MantisBT - ParaView
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0009564ParaViewFeaturepublic2009-09-17 18:122016-08-12 09:57
ReporterKen Moreland 
Assigned Touser3172 
PriorityhighSeverityminorReproducibilityhave not tried
StatusclosedResolutionmoved 
PlatformOSOS Version
Product Version 
Target VersionFixed in Version 
Project
Topic Name
Type
Summary0009564: Append attributes should work with arrays of same name
DescriptionIf you try to run append attributes on two or more data sets with arrays of the same name, then only one will be accessible in the output. For a simple test, do the following.

1. Create a Wavelet source. Apply with default values.
2. Create another Wavelet source. Change one or more of the Mag parameters. Apply.
3. Select both Wavelet sources in the pipeline browser and add the Append Attributes filter. Apply.

Note that in the output, only one of the RTData arrays is visible. In cases like this, the filter should mangle the names so that both are visible. For example, change the names to RTData_0 and RTData_1 for the respective inputs.

Note that this situation is not all that uncommon. It can happen, for example, when running the same simulation with different initial conditions. You get results with the same mesh and field values with the same names but different values. The append attributes could be a helpful feature in comparing the values, if it worked.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
has duplicate 0006891closed Utkarsh Ayachit Append attributes should handle attributes with the same name 
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2009-09-17 18:12Ken MorelandNew Issue
2010-04-16 12:34Utkarsh AyachitStatusbacklog => tabled
2010-04-16 12:34Utkarsh AyachitAssigned To => user3172
2010-04-16 12:34Utkarsh AyachitTarget Version3.8 => 3.8.2
2010-06-15 09:44Utkarsh AyachitTarget Version3.8.2 => 3.8.1
2010-07-12 14:00Utkarsh AyachitTarget Version3.8.1 => 3.10
2010-09-10 14:23Alan ScottNote Added: 0022183
2010-09-10 14:23Alan ScottPrioritynormal => high
2010-11-08 12:08Utkarsh AyachitNote Added: 0023064
2010-11-08 13:15Utkarsh AyachitTarget Version3.10 => 4.0
2011-01-18 10:40Robert MaynardRelationship addedhas duplicate 0006891
2011-06-16 13:10Zack GalbreathCategoryFeature Request => Feature
2016-08-12 09:57Kitware RobotNote Added: 0037714
2016-08-12 09:57Kitware RobotStatusexpired => closed
2016-08-12 09:57Kitware RobotResolutionopen => moved

Notes
(0022183)
Alan Scott   
2010-09-10 14:23   
Changing priority to high. I have a user asking for it.
(0023064)
Utkarsh Ayachit   
2010-11-08 12:08   
This is not as trivial as it sounds. If any filter changes the name of an input array, then it has to create a "deep copy" of the input data array and then change the name. Thus just to change the name, we are increasing the memory footprint by twice.

So before we do this, we need to fix vtkAbstractArray's to use vtkArray as the internal data container and then add API to ShallowCopy() the heavy data around. This would require ensuring that vtkArray does the realloc and other memory management as optimally as data-array-template.
(0037714)
Kitware Robot   
2016-08-12 09:57   
Resolving issue as `moved`.

This issue tracker is no longer used. Further discussion of this issue may take place in the current ParaView Issues page linked in the banner at the top of this page.