[vtkusers] Unstructured grid to structured grid?

[Chris Marsh]

Everything successfully works with the existing unstructured mesh instead
of creating a 2nd Delaunauy2D. I did this via a geometry filter and a
flattening transform filter.
Are contributions to the wiki welcome? If so, I'd like to contribute a
minimum example that is slightly more in-depth versus the existing examples
for others like myself.


Chris Marsh
PhD Student
chrismarsh.ca

13 Kirk Hall
University of Saskatchewan

On Wed, Sep 17, 2014 at 12:55 PM, Chris Marsh <chris.marsh at usask.ca> wrote:

> Because, at this point, I just wanted to get something working, and all
> the example code has Delaunay2D. Figured it was a good start :) Agreed
> though that it is non-optimal in the long run.
>
> Chris Marsh
> PhD Student
> chrismarsh.ca
>
> 13 Kirk Hall
> University of Saskatchewan
>
> On Wed, Sep 17, 2014 at 12:53 PM, David Gobbi <david.gobbi at gmail.com>
> wrote:
>
>> I'm not sure if I've understood, but if your original data already has
>> connectivity (i.e. faces), then why are you using vtkDelaunay2D at all?
>>
>> On Wed, Sep 17, 2014 at 12:49 PM, Chris Marsh <chris.marsh at usask.ca>
>> wrote:
>> > Hi all,
>> >
>> > Thanks for persevering with me. I got it working correctly with
>> > vtkPlaneSource.
>> >
>> > The problem ended up being a PEBKAC with how I was using the Delaunay2D
>> > class: When converting from CGAL to vtkUnstructuredgrid, you need to
>> index
>> > the face values and vertices separately while maintaining connectivity.
>> > Because you are defining the connectivity, this works. However! with
>> > vtkDelaunay2D, you can't do this. Rather, for each vertex that is added
>> to
>> > 'points', the face data must be added 3-times. Which this isn't quite
>> > perfect, it should work fine, unless there is a way of using an
>> unstructured
>> > mesh and a probe filter?
>> > http://www.vtk.org/Wiki/VTK/Examples/Cxx/PolyData/GeometryFilter
>>  suggests
>> > you can? So I will have to play around with that more.
>> >
>> > Thanks again, much appreciated.
>> >
>> > Chris Marsh
>> > PhD Student
>> > chrismarsh.ca
>> >
>> > 13 Kirk Hall
>> > University of Saskatchewan
>> >
>> > On Wed, Sep 17, 2014 at 11:00 AM, David Gobbi <david.gobbi at gmail.com>
>> wrote:
>> >>
>> >> On Wed, Sep 17, 2014 at 10:45 AM, Chris Marsh <chris.marsh at usask.ca>
>> >> wrote:
>> >> > Hi David,
>> >> >
>> >> > Thanks for the reply, this is slowly starting to make sense. That
>> call
>> >> > did
>> >> > in fact fix the bounds issues. However, the output still looks like
>> it
>> >> > did
>> >> > in the above screenshot.
>> >> >
>> >> >>The plane is already composed of cells, so you should not run it
>> >> > through Delaunay.
>> >> >
>> >> > This is what perplexed me about the above linked example.
>> >> >
>> >> > What is the best way to write the probe results to a file then?
>> >>
>> >> At the beginning, I was hinting that you should probe with a
>> >> vtkImageData, which you could have written out as an image
>> >> (either as raw data, or as a png, etc).  But now that you have
>> >> the probed data as a polydata plane, you'll have to get the
>> >> scalars that are associated with the points that make up the
>> >> polydata.  GetPointData()->GetScalars() will return an array
>> >> that contains all the scalars.  See the VTK text for an in-depth
>> >> description of how data is stored in a VTK data set.
>> >>
>> >>  - David
>> >
>> >
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtkusers/attachments/20140917/d7176640/attachment.html>