11.4 VTK 9.2.6
(I also see a similar or same segmentation fault with 9.3.0.rc1)
I am building locally from source. However, installing vtk from pypi also gives this segmentation fault.
This is lik...Windows 11 WSL2 Ubuntu 20.04 Python 3.11.4 VTK 9.2.6
(I also see a similar or same segmentation fault with 9.3.0.rc1)
I am building locally from source. However, installing vtk from pypi also gives this segmentation fault.
This is likely a WSL issue but I am submitting here in case it is of any help to the development team. I understand that you may not yet support WSL too. Feel free to close immediately if you feel this is not actionable on your end.
I get a segmentation fault when running python examples (such as the boolean example): the dialog opens and correctly shows the rendered object. I can interact with object in the usual manner. When closing the dialog using the x in the top right corner I get the segmentation fault after the dialog disappears.
The crash does not occur when running the example in gdb, the process exits normally, suggesting a multi-thread timing issue. When I dump a core file I get the following information when loaded into gdb:
```plaintext
Reading symbols from python3...
```
```plaintext
warning: core file may not match specified executable file.
```
```plaintext
[New LWP 11528]
[New LWP 11530]
[New LWP 11531]
[New LWP 11532]
[New LWP 11529]
[New LWP 11525]
```
```plaintext
[Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
```
```plaintext
Core was generated by `python3 boolean.py'.
```
```plaintext
Program terminated with signal SIGSEGV, Segmentation fault. #0 0x00007f121e47c1ca in ?? () from /usr/lib/wsl/drivers/nvdm.inf_amd64_47b4de8ce9fa1909/libnvwgf2umx.so
```
```plaintext
[Current thread is 1 (Thread 0x7f121b9dd700 (LWP 11528))]
```https://gitlab.kitware.com/vtk/vtk/-/issues/19060build failure with emscripten 3.1.452023-10-19T17:54:30-04:00Raymond Chenbuild failure with emscripten 3.1.45I am tring to use webgpu, but can not build .
### commit `1221eaeeb7`
### the build error info
```
vtk/Rendering/WebGPU/vtkWebGPUInternalsShaderModule.cxx:11:12: error: no member named 'source' in 'wgpu::ShaderModuleWGSLDescriptor'
...I am tring to use webgpu, but can not build .
### commit `1221eaeeb7`
### the build error info
```
vtk/Rendering/WebGPU/vtkWebGPUInternalsShaderModule.cxx:11:12: error: no member named 'source' in 'wgpu::ShaderModuleWGSLDescriptor'
11 | wgslDesc.source = source.c_str();
| ~~~~~~~~ ^
1 error generated.
```
### in `webgpu_cpp.h`
![image](/uploads/dd48a1c05beb00cf07a6f3a679effa1c/image.png)
![image](/uploads/7b1d76040bddaefa106cbef2500a73b8/image.png)
### emcc
```
emcc -v
emcc (Emscripten gcc/clang-like replacement + linker emulating GNU ld) 3.1.45 (ef3e4e3b044de98e1811546e0bc605c65d3412f4)
clang version 18.0.0 (https://github.com/llvm/llvm-project d1e685df45dc5944b43d2547d0138cd4a3ee4efe)
Target: wasm32-unknown-emscripten
Thread model: posix
```Jaswant Panchumarti (Kitware)Jaswant Panchumarti (Kitware)https://gitlab.kitware.com/vtk/vtk/-/issues/19059vtk 9.3.0.rc1 java build failure on Fedora Rawhide2023-09-01T20:04:31-04:00Orion Poplawskivtk 9.3.0.rc1 java build failure on Fedora RawhideI'm getting this trying to update the Fedora package:
```
[ 28%] Building Java object Wrapping/Java/CMakeFiles/vtkjava.dir/vtk/sample/rendering/annotation/LabeledCubeAxesActor.class
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/J...I'm getting this trying to update the Fedora package:
```
[ 28%] Building Java object Wrapping/Java/CMakeFiles/vtkjava.dir/vtk/sample/rendering/annotation/LabeledCubeAxesActor.class
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/Java && /usr/bin/javac -classpath /builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/Java -source 1.7 -target 1.7 /builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/Java/vtk/sample/rendering/annotation/LabeledCubeAxesActor.java -d CMakeFiles/vtkjava.dir
warning: [options] bootstrap class path not set in conjunction with -source 7
warning: [options] source value 7 is obsolete and will be removed in a future release
warning: [options] target value 7 is obsolete and will be removed in a future release
warning: [options] To suppress warnings about obsolete options, use -Xlint:-options.
/builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/Java/vtk/sample/rendering/annotation/LabeledCubeAxesActor.java:69: error: incompatible types: int cannot be converted to boolean
cubeAxesActor.SetXAxisMinorTickVisibility(0);
^
/builddir/build/BUILD/VTK-9.3.0.rc1/build/Wrapping/Java/vtk/sample/rendering/annotation/LabeledCubeAxesActor.java:72: error: incompatible types: int cannot be converted to boolean
cubeAxesActor.SetYAxisMinorTickVisibility(0);
^
Note: Some messages have been simplified; recompile with -Xdiags:verbose to get full output
2 errors
4 warnings
```https://gitlab.kitware.com/vtk/vtk/-/issues/19056render width EdgeVisiblilityOn in WASM with shaderProgram error2023-09-05T09:49:35-04:00Raymond Chenrender width EdgeVisiblilityOn in WASM with shaderProgram errorthe error info as follows
![image](/uploads/509b883f468e6f4f9123c635fe8faba0/image.png)
![image](/uploads/8f079c909d576fcfedd03edcb62f42e5/image.png)
![image](/uploads/825552f49b2796d9b2d852cfd94308ea/image.png)
the commit is `94b78e...the error info as follows
![image](/uploads/509b883f468e6f4f9123c635fe8faba0/image.png)
![image](/uploads/8f079c909d576fcfedd03edcb62f42e5/image.png)
![image](/uploads/825552f49b2796d9b2d852cfd94308ea/image.png)
the commit is `94b78e3034`https://gitlab.kitware.com/vtk/vtk/-/issues/19055vtkDecimatePolylineFilter incorrectly copying cell data2023-08-23T09:43:14-04:00Matthew Heinsen EganvtkDecimatePolylineFilter incorrectly copying cell datavtkDecimatePolylineFilter seems to be incorrectly copying cell data by using the current polyline's first vertex index rather than the current polyline's source cell id. This appears to have been accidentally introduced by c66b00ab34dad3...vtkDecimatePolylineFilter seems to be incorrectly copying cell data by using the current polyline's first vertex index rather than the current polyline's source cell id. This appears to have been accidentally introduced by c66b00ab34dad326d5dbdd7a26639b15a75a3c47. Fixed by the small change in !10430
[testdecimateline.cpp](/uploads/ebcb6f2433b05ec28286db93943f3a23/testdecimateline.cpp)https://gitlab.kitware.com/vtk/vtk/-/issues/19053Problem using external fast_float library2023-08-23T16:11:11-04:00Orion PoplawskiProblem using external fast_float libraryI'm testing building the Fedora vtk package with 9.3.0.rc1 but ending up with the following error:
```
[ 6%] Building CXX object Common/Core/CMakeFiles/CommonCore.dir/vtkValueFromString.cxx.o
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build...I'm testing building the Fedora vtk package with 9.3.0.rc1 but ending up with the following error:
```
[ 6%] Building CXX object Common/Core/CMakeFiles/CommonCore.dir/vtkValueFromString.cxx.o
cd /builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Core && /usr/lib64/ccache/g++ -DCommonCore_EXPORTS -DVTK_MODULE_ENABLE_VTK_loguru=1 -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/Common/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Core -I/builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/loguru/vtkloguru -I/builddir/build/BUILD/VTK-9.3.0.rc1/ThirdParty/loguru/vtkloguru -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/Utilities/KWIML -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/Utilities/KWIML -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/Utilities/KWSys -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/Utilities/KWSys -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/utf8 -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/ThirdParty/utf8 -isystem /usr/include/utf8cpp -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/loguru -isystem /builddir/build/BUILD/VTK-9.3.0.rc1/ThirdParty/loguru -O2 -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wno-complain-wrong-lang -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -D_UNICODE -DHAVE_UINTPTR_T -g -std=c++11 -fPIC -fvisibility=hidden -fvisibility-inlines-hidden -MD -MT Common/Core/CMakeFiles/CommonCore.dir/vtkValueFromString.cxx.o -MF CMakeFiles/CommonCore.dir/vtkValueFromString.cxx.o.d -o CMakeFiles/CommonCore.dir/vtkValueFromString.cxx.o -c /builddir/build/BUILD/VTK-9.3.0.rc1/Common/Core/vtkValueFromString.cxx
gmake[2]: Leaving directory '/builddir/build/BUILD/VTK-9.3.0.rc1/build'
/builddir/build/BUILD/VTK-9.3.0.rc1/Common/Core/vtkValueFromString.cxx:11:10: fatal error: vtkfast_float.h: No such file or directory
11 | #include <vtkfast_float.h>
| ^~~~~~~~~~~~~~~~~
```
This file gets created in `/builddir/build/BUILD/VTK-9.3.0.rc1/build/ThirdParty/fast_float/vtkfast_float.h`, but no include path seems to be specified for that directory. We are attempting to use the system version of fast_float.https://gitlab.kitware.com/vtk/vtk/-/issues/19052vtkWorldPointPicker will stuck when the pick function get -1 in release mode2023-08-18T06:13:27-04:00ty tyvtkWorldPointPicker will stuck when the pick function get -1 in release modeHello,
I'm using a **vtkWorldPicker** to get world coordinates, and if I pass **\-1** to the **Pick** function, the software is unresponsive, but doesn't crash! this happens in **release** mode, in debug mode, it works. The problematic ...Hello,
I'm using a **vtkWorldPicker** to get world coordinates, and if I pass **\-1** to the **Pick** function, the software is unresponsive, but doesn't crash! this happens in **release** mode, in debug mode, it works. The problematic code looks like this:
```plaintext
vtkNew<vtkWorldPointPicker> picker;
picker->Pick(-1,1,0,renderer)
```https://gitlab.kitware.com/vtk/vtk/-/issues/19048vtkBlockSortHelper needs more documentation2023-11-06T21:59:47-05:00Ken MartinvtkBlockSortHelper needs more documentationvtkBlockSortHelper could use more documentation. Here is some text etc that may help.
vtkBlockSortHelper is a helper class designed to sort blocks into back to front ordering as is typically required for correct translucent object compo...vtkBlockSortHelper could use more documentation. Here is some text etc that may help.
vtkBlockSortHelper is a helper class designed to sort blocks into back to front ordering as is typically required for correct translucent object compositing. This can be a tricky problem which in the general case may not have a solution without changing the underlying geometry (such as with three translucent triangles that each overlap one other triangle and are overlapped by one other). This class will not modify the geometry but tries to produce an ordering that will work. One approach would be to use the distance from camera to block center but this can produce incorrect orderings. Consider the example below
![IMG_3938.jpg](/uploads/af1b72611e0f4cc75439ef11b9ddd801/IMG_3938.jpg)
assume the blocks A and C are the same in both cases and only block B is changing. The correct rendering order between A and C can only be determined with knowledge of block B's geometry. Using centroids (or closest point to camera) would not work. This class is designed to handle the case above, but does not handle the case of fully disconnected blocks.
The basic algorithm is as follows
* put all blocks into the working vector
* loop over the working vector -\> it
* loop over the working vector -\> it2
* compare it and it2 where the results can be -1 first closer, +1 second closer, 0 unknown order
* if no value of it2 is in front of it then we know it is the front-most of what is remaining in the working vector, so add it to the results vector and remove it from the working vector and restart, otherwise move to the next it
* reverse copy the result vector to get back to front ordering and return
The comparison operator is critical to this working correctly and works as follows
* compute the projection of block A onto block B
* compute the projection of block B onto block A
* determine the dimensionality of the projection
* 3 dimensions overlap = volume
* 2 dimensions overlap = plane
* 1 dimension overlaps = line
* 0 dimensions overlap = no overlap
* for 3 dimension overlap take the two largest dimensions and use them to collapse down to a 2d overlap
* for 1 or 0 dimensions of overlap that means the two blocks cannot share a surface so return 0 = unknown ordering
* compute the distance between the two projections, if they are not adjacent then return 0 = unknown ordering
* at this point we have a shared plane where the two blocks touch, we compare this plane to the camera model to determine which side of the plane is in front, and return -1 or +1 based on which block is on the front side of the plane
So this algorithm works only when all the blocks are connected at least transitively to each other by some amount of face connectivity. The drawing up above also shows why the comparison operator cannot determine a correct ordering of A and C without knowledge of B (or in more complex cases many intermediate blocks).
There are other ways to write this algorithm (such as providing the working set to the comparison operator) which would yield the same results.
It is also possible to improve this helper to handle disconnected blocks to varying degrees. Such as
* examine the blocks and break them into a vector of vectors of connected blocks (each vector only contains face connected blocks)
* Sort each vector of connected blocks as is currently done
* compute the overall bounding box of each vector
* sort the vectors based on their bounding boxes using the same algorithm but without the requirement that the blocks must be touching. e.g. you could just define the plane as the mid-point on the line connecting the centers of the two blocks that is perpendicular to the line.
There are other completely different approaches as well. You could convert all the blocks into 2d hexagons with z values corresponding to camera depth and then do some sort of ordering from that, etc.Jaswant Panchumarti (Kitware)Jaswant Panchumarti (Kitware)https://gitlab.kitware.com/vtk/vtk/-/issues/19045VTK rendering window inside a Qt application freezes the Operating System up ...2023-08-17T03:50:31-04:00Sayan BhattacharjeeVTK rendering window inside a Qt application freezes the Operating System up when resized opaquely/continously.Hello everyone,
In VTK 9.2.6 debug build, with Qt 5.12.8, when I try to resize the VTK window using the vertical Qt QSplitter widget _"opaquely"_, I have observed that the VTK window freezes up, and the Operating System becomes unrespon...Hello everyone,
In VTK 9.2.6 debug build, with Qt 5.12.8, when I try to resize the VTK window using the vertical Qt QSplitter widget _"opaquely"_, I have observed that the VTK window freezes up, and the Operating System becomes unresponsive, and has to be hard reset (restarted after shutting off the power to the computer).
By _"opaquely"_, I mean that it's the `opaqueResize` feature of the QSplitter widget, and as the QSplitter is moved, the VTK window is resized continuously, and due to this continuous resizing, something goes wrong, and the VTK window freezes up, as described above.
Qt Documentation on `opaqueResize`: https://doc.qt.io/qt-5/qsplitter.html#opaqueResize-prop
The VTK window being used here is a `QVTKOpenGLNativeWidget` that has been promoted to be used in Qt.
The error is reduced and doesn't happen as often when we disable opaque resizing for the QSplitter, but still I have seen it happen once in a while, but the frequency is somewhat acceptable for now, though it would be good if it was fixed completely. The image below shows how it looks when opaque resizing is disabled. The resizing only happens when we let go off the mouse.
Since the VTK window no longer being continuously resized, the code is more stable, so by deduction, we can say that the continuous resizing is the cause of the bug.
I have a Proof of Concept (POC) code, but the bug is self descriptive and would probably not require a POC. But if you require a POC, I would be happy to submit it.
PS: Interestingly, I don't have a proof, but I was able to switch to a TTY terminal when a freeze up happened, and saw an error message that said that "The system was Out of Memory (OOM)." Meaning that something happens which probably causes VTK to use up a lot of memory. I don't have a proof that that's what happens, but maybe the continuous resizing causes too many buffers to be allocated or something internally.
```
System : Linux Ubuntu 20.04.01
VTK : 9.2.6 built from source in debug mode
Compiled with : g++ (Ubuntu 9.3.0-17ubuntu1~20.04) 9.3.0
```
![qtvtk](/uploads/d1af5523aa27e1f4106c486977cd6ba1/qtvtk.png)https://gitlab.kitware.com/vtk/vtk/-/issues/19039OSPRay 2.12: nothing is rendered2024-03-13T18:55:40-04:00SkydeOSPRay 2.12: nothing is renderedI have tested the latest version of OSPRay with pathtracer (with the master of VTK) and no dataset will be render. To reproduce that I have attached an example :
[vtk-ospray.zip](/uploads/28e9d64919e2eb970a6e0ce365926103/vtk-ospray.zip...I have tested the latest version of OSPRay with pathtracer (with the master of VTK) and no dataset will be render. To reproduce that I have attached an example :
[vtk-ospray.zip](/uploads/28e9d64919e2eb970a6e0ce365926103/vtk-ospray.zip).
Here is what I have with this code with different version of OSPray:
- With OSPRay 2.7.1:
![image](/uploads/517a8fd01f51e3edc3aa283e8518a333/image.png)
- With OSPRay 2.12.0:
![image](/uploads/3e8ee96422de67fd3bfa9db3b6d8f7a1/image.png)
Configs:
- **windows 11**
- vtk **9.2.20230730**https://gitlab.kitware.com/vtk/vtk/-/issues/19038VTK Nightly Documentation - Missing Online Examples2023-08-09T11:36:52-04:00Andrew MacleanVTK Nightly Documentation - Missing Online ExamplesFor example: In https://vtk.org/doc/nightly/html/classvtkFlyingEdges3D.html the section **Online Examples:** is now blank. This section was working about a week ago.
The file used to generate the online examples is here: [vtk_vtk-exampl...For example: In https://vtk.org/doc/nightly/html/classvtkFlyingEdges3D.html the section **Online Examples:** is now blank. This section was working about a week ago.
The file used to generate the online examples is here: [vtk_vtk-examples_xref.json](https://github.com/Kitware/vtk-examples/blob/gh-pages/src/Coverage/vtk_vtk-examples_xref.json). I downloaded this file and tested it using [SelectExamples](https://examples.vtk.org/site/Python/Utilities/SelectExamples/) and it all works Ok, so no issues there.
Just guessing @jcfr but could it have been a change in this MR: !10281, Merge topic `consolidate-markdown-files-into-docs`?https://gitlab.kitware.com/vtk/vtk/-/issues/19035rename vtkButtonWidget method2023-12-11T21:26:29-05:00inglis-dlrename vtkButtonWidget methodchange vtkButtonWidget GetSliderRepresentation method name to GetButtonRepresentation.
Looks like a copy paste issue from vtkSliderWidget that was never resolved.change vtkButtonWidget GetSliderRepresentation method name to GetButtonRepresentation.
Looks like a copy paste issue from vtkSliderWidget that was never resolved.David GobbiDavid Gobbihttps://gitlab.kitware.com/vtk/vtk/-/issues/19033[Feature request] VTX: reuse previously saved mesh2023-08-03T06:43:45-04:00Massimiliano Leoni[Feature request] VTX: reuse previously saved meshI am using ADIOS2 to save some time-dependent simulation results in VTX format and I noticed that the mesh is saved multiple times even though it does not change. Does VTX support reusing a mesh saved at a previous time step?
Fides rece...I am using ADIOS2 to save some time-dependent simulation results in VTX format and I noticed that the mesh is saved multiple times even though it does not change. Does VTX support reusing a mesh saved at a previous time step?
Fides recently added support for this functionality [see https://gitlab.kitware.com/vtk/fides/-/issues/15 ] so I was wondering if it would be possible to have it in VTX too.
I am not entirely sure whether this is a VTK issue, an ADIOS2 issue or something else entirely so apologies if I'm asking in the wrong place :smile:https://gitlab.kitware.com/vtk/vtk/-/issues/19031VTK static lib bug report about vtkInformationXXKey when using more than one ...2023-07-26T21:52:51-04:00Liming ZhouVTK static lib bug report about vtkInformationXXKey when using more than one copy of vtk libsI have problem running my exe linked against static vtk libs, involving such as vtkImageSlice.
The case is I have a project consisting of a core lib and a test exe.
The core lib links against vtk static libs and the test exe links agains...I have problem running my exe linked against static vtk libs, involving such as vtkImageSlice.
The case is I have a project consisting of a core lib and a test exe.
The core lib links against vtk static libs and the test exe links against vtk static or dynamic libs.
After my debugging efforts, I found the problem is systematic.
The cause is each copy of vtk lib batch have their own static vtkInformationXXKey* (in vtkDataObject.h).
When passing a vtkImageData pointer from the test exe to the core lib, as input for vtkImageSlice, then running vtkImageSlice::update().
During which process, the codes `MapType::const_iterator i = this->Internal->Map.find(key);` in vtkInformation.cxx will result invalid iter.
that is because the Map using vtkInformationKey*(static vtkInformationXXKey* in vtkDataObject.h) as key instead of just std::string.
That means vtkInformation is systematiclly not fit for crossing copies of vtk libs(translation units)