ParaView issueshttps://gitlab.kitware.com/paraview/paraview/-/issues2017-11-19T19:27:21-05:00https://gitlab.kitware.com/paraview/paraview/-/issues/17235Find Data dialog is very slow to show up2017-11-19T19:27:21-05:00Utkarsh AyachitFind Data dialog is very slow to show upSteps:
1. Start ParaView
2. Open **can.ex2**, Apply
3. Using [![image](/uploads/576976e42f0a9f1046d96c8cab20f4e2/image.png)](*), create a selection on the block.
4. Now hit **v** to show the Find Data dialog -- it takes a long time t...Steps:
1. Start ParaView
2. Open **can.ex2**, Apply
3. Using [![image](/uploads/576976e42f0a9f1046d96c8cab20f4e2/image.png)](*), create a selection on the block.
4. Now hit **v** to show the Find Data dialog -- it takes a long time to show up.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17222Screenshots of 2d views fails2017-06-27T05:16:26-04:00W. Alan ScottScreenshots of 2d views fails@cory-quammen @utkarsh.ayachit Showstopper bug for 5.3.0.
We can't take screenshots of 2d views. This worked on 5.2.0. Here is how to replicate.
* Windows or Linux, 5.3.0-RC2, local server. Windows binaries are from Kitware, Lin...@cory-quammen @utkarsh.ayachit Showstopper bug for 5.3.0.
We can't take screenshots of 2d views. This worked on 5.2.0. Here is how to replicate.
* Windows or Linux, 5.3.0-RC2, local server. Windows binaries are from Kitware, Linux is a superbuild.
* disk_out_ref.exo. (I turned on all variables, not needed.) Apply. (Wavelet would work).
* Plot over line. Apply.
* (Make sure the Plot over line filter and view is selected. I.e,, don't click around and unselect stuff.)
* File/ Save Screenshot. Make sure the Save only Selected View is selected.
* On Windows, this screenshot is black. On Linux, it is a weird patterned gray.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17204Color bar not displaying minus symbol for negative numbers2017-04-24T15:00:30-04:00Kenneth MorelandColor bar not displaying minus symbol for negative numbersWhile playing with ParaView 5.3 RC1 I noticed that the color bar appears to not show the minus sign for negative numbers (when running on my Windows desktop). To replicate, do the following:
1. Create a Plane source. Apply.
2. Create an...While playing with ParaView 5.3 RC1 I noticed that the color bar appears to not show the minus sign for negative numbers (when running on my Windows desktop). To replicate, do the following:
1. Create a Plane source. Apply.
2. Create an Elevation filter. Change Scalar Range to -1 to 1. Apply.
If you look at the resulting color bar, the minus symbols are missing so that it looks like the scalar range is from 1 to 1.
![ColorBarNoNegative](/uploads/e537bef8b74bef021daf390f13831439/ColorBarNoNegative.png)5.3https://gitlab.kitware.com/paraview/paraview/-/issues/17202Charts rendering without anti-aliasing with Qt 52018-12-29T13:14:30-05:00Utkarsh AyachitCharts rendering without anti-aliasing with Qt 5See following:
**Qt 4**:
![image](/uploads/63682ca0c6f1953f6c351ba2c81ec7a1/image.png)
**Qt 5**:
![image](/uploads/a06160f48ab40de26ea00a6e068e767e/image.png)
Notice how the lines are more jagged with Qt 5 than Qt 4 (both are ...See following:
**Qt 4**:
![image](/uploads/63682ca0c6f1953f6c351ba2c81ec7a1/image.png)
**Qt 5**:
![image](/uploads/a06160f48ab40de26ea00a6e068e767e/image.png)
Notice how the lines are more jagged with Qt 5 than Qt 4 (both are rendered with OpenGL2 backend).5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17201Delete splash screen2017-07-31T12:48:58-04:00W. Alan ScottDelete splash screenThe splash screen has outlived it's functionality. We now have the Startup screen, which presents the same information, along with beginning links, in a better format. Surprisingly, the splash screen still lives. Please delete it.The splash screen has outlived it's functionality. We now have the Startup screen, which presents the same information, along with beginning links, in a better format. Surprisingly, the splash screen still lives. Please delete it.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17157Static Qt 5 builds fail.2017-11-19T19:27:19-05:00Utkarsh AyachitStatic Qt 5 builds fail.ParaView version: **v5.2.0-350-gf4b1fec**
Static builds with Qt5 fail with following link errors:
```sh
lib/libCatalystScriptGeneratorPlugin.a(CatalystScriptGeneratorPlugin_Plugin.cxx.o): In function `CatalystScriptGeneratorPlugin_...ParaView version: **v5.2.0-350-gf4b1fec**
Static builds with Qt5 fail with following link errors:
```sh
lib/libCatalystScriptGeneratorPlugin.a(CatalystScriptGeneratorPlugin_Plugin.cxx.o): In function `CatalystScriptGeneratorPlugin_Plugin::CatalystScriptGeneratorPlugin_Plugin()::StaticCatalystScriptGeneratorPluginPluginInstance::StaticCatalystScriptGeneratorPluginPluginInstance()':
/home/utkarsh/Kitware/ParaView3/ParaViewBin-static/Plugins/CatalystScriptGenerator/CatalystScriptGeneratorPlugin_Plugin.cxx:187: undefined reference to `qt_static_plugin_CatalystScriptGeneratorPlugin()'
lib/libLagrangianParticleTracker.a(LagrangianParticleTracker_Plugin.cxx.o): In function `LagrangianParticleTracker_Plugin::LagrangianParticleTracker_Plugin()::StaticLagrangianParticleTrackerPluginInstance::StaticLagrangianParticleTrackerPluginInstance()':
/home/utkarsh/Kitware/ParaView3/ParaViewBin-static/Plugins/LagrangianParticleTracker/LagrangianParticleTracker_Plugin.cxx:190: undefined reference to `qt_static_plugin_LagrangianParticleTracker()'
lib/libNonOrthogonalSource.a(NonOrthogonalSource_Plugin.cxx.o): In function `NonOrthogonalSource_Plugin::NonOrthogonalSource_Plugin()::StaticNonOrthogonalSourcePluginInstance::StaticNonOrthogonalSourcePluginInstance()':
/home/utkarsh/Kitware/ParaView3/ParaViewBin-static/Plugins/NonOrthogonalSource/NonOrthogonalSource_Plugin.cxx:188: undefined reference to `qt_static_plugin_NonOrthogonalSource()'
lib/libSLACTools.a(SLACTools_Plugin.cxx.o): In function `SLACTools_Plugin::SLACTools_Plugin()::StaticSLACToolsPluginInstance::StaticSLACToolsPluginInstance()':
/home/utkarsh/Kitware/ParaView3/ParaViewBin-static/Plugins/SLACTools/SLACTools_Plugin.cxx:189: undefined reference to `qt_static_plugin_SLACTools()'
lib/libSierraPlotTools.a(SierraPlotTools_Plugin.cxx.o): In function `SierraPlotTools_Plugin::SierraPlotTools_Plugin()::StaticSierraPlotToolsPluginInstance::StaticSierraPlotToolsPluginInstance()':
/home/utkarsh/Kitware/ParaView3/ParaViewBin-static/Plugins/SierraPlotTools/SierraPlotTools_Plugin.cxx:186: undefined reference to `qt_static_plugin_SierraPlotTools()'
collect2: error: ld returned 1 exit status
ninja: build stopped: subcommand failed.
```5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17156Text in 3D View not rendered correctly2017-11-19T19:27:19-05:00Utkarsh AyachitText in 3D View not rendered correctlySteps to reproduce:
1. Text source, set **Text** to 0.467, Apply
2. Change **Font Properties** to change to size to 7. (you may have to incrementally lower this to get to buggy state if 7 doesn't show the issue).
![image](/uploads/6...Steps to reproduce:
1. Text source, set **Text** to 0.467, Apply
2. Change **Font Properties** to change to size to 7. (you may have to incrementally lower this to get to buggy state if 7 doesn't show the issue).
![image](/uploads/625a1028567a82c4bf94736d6a2788d2/image.png)
That's bad. The same bug affects labels in Scalar Bar too.5.3Shawn WaldonShawn Waldonhttps://gitlab.kitware.com/paraview/paraview/-/issues/17155If server dies during filter execution, the "Save State" prompt is not useable.2017-11-19T19:27:21-05:00Utkarsh AyachitIf server dies during filter execution, the "Save State" prompt is not useable.Try this:
1. Connect ParaView to pvserver (single rank is fine)
2. Create `Programmable Source` with the following script:
```python
import sys
sys.exit(1)
```
3. Apply
We're simulating a server crash here. On **Apply**, the se...Try this:
1. Connect ParaView to pvserver (single rank is fine)
2. Create `Programmable Source` with the following script:
```python
import sys
sys.exit(1)
```
3. Apply
We're simulating a server crash here. On **Apply**, the server will exit and you should get the following prompt
![image](/uploads/c3fcda5f38ae81603665b9bb2be5fc43/image.png)
However the dialog is not usable! One cannot click on any of the buttons!
Now try this:
1. Connect to pvserver (1 rank).
2. Just `Ctrl-C` to kill the pvserver processes.
This time the same dialog pops up, but is usable. I can click on it to save state etc.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17149color legend, Information tab and spreadsheet giving incorrect information2017-04-24T15:00:30-04:00W. Alan Scottcolor legend, Information tab and spreadsheet giving incorrect informationThis bug is set up by #15887. Here is a copy of that bug report, setting up this one.
* Dataset is Resample-Cone-DoNotRelease. This has two files, donor.g and recipient.g.
* Linux, master (pre 5.3.0), local server.
* Timestep 0.
*...This bug is set up by #15887. Here is a copy of that bug report, setting up this one.
* Dataset is Resample-Cone-DoNotRelease. This has two files, donor.g and recipient.g.
* Linux, master (pre 5.3.0), local server.
* Timestep 0.
* (Do not do this step, this functionality is broken.) (If desired, but not necessary, Edit/ Settings/ Clamp and update every timestep).
* Open donor.g. All vars on. Apply.
* Open recipient.g. All vars on. Apply.
* Filter/ Alphabetical/ ResampleWithDataset.
- Input == donor.g
- Source == recipient.g
- OK
- Advanced Properties tab.
- Turn on Pass Cell Array
- Turn on Pass Point Arrays
- Turn off Compute Tolerance. 1e-3.
- Apply.
* Paint by TEMPK.
Lets look at our data for the donor.g.
* Turn visibility off for ResampleWithDataset.
* Click on and turn visibility on for donor.g.
* Paint by TEMPK.
* Rescale to Data Range.
**First bug.** Notice that the min and max, as presented by the Color Legend, is 293.9 to 295.9. But, the Information tab shows the data running from 293.9 to 293.9. Try manually changing this to 293.9 to 294.9. Fails. Question - is this a bug, or is this just a feature of our "don't divide by zero" Color Legend implementation? Either way, needs to be fixed.
Lets look at our data for the ResampleWithDataset.
* Turn visibility off for donor.g.
* Click on and turn visibility on for ResampleWithDataset.
* Paint by TEMPK.
* Rescale to Data Range.
**Second bug.** The max has jumped down by two degrees. Range is now 0 to 293.9. I believe the reason is that the base of the cone is actually outside of the tolerance, and is being padded by zero. With regards to this bug, lets assume 0 is OK. Max range should not have jumped.
Lets look at our data in the Spreadsheet.
* Split horizontal. Spreadsheet. Should already be set up correctly. ResampleWithDataset, Point Data.
**Third bug.** Sort on TEMPK. Notice that everything is zero. This is wrong, the outside of the cone should be 293.9 degrees! Same problem with all of the other variables.
Note that I have marked this as required. The color legend min/max issue is required for the 5.3 release, the Spreadsheet may slip to 5.4.
Thanks team!5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17145Variables disappear on extract selection.2017-11-19T19:27:21-05:00W. Alan ScottVariables disappear on extract selection.A bug from a user. I will give the dataset to Utkarsh. (Note to self - workstation, mst..., cylinder_a_v3.0a.5.e.) (Note to Utkarsh, same dataset as #17081. Same restrictions apply.)
* ParaView master, local server, Linux.
* This d...A bug from a user. I will give the dataset to Utkarsh. (Note to self - workstation, mst..., cylinder_a_v3.0a.5.e.) (Note to Utkarsh, same dataset as #17081. Same restrictions apply.)
* ParaView master, local server, Linux.
* This dataset has three vectors of interest - Q_XX, Q_YY and Q_ZZ, which are orthogonal to each other.
* Read data, all vars on, apply.
* Select one cell. Extract Selection. Apply. Now, I have one cell to play with.
Notice that cell variable Q_XX has disappeared. It doesn't show up as a color to paint with, or in the Information tab. This is a bug.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17144Clamp and update every timestep is broken2017-11-19T19:27:19-05:00W. Alan ScottClamp and update every timestep is brokenClamp and update every timestep is broken. This is needed by a customer. Thus, marking must have.
* Master, Linux, local server.
* Edit/ Settings/ Transfer Function Reset Mode/ change to "Clamp and update every timestep".
* Open...Clamp and update every timestep is broken. This is needed by a customer. Thus, marking must have.
* Master, Linux, local server.
* Edit/ Settings/ Transfer Function Reset Mode/ change to "Clamp and update every timestep".
* Open Can.exo. All vars on. Apply.
* Color by ACCL. Step through numerous steps. The Color legend should update every timestep.
This is critical functionality for some users. Functionality lost after the 5.2.0 release. Marking as required for the 5.3 release.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17135Maximum time limit of 4095 for annotated .vtu file2017-11-19T19:27:20-05:00David BennMaximum time limit of 4095 for annotated .vtu fileUpon loading an annotated .vtu file with TimeStep annotations such as this into Paraview 5.2.0 (under 64 bit SUSE Linux):
`<DataArray type="Float64" Name="V" TimeStep="10560" format="appended" RangeMin="-87.934331219" RangeMax="-87.44...Upon loading an annotated .vtu file with TimeStep annotations such as this into Paraview 5.2.0 (under 64 bit SUSE Linux):
`<DataArray type="Float64" Name="V" TimeStep="10560" format="appended" RangeMin="-87.934331219" RangeMax="-87.443353026" offset="7801035604" />`
I see the following in Paraview:
![timemax4095](/uploads/7da3e44c98f680a4198757dbf81281b4/timemax4095.png)
such that a maximum of 4095 time steps are available, as opposed to the expected 10,560 time steps.
The number of time steps can be found in the TimeValues attribute of the UnstructuredGrid element:
`<UnstructuredGrid TimeValues="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 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 ...`
When the *non-annotated* .vtu file is loaded, the expected number of *frames* are visible.
There is an inherent limitation within Paraview that prevents more than 4095 time steps from being loaded/visualised.
A colleague found the place in the code where the 4095 limit occurs:
```
int vtkXMLReader::ReadPrimaryElement(vtkXMLDataElement* ePrimary)
{
const int tsMax = 4096;
double timevalues[tsMax];
int numTimeSteps =
ePrimary->GetVectorAttribute("TimeValues", tsMax, timevalues);
assert( numTimeSteps <= tsMax);
this->SetNumberOfTimeSteps( numTimeSteps );
...
```
Changing 4096 to a larger value (e.g. 65536) allows more time steps to be used within Paraview, but I wondered about a solution without an upper bound on the number of time steps. I spent time looking at the code being called and what was required.
vtkXMLReader::ReadPrimaryElement calls vtkXMLDataElement::GetVectorAttribute() which calls vtkXMLDataElement::vtkXMLDataElementVectorAttributeParse().
In ReadPrimaryElement(), only the number of time steps is required. This corresponds to the number of space-delimited tokens (time step numbers) in the TimeValues attribute. The timevalues double array passed to GetVectorAttribute() is not used for anything within ReadPrimaryElement().
Accordingly, I've modified the code locally to this:
```
int vtkXMLReader::ReadPrimaryElement(vtkXMLDataElement* ePrimary)
{
const char* timeValuesAttr = ePrimary->GetAttribute("TimeValues");
int numTimeSteps =
ePrimary->SpaceDelimitedTokenCount(timeValuesAttr);
this->SetNumberOfTimeSteps( numTimeSteps );
...
```
and added vtkXMLDataElement::SpaceDelimitedTokenCount():
```
int vtkXMLDataElement::SpaceDelimitedTokenCount(const char* str)
{
const int find_token = 1;
const int skip_token = 2;
const int end = 3;
int state = find_token;
int count = 0;
while(str && state != end)
{
switch(state)
{
case find_token:
// Iterate until non-space (token or EOLN: 0)
while(*str == ' ') ++str;
state = *str != 0 ? skip_token : end;
break;
case skip_token:
++count;
// Iterate past token until space or EOLN
while(*str && *str != ' ') ++str;
state = *str != 0 ? find_token : end;
break;
default:
state = end;
}
}
return count;
}
```
This may not be the best class for it. In any case, it works well.
Initially I wrote this in terms of a variant of vtkXMLDataElement::vtkXMLDataElementVectorAttributeParse():
```
int vtkXMLDataElement::SpaceDelimitedTokenCount(const char* str)
{
if(!str) { return 0; }
std::stringstream vstr;
vstr.imbue(std::locale::classic());
vstr << str;
char ch;
int count = 0;
while(vstr)
{
vstr >> ch;
count++;
}
return count-1;
}
```
I used the following simple test cases (with spaceDelimitedTokenCount2() referring to the second implementation above passed as a non-member function to testTokenCounter()):
```
void testTokenCounter(int (*f)(const char*))
{
const char* str1 = "1 2 3 4 5";
std::cout << f(str1) << std::endl;
const char* str2 = "1 2 3 4 5";
std::cout << f(str2) << std::endl;
const char* str3 = " 1 2 3 4 5 ";
std::cout << f(str3) << std::endl;
// Fails with spaceDelimitedTokenCount2()
const char* str4 = " 1 2 3 4 52 ";
std::cout << f(str4) << std::endl;
const char* str5 = " ";
std::cout << f(str5) << std::endl;
const char* str6 = " ";
std::cout << f(str6) << std::endl;
const char* str7 = "";
std::cout << f(str7) << std::endl;
const char* str8 = 0;
std::cout << f(str8) << std::endl;
std::cout << std::endl;
}
```
and found that the second approach yielded a count of 6 tokens instead of 5. I think the first implementation is less opaque in any case, at the cost of being a little more verbose.
I submit this problem and potential solution for your consideration.
From a git clone of VTK, I see:
> $ git rev-parse HEAD
> c0a653bf5fd488f7ae7bcd24e8321b542f4b5f6b5.3Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/17128quadratic tetrahedra failing with ghost cell generation2017-11-19T19:27:21-05:00W. Alan Scottquadratic tetrahedra failing with ghost cell generationI have a dataset that is failing with ghost cell generation. This is true for D3, Ghost Cell Generation, and Extract Surface. Ken and my speculation is that this case is either not being caught, or needs to be added.
Test as follows...I have a dataset that is failing with ghost cell generation. This is true for D3, Ghost Cell Generation, and Extract Surface. Ken and my speculation is that this case is either not being caught, or needs to be added.
Test as follows:
* Linux, 5.1.2, remote server. I am using 8 ranks.
* Open the spread file hollowSphere_fieldVars.e.48. Apply. (This dataset looks like the ice layer of Europa or a pressure vessel.).
* D3 or Ghost Cell Generator. Apply.
* Change Opacity to 0.2. Notice the processor boundaries.
Now, to show it work correctly, do the following:
* Linux, 5.1.2, remote server. I am using 8 ranks.
* Open the spread file hollowSphere.g.48. Apply. (This dataset looks like the ice layer of Europa).
* Tesselate. Apply.
* D3 or Ghost Cell Generator. Apply.
* Change Opacity to 0.2. Notice the processor boundaries are gone.
I will give the dataset to Utkarsh. It is UUR.
I need either this bug fixed (this one preferrably) or #17137 and #17136 5.3Cory Quammencory.quammen@kitware.comCory Quammencory.quammen@kitware.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/17098Select Points Through slowed down by Find Data2017-04-24T15:00:31-04:00Andrew BauerSelect Points Through slowed down by Find DataSteps to reproduce:
* Create Sphere source
* Set Phi resolution to 800
* Use Select Points Through and select bottom half of sphere
* Open Find Data dialog
* Close Find Data dialog (no need to do anything beyond this)
* Use Select...Steps to reproduce:
* Create Sphere source
* Set Phi resolution to 800
* Use Select Points Through and select bottom half of sphere
* Open Find Data dialog
* Close Find Data dialog (no need to do anything beyond this)
* Use Select Points Through to roughly select the same points again
The first Select Points Through is significantly faster than the second one. Note that this difference may not be that noticeable if the Select Points Through is very fast. This has been noticed on both Linux and Windows with PV 5.2.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17088SurfaceLIC doesn't work when FXAA is enabled2017-04-24T15:00:31-04:00Utkarsh AyachitSurfaceLIC doesn't work when FXAA is enabled+ Start ParaView, load `SurfaceLIC` plugin
+ Open disk_out_ref.ex2, all variables, apply, render as representation = "Surface LIC". All's good.
+ Now enable `Use FXAA` from the Settings |Render View pane, and interact with the view. Th...+ Start ParaView, load `SurfaceLIC` plugin
+ Open disk_out_ref.ex2, all variables, apply, render as representation = "Surface LIC". All's good.
+ Now enable `Use FXAA` from the Settings |Render View pane, and interact with the view. The rendering will be messed up.5.3Allison Vacantialliepiper16@gmail.comAllison Vacantialliepiper16@gmail.comhttps://gitlab.kitware.com/paraview/paraview/-/issues/17081Glyph filter failing on cell data2017-04-24T15:00:31-04:00W. Alan ScottGlyph filter failing on cell dataA bug from a user. I will give the dataset to Utkarsh. (Note to self - workstation, mst..., cylinder_a_v3.0a.5.e.)
* ParaView 5.1.2, local server, Linux.
* This dataset has three vectors of interest - Q_XX, Q_YY and Q_ZZ, which are...A bug from a user. I will give the dataset to Utkarsh. (Note to self - workstation, mst..., cylinder_a_v3.0a.5.e.)
* ParaView 5.1.2, local server, Linux.
* This dataset has three vectors of interest - Q_XX, Q_YY and Q_ZZ, which are orthogonal to each other.
* Read data, all vars on, apply.
* Select one cell. I selected one on the surface, down at the bottom edge of the can. Extract Selection. Apply. Now, I have one cell to play with.
* Glyph. Vector == Q_XX, Scale Mode vector. Surprisingly, I got no vector at all. This is a bug.
* Choose incoming dataset. Split screen. Spreadsheet view. Look at Q_XX, Q_YY and Q_ZZ. Spread the Q_XX, and Q_YY, and Q_ZZ columns. Yep, they look orthogonal to me.
* Reset Session.
* Read data, all vars on, apply.
* Cell data to Point data filter. We now have the vectors at the corners of the cells.
* Glyph as above. Q_XX. Apply.
* Be sure pipeline browser is on the Cell Data to Point Data filter. Glyph as above. Q_YY. Apply.
* Be sure pipeline browser is on the Cell Data to Point Data filter. Glyph as above. Q_ZZ. Apply.
The three vectors are well behaved, and furthermore, are orthogonal.
So, with this dataset, glyphing of cell data doesn't work correctly, glyphing point data does work correctly.
Marking as important, since it misrepresents data.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17079Clear Opengl framebuffers on open2017-11-19T19:27:20-05:00W. Alan ScottClear Opengl framebuffers on openPlease clear the Opengl frame buffers upon ParaView open. Actively overwrite the front and back frame buffers. Currently, on Linux, when you run Paraview , visualize something and close ParaView, (or any other 3d program), and then reo...Please clear the Opengl frame buffers upon ParaView open. Actively overwrite the front and back frame buffers. Currently, on Linux, when you run Paraview , visualize something and close ParaView, (or any other 3d program), and then reopen ParaView, the old frame buffer pictures show. This is potentially a big issue if someone is showing a coworker something, and said co-worker sees something he/she should not see.
Talk to me if you want a more specific example.5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17077*.pvsm files not being shown in `Load State` dialog2017-11-19T19:27:20-05:00Utkarsh Ayachit*.pvsm files not being shown in `Load State` dialogSee the images below. The `state.pvsm` file doesn't show up until I choose `All files(*)`.
![image](/uploads/34f19e7a011f1f4aa558e69074de53c4/image.png)
![image](/uploads/f0a9e8a2b6e173cef2c1651072013c30/image.png)See the images below. The `state.pvsm` file doesn't show up until I choose `All files(*)`.
![image](/uploads/34f19e7a011f1f4aa558e69074de53c4/image.png)
![image](/uploads/f0a9e8a2b6e173cef2c1651072013c30/image.png)5.3Shawn WaldonShawn Waldonhttps://gitlab.kitware.com/paraview/paraview/-/issues/17028PVD files pointing to .pvtu files with multiple pieces are not read correctly...2017-11-21T16:31:57-05:00W. Alan ScottPVD files pointing to .pvtu files with multiple pieces are not read correctly on parallel remote serverWhen you have a .pvd file, which has time data and calls a pvtu file per timestep, and these pvtu files point to numerous .vtu files, ParaView acts differently remote server than local server.
* Local server version of ParaView. tr...When you have a .pvd file, which has time data and calls a pvtu file per timestep, and these pvtu files point to numerous .vtu files, ParaView acts differently remote server than local server.
* Local server version of ParaView. tried with 5.1.2 and older version of 5.2.0 master. Linux.
* Open file rodcylinder.pvtu from the dataset rodcylinder. (I will give this to Utkarsh).
* Information tab. Note that we have 61343 cells.
* Remote server. I used 16 ranks, but I bet it shows with 1 or 2. Linux.
* Open file rodcylinder.pvtu from the dataset rodcylinder.
* Information tab. Note that we have 43218 cells. Only the first of the two .vtu files were read!5.3Utkarsh AyachitUtkarsh Ayachithttps://gitlab.kitware.com/paraview/paraview/-/issues/17023Warp Scale Factor should default to 12017-11-21T16:31:54-05:00Kenneth MorelandWarp Scale Factor should default to 1The Warp By Vector and Warp By Scalar filters each has an option named Scale Factor that scales the displacement by a multiplier. The initial value in the properties panel is set to some odd value based on the bounds of the data and scal...The Warp By Vector and Warp By Scalar filters each has an option named Scale Factor that scales the displacement by a multiplier. The initial value in the properties panel is set to some odd value based on the bounds of the data and scale of the displacement field (I guess). This makes absolutely no sense.
Displacement vectors are almost always defined in spatial units. Thus, the only default that makes sense is 1.5.3Shawn WaldonShawn Waldon