More cleaning required

[John Biddiscombe]

With regard to new vtkCleanPolyData and vtkQuantizePolyDataPoints, it occurs
to me that suppose you have a long thin box...

if the points are quantized (or cleaned) to some suitable value, its
possible that the box will collapse to form a sheet. If the quantization is
chosen such that the four corners of the long side are separated by more
than the quantization, then the side will be fine, but the thin separation
of
the two box sides may be small and the two facets will in fact become
identical.

The ends will collapse to a line and be correctly handled, but the two faces
that are now identical are effectively a degenerate pair.

I'd like to make a seperate
vtkDegenerateCellFilter which would check each cell (against the others) and
remove any which were identical (note that cell ID order would be immaterial
in this case).

The sensible option would be to put the Cell point ID's into a set and do a
difference operation on each other set (using a cell locator to speed up
matters), any cells which were in effect identical could be removed
painlessly.

Firstly - is there a better way - and secondly, should one use the STL <set>
container classes for ease or is there an objection to using STL inside vtk
filters.

The cleanpolydata and quantize filters are ready to be checked in, I've made
a regression test for the quantize filter it's a cxx file not a TCL one as I
can't get tcl/vtk running (sob sob blubber), I'm not sure what to do with
the "valid" test ppm image generated. Can anyone advise.

Thanks

John B



--------------------------------------------------------------------
This is the private VTK discussion list. Please keep messages on-topic.
Check the FAQ at: <http://public.kitware.com/cgi-bin/vtkfaq>
To UNSUBSCRIBE, send message body containing "unsubscribe vtkusers" to
<majordomo at public.kitware.com>. For help, send message body containing
"info vtkusers" to the same address.
--------------------------------------------------------------------