[vtk-developers] Problem with vtkStreamingDemandDrivenPipeline

[Crossno, Patricia J]

I am writing a filter for unstructured data (vtkVortexCore) and I
recently (Friday) updated my copy of VTK.  Since that time, I've been
getting the following error message:

ERROR: In
C:\vtk_stuff\ParaView\VTK\Filtering\vtkStreamingDemandDrivenPipeline.cxx
, line 549
vtkStreamingDemandDrivenPipeline (0x03E8DDE8): No maximum number of
pieces has been set in the information for output port 0 on algorithm
vtkVortexCore(0x03E85FA0).

Looking at the revision log for vtkStreamingDemandDrivenPipeline.cxx, I
see the following last entry:
Revision 1.28 - (view
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?rev
=1.28&view=markup> ) (download
</cgi-bin/viewcvs.cgi/*checkout*/Filtering/vtkStreamingDemandDrivenPipel
ine.cxx?rev=1.28> ) (as text
</cgi-bin/viewcvs.cgi/*checkout*/Filtering/vtkStreamingDemandDrivenPipel
ine.cxx?content-type=text%2Fplain&rev=1.28> ) (annotate
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?ann
otate=1.28> ) - [select for diffs]
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?r1=
1.28&view=log>  
Tue Mar 29 15:48:48 2005 UTC (6 days, 5 hours ago) by blue 
Branch: MAIN
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?onl
y_with_tag=MAIN>  
CVS Tags: HEAD
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?onl
y_with_tag=HEAD> , ParaView-RenderModule2
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?onl
y_with_tag=ParaView-RenderModule2>  
Branch point for: ParaView-RenderModule2-branch
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?onl
y_with_tag=ParaView-RenderModule2-branch>  
Changes since 1.27: +14 -4 lines 
Diff to previous 1.27
</cgi-bin/viewcvs.cgi/Filtering/vtkStreamingDemandDrivenPipeline.cxx?r1=
1.27&r2=1.28>  
BUG: Fixing a bug in the CopyEntry functionality of Integer keys
revealed that MAXIMUM_NUMBER_OF_PIECES was not correctly being handled
if the input was structured data.

Is it possible that this change has created a problem for unstructured
data, or that the extent of the original bug was not limited to
structured data?
		Pat

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20050404/abb33532/attachment.html>