MantisBT - ParaView
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0007799ParaView(No Category)public2008-10-10 09:582009-05-13 13:46
ReporterKen Moreland 
Assigned ToUtkarsh Ayachit 
PrioritylowSeverityminorReproducibilityhave not tried
StatusclosedResolutionfixed 
PlatformOSOS Version
Product Version 
Target Version3.6Fixed in Version 
Project
Topic Name
Type
Summary0007799: Snap to timesteps should only snap to timesteps of visibile objects
DescriptionAn annoying "feature" of snap to timesteps animation is that it will snap to timesteps of objects that no one is looking at. This is problematic when, say, someone loads in data and then uses the temporal shift scale filter to change the time range. You really only want to snap to the timesteps after the temporal shift scale, but ParaView also snaps to timesteps of the original data, which may even be outside the range of the desired time.

To observe, load the blow file series (found FileSeries directory of the ParaView data), warp by vector, and add a temporal shift scale with shift and scale both 10. Play the animation. You will see that the first part of the animation does nothing because it is playing timesteps from the original data set that are outside of the range of the desired time.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
parent of 0007361closed Utkarsh Ayachit animation inspector 'snap to time' steps madness 
child of 0007362closed Utkarsh Ayachit Time modes not useful 
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2008-10-10 09:58Ken MorelandNew Issue
2008-10-10 10:00Ken MorelandNote Added: 0013838
2008-10-14 02:24John BiddiscombeRelationship addedchild of 0007362
2008-10-14 02:25John BiddiscombeRelationship addedparent of 0007361
2009-02-17 14:18Utkarsh AyachitCategory3.6 => Pending
2009-02-20 13:01Utkarsh AyachitStatusbacklog => tabled
2009-02-20 13:01Utkarsh AyachitAssigned To => Utkarsh Ayachit
2009-02-20 13:01Utkarsh AyachitPrioritynormal => low
2009-02-20 13:01Utkarsh AyachitCategoryPending => 3.6
2009-02-26 15:14Utkarsh AyachitNote Added: 0015433
2009-02-26 15:14Utkarsh AyachitStatustabled => @80@
2009-02-26 15:14Utkarsh AyachitResolutionopen => fixed
2009-02-26 19:24Ken MorelandNote Added: 0015441
2009-02-27 08:08Utkarsh AyachitNote Added: 0015450
2009-03-10 16:24Alan ScottStatus@80@ => closed
2009-03-10 17:59Alan ScottStatusclosed => @80@
2009-03-10 17:59Alan ScottNote Added: 0015626
2009-03-10 17:59Alan ScottStatus@80@ => closed
2009-05-13 13:46Utkarsh AyachitTarget Version => 3.6
2011-06-16 13:10Zack GalbreathCategory => (No Category)

Notes
(0013838)
Ken Moreland   
2008-10-10 10:00   
One complication to implementing this is dealing with the situation where all objects with time steps become invisible. In this case, the animation mode has to switch from snap to timesteps back to something like sequence. This situation is common when, for example, loading a time series and then adding the temporal interpolator.
(0015433)
Utkarsh Ayachit   
2009-02-26 15:14   
Now the context menu on the pipeline browser
has a new checkbox "Ignore Time". If checked, the source will no longer be
considered when determining the timesteps or timeranges for the animation.
Whenever a temporal filter i.e. a filter with "TimestepValues" property is
created, its input is automatically updated to "ignore time" so that the inputs
time will not be considered in the animation (unless of course the user forces
it). This may be a more acceptable solution than creating a new play mode (as
suggedsted by BUG 0007362) or using timesteps from visible objects (as suggested
by BUG 0007799).

/cvsroot/ParaView3/ParaView3/Applications/OverView/Core/MainWindow.cxx,v <-- Applications/OverView/Core/MainWindow.cxx
new revision: 1.2; previous revision: 1.1
/cvsroot/ParaView3/ParaView3/Qt/Components/pqClientMainWindow.cxx,v <-- Qt/Components/pqClientMainWindow.cxx
new revision: 1.5; previous revision: 1.4
/cvsroot/ParaView3/ParaView3/Qt/Components/pqMainWindowCore.cxx,v <-- Qt/Components/pqMainWindowCore.cxx
new revision: 1.307; previous revision: 1.306
/cvsroot/ParaView3/ParaView3/Qt/Components/pqMainWindowCore.h,v <-- Qt/Components/pqMainWindowCore.h
new revision: 1.101; previous revision: 1.100
/cvsroot/ParaView3/ParaView3/Qt/Components/pqPipelineBrowserContextMenu.cxx,v <-- Qt/Components/pqPipelineBrowserContextMenu.cxx
new revision: 1.20; previous revision: 1.19
/cvsroot/ParaView3/ParaView3/Qt/Components/pqPipelineBrowserContextMenu.h,v <-- Qt/Components/pqPipelineBrowserContextMenu.h
new revision: 1.8; previous revision: 1.7
/cvsroot/ParaView3/ParaView3/Qt/Components/pqPipelineMenu.cxx,v <-- Qt/Components/pqPipelineMenu.cxx
new revision: 1.7; previous revision: 1.6
/cvsroot/ParaView3/ParaView3/Qt/Components/pqPipelineMenu.h,v <-- Qt/Components/pqPipelineMenu.h
new revision: 1.7; previous revision: 1.6
/cvsroot/ParaView3/ParaView3/Qt/Components/Resources/UI/pqClientMainWindow.ui,v <-- Qt/Components/Resources/UI/pqClientMainWindow.ui
new revision: 1.2; previous revision: 1.1
/cvsroot/ParaView3/ParaView3/Qt/Core/pqPendingDisplayManager.cxx,v <-- Qt/Core/pqPendingDisplayManager.cxx
new revision: 1.28; previous revision: 1.27
/cvsroot/ParaView3/ParaView3/Qt/Core/pqPipelineFilter.cxx,v <-- Qt/Core/pqPipelineFilter.cxx
new revision: 1.10; previous revision: 1.9
/cvsroot/ParaView3/ParaView3/Qt/Core/pqPipelineFilter.h,v <-- Qt/Core/pqPipelineFilter.h
new revision: 1.8; previous revision: 1.7
/cvsroot/ParaView3/ParaView3/Qt/Core/pqTimeKeeper.cxx,v <-- Qt/Core/pqTimeKeeper.cxx
new revision: 1.18; previous revision: 1.17
/cvsroot/ParaView3/ParaView3/Qt/Core/pqTimeKeeper.h,v <-- Qt/Core/pqTimeKeeper.h
new revision: 1.11; previous revision: 1.10
(0015441)
Ken Moreland   
2009-02-26 19:24   
What happens when the said temporal filter gets deleted? Does the input have its Ignore Time set to 1 or does it stay at 0? Either answer seems wrong.
(0015450)
Utkarsh Ayachit   
2009-02-27 08:08   
When the temporal filter gets deleted, it's input's "Ignore Time" gets set back to 0 -- kind of analogous to what happens with visibility. If we changed the timekeeper to use timesteps from "visible" sources, that's exactly what will happen as well. Any alternatives?
(0015626)
Alan Scott   
2009-03-10 17:59   
Tested client/remote server using test described above.