MantisBT - VTK
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0003691VTK(No Category)public2006-09-01 07:342013-04-05 19:57
Reportersoerengebbert 
Assigned ToZhanping Liu 
PriorityurgentSeveritymajorReproducibilityalways
StatusclosedResolutionopen 
PlatformOSOS Version
Product Version 
Target VersionFixed in Version 
Project
Type
Summary0003691: Contouring of new nonlinear cells incorrect
DescriptionThe contour is not calculated correctly, because the point id numbering
while contouring is incorrect in most of the new nonlinear cell implementations.
I have fixed this bug.

I didnt recognized this bug ealier, because i have tested the contouring
with only one cell .... .

Futhermore i have extended the quadCellConsistency.cxx test with most of
the new cell types.
Unfortunately the vtkQuadraticLinearWedge fails the "face" test, because
the of vtkQuadraticLineareQuad
numbering.

The cvs diff is attached.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Filesdiff VTK-CVS_Filtering.diff (9,946) 1969-12-31 19:00
https://www.vtk.org/Bug/file/5643/VTK-CVS_Filtering.diff
? vtkTriQuadraticHexahedronBoard.vtk (28,866) 1969-12-31 19:00
https://www.vtk.org/Bug/file/5653/vtkTriQuadraticHexahedronBoard.vtk
? vtkQuadraticLinearWedgeBoard.vtk (26,146) 1969-12-31 19:00
https://www.vtk.org/Bug/file/5654/vtkQuadraticLinearWedgeBoard.vtk
Issue History
Date ModifiedUsernameFieldChange
2008-02-06 10:00Jeff BaumesAssigned ToBerk Geveci => David Thompson (old)
2008-09-23 11:27Berk GeveciAssigned ToDavid Thompson (old) => Zhanping Liu
2008-10-14 17:50Zhanping LiuNote Added: 0013880
2008-10-14 17:50Zhanping LiuStatustabled => @80@
2011-06-16 13:11Zack GalbreathCategory => (No Category)
2013-04-05 19:57Berk GeveciStatuscustomer review => closed

Notes
(0004863)
Mathieu Malaterre   
2006-09-05 19:36   
Hi Soern,

  Thanks a lot for looking into this. Could you please send me the dataset your are using to reproduce the bug. In the past I had to go through the cell and indeed did the exact opposite fix to fix an out of bound read access. I believe the current code is correct.

Mathieu
(0004867)
soerengebbert   
2006-09-06 09:22   
Dear Mathieu,
i have attached two vtk files with two different
celltypes.
You can verify this bug with ParaView and
the cut filter.
With this fix we are now able to cut
correctly with our FEM tool.

Soeren
(0013880)
Zhanping Liu   
2008-10-14 17:50   
The fix suggested by the reporter, "VTK-CVS_Filtering.diff", involves the following five files:

    ===============================================================
    vtkQuadraticLinearWedge.cxx --- VTK/Filtering/
    vtkBiQuadraticQuad.cxx --- VTK/Filtering/
    vtkBiQuadraticQuadraticWedge.cxx --- VTK/Filtering/
    vtkTriQuadraticHexahedron.cxx --- VTK/Filtering/
    quadCellConsistency.cxx --- VTK/Filtering/Testing/Cxx/
    ===============================================================

    These files were last updated on Jan 16, Jan 16, May 1, Jan 18, and Jan 16, respectively, all in 2008. All the suggested segments of code (based on "VTK-CVS_Filtering.diff") were fully incorporated into these five files at the respective times and they continue to be in effect now.

    Thus the reported problem does not exist any more and these five updated files could and can pass CTests.

    The previous developer(s) just forgot to upchage the status of this problem to "resolved". Now this "bug" report can be closed.

    Thank soerengebbert so much for the suggested fix.