MantisBT - ParaView
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0013386ParaView(No Category)public2012-08-22 21:042012-10-29 17:04
ReporterAlan Scott 
Assigned ToUtkarsh Ayachit 
PriorityhighSeverityminorReproducibilityhave not tried
StatusclosedResolutionfixed 
PlatformOSOS Version
Product Versiongit-master 
Target VersionFixed in Version3.98.0 
ProjectSandia
Topic Name13386_fix_cth_derived_quantities
Typeincorrect functionality
Summary0013386: Failure reading CTH files - Compute Derived Variables
DescriptionI have a CTH dataset that is failing on read.
Windows (Linux also fails), local server, master.
Open dataset. All vars on. Apply.

Error print is as follows:
ERROR: In ..\..\..\..\..\src\ParaView3\VTK\Common\DataModel\vtkDataSet.cxx, line 437
vtkRectilinearGrid (189FF048): Cell array Derived Density - 1 with 1 components, has only 0 tuples but there are 74884 cells


The error is here:
file: vtkDataSet.cxx, line 437.

The cause of the error message is as follows:
array name == Derived Density - 1
number of tuples == 0
number of cells = 74884

We are failing on an if test making sure that the number of tuples == number of cells.

Workaround - in the reader, before clicking "Apply", deselect "Compute Derived Variables".

This is a show stopper for 4.0.

I will give Utkarsh the dataset. It is named flat_2d-crash-snl-do-not-release.tar.gz

Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2012-08-22 21:04Alan ScottNew Issue
2012-08-22 21:05Alan ScottNote Added: 0029048
2012-08-22 21:05Alan ScottAssigned To => Alan Scott
2012-08-22 21:05Alan ScottStatusbacklog => todo
2012-09-05 18:23Utkarsh AyachitTopic Name => 13386_fix_cth_derived_quantities
2012-09-05 18:23Utkarsh AyachitNote Added: 0029133
2012-09-05 18:23Utkarsh AyachitStatustodo => gatekeeper review
2012-09-05 18:23Utkarsh AyachitFixed in Version => git-next
2012-09-05 18:23Utkarsh AyachitResolutionopen => fixed
2012-09-05 18:23Utkarsh AyachitAssigned ToAlan Scott => Utkarsh Ayachit
2012-09-07 14:50Utkarsh AyachitFixed in Versiongit-next => git-master
2012-09-07 14:50Utkarsh AyachitStatusgatekeeper review => customer review
2012-09-07 14:50Utkarsh AyachitNote Added: 0029143
2012-09-19 19:46Alan ScottNote Added: 0029225
2012-09-19 19:46Alan ScottStatuscustomer review => closed
2012-10-29 17:04Utkarsh AyachitFixed in Versiongit-master => 3.98.0

Notes
(0029048)
Alan Scott   
2012-08-22 21:05   
Changing to todo.
(0029133)
Utkarsh Ayachit   
2012-09-05 18:23   
commit 5c0213b0d59942514253ec02f1b6d466c3544c13
Author: Utkarsh Ayachit <utkarsh.ayachit@kitware.com>
Date: Wed Sep 5 18:17:42 2012 -0400

    BUG 0013386. Fixed computation of derived values.
    
    1. vtkSpyPlotBlock was using a incorrect SavedRealDims size of 6. We just need
       3 elements for a dimension. Fixed that. That cleans up the incorrect
       initialization of SavedRealDims too.
    
    2. vtkSpyPlotBlock::FixInformation() was increamenting "j" for 1 extra time when
       this->Dimensions[] was 1. Fixed that. This would have caused invalid writes
       for 2D datasets no in the XY plane.
    
    3. SavedRealDims is set to valid values only when ghost cells were present in
       the data and we need to "fix" them. Hence fixed GetCellVolume() and
       ComputeDerivedVariables() calls to use SavedRealDims only when valid. This
       was causing the bug reported. Since some grids didn't have the extra ghost
       cells that needed to be fixed, the SavedRealDims was not set for those grids
       and we ended up with invalid dimensions being used when computing the derived
       quantities.
(0029143)
Utkarsh Ayachit   
2012-09-07 14:50   
merged in master if applicable
(0029225)
Alan Scott   
2012-09-19 19:46   
Excellent.

Tested remote server, Linux, master.