Hi Matt,
The output of the tube filter is not triangulated: it is represented in a
different mesh structure (don't ask me which, I'd have to dig up my VTK
documentation and loose a couple of hours looking in it).
The quick workaround is to apply a triangle filter to triangulate the
mesh:
poly_data = mlab.pipeline.triangle_filter(ply_data).outputs[0]
then the 'polys' attribute of the poly_data is well-behaved.
A as side remark, it would be useful to apply a stripper module to your
source before running the tube filter: it will connect the segment into
continuous lines, and the resulting tubes will be more continuous:
tube = mlab.pipeline.tube(mlab.pipeline.stripper(pts))
Finally, the warning about computing normals is fairly harmless. However,
a bit of CleanPolyData can remove the coincident geometry that VTK
dislikes:
lines = mlab.pipeline.user_defined(pts, filter='CleanPolyData')
Hope this helps,
Gaël
On Fri, Feb 24, 2012 at 05:50:36PM -0800, Matt Goodman wrote:
> Much thanks. That was the first place that I looked, but I was a bit
> confused by the following output, so I figured I was poking about in the
> wrong place.
> >From there, I do:
> ply_data = tube.outputs[0]
> print ply_data.points, ply_data.points.to_array()
> print ply_data.polys, ply_data.polys.to_array()
> When I print the poly-data's points attribute I get what I expect, but the
> polys attribute seems devoid of the connectivity data that I am used to
> seeing there:
> (output follows)
> vtkPoints (000000000E89F190)
> Debug: Off
> Modified Time: 813051
> Reference Count: 4
> Registered Events:
> Registered Observers:
> vtkObserver (000000000E7DBB00)
> Event: 33
> EventName: ModifiedEvent
> Command: 000000000E7DB0B0
> Priority: 0
> Tag: 1
> Data: 00000000108208D0
> Data Array Name: Points
> Number Of Points: 85404
> Bounds:
> Xmin,Xmax: (-219.164, 618.399)
> Ymin,Ymax: (-116.33, 665.014)
> Zmin,Zmax: (-234.603, 904.209)
> [[ 478.18218994 249.84048462 236.28182983]
> [ 478.12219238 250.10919189 236.24630737]
> [ 478.32241821 250.28787231 236.17526245]
> ...,
> [ 252.15283203 62.9306488 252.89715576]
> [ 247.82270813 60.4306488 252.89715576]
> [ 247.82270813 55.4306488 252.89715576]]
> vtkCellArray (000000000E8A0700)
> Debug: Off
> Modified Time: 804722
> Reference Count: 52
> Registered Events:
> Registered Observers:
> vtkObserver (000000000E7DBB00)
> Event: 33
> EventName: ModifiedEvent
> Command: 000000000E7DB350
> Priority: 0
> Tag: 3
> Number Of Cells: 0
> Insert Location: 0
> Traversal Location: 0
> []
> (note empty polys object . . .)
> I get a few warnings about being able to compute surface normals (there may
> be a few coincident points kicking around in there), but the figure draws
> just fine (attached), so I know the data has to be somewhere in that
> object. . .
> The warnings look like this:
> Generic Warning: In
> C:\pisi\tmp\VTK-5.6.0-1\work\VTK\Filtering\vtkPolyLine.cxx, line 104
> Coincident points in polyline...can't compute normals
> Warning: In C:\pisi\tmp\VTK-5.6.0-1\work\VTK\Graphics\vtkTubeFilter.cxx,
> line 225
> vtkTubeFilter (000000000EE0C330): Could not generate normals for line.
> Skipping to next.
> Are these possibly related?
> Code and data attached, thanks for reading. . .
> --Matthew Goodman
> =====================
> Check Out My Website: http://craneium.net
> Find me on LinkedIn: http://tinyurl.com/d6wlch
> On Fri, Feb 24, 2012 at 3:15 PM, Gael Varoquaux <
> gael.varoqu...@normalesup.org> wrote:
> > Hi Matt,
> > On Fri, Feb 24, 2012 at 02:24:55PM -0800, Matt Goodman wrote:
> >> I am trying to extract the mesh data from the generated tube object,
> >> but I am having trouble finding the generated polygon data.
> > The tube filter, as any filter, shoudl have an 'outputs' attribute,
> > giving the list of VTK datasets that it outputs. The first on of them
> > should be a polydata:
> > http://github.enthought.com/mayavi/mayavi/data.html#polydata
> > Hope this helps,
> > Gaėl
--
Gael Varoquaux
Researcher, INRIA Parietal
Laboratoire de Neuro-Imagerie Assistee par Ordinateur
NeuroSpin/CEA Saclay , Bat 145, 91191 Gif-sur-Yvette France
Phone: ++ 33-1-69-08-79-68
http://gael-varoquaux.info
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
MayaVi-users mailing list
MayaVi-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mayavi-users
