octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Latest port of Qhull/goemetry functions from octave-forge


From: John W. Eaton
Subject: Latest port of Qhull/goemetry functions from octave-forge
Date: Thu, 23 Aug 2007 13:38:53 -0400

On 22-Aug-2007, David Bateman wrote:

| Find attached my latest port of the geometry functions based on Qhull
| from octave-forge to Octave. The things I changed while porting this
| code are
| 
| * Style changes to more closely meet the Octave coding style
| * Some texinfo help string fixes
| * Changes in the configure process to use the existing Octave mechanisms
| to pass configuration information
| * Some attempt to use octave_idx_type rather than int where appropriate,
| though as Qhull uses int, I'm unsure that this will mean that the code
| will be 64 bit safe.
| * move delaunayn.cc to __delaunay__.cc and wrap in delaunayn.m. In the
| new delaunayn function remove zero volume simplices for the tesselation.
| * Inclusion of the triplot and trimesh functions
| * Addition of naive versions of dsearch, dsearchn, tsearchn though they
| appear to perform quite well
| * Addition of griddata3 and griddatan functions
| * Add test code and check for compatibility with Matlab and fix code
| where needed.
| * Initial documentation
| 
| Things that still need to be done are
| 
| * Write more complete documentation
| * Adapt voronoi.m and voronoin.m so that it returns the diagram with the
| segments of the diagram with one end point at infinity.
| 
| However, given the above provisos the code is pretty complete and I've
| tried to fully test it. I'd therefore suggest that this patch is
| committed and the fixes above are applied later.

I agree.  Please check in these changes.

Thanks,

jwe


reply via email to

[Prev in Thread] Current Thread [Next in Thread]