[Bf-python] Re: Get faces from vertex (Roland Hess)
Tom Musgrove
tommusgrove__ at hotmail.com
Tue Apr 6 10:08:40 CEST 2004
>Now the real question for this whole problem is: what is the best
>information structure for the index?
Did some searching today, it sounds like the primary approaches people are
taking are winged edge, half edge, quad edge, and hashing.
Hashing has the benefit that is probably in the base python install and if I
can't find a module that I can use that implements one of the others then I
might go that route, at least for now. From reading usenet it sounds like
someone has coded an open source half edge module implementation for wings
3d.
I've emailed the author to see what license it is under and where I can find
it.
Anthony D'Agostino scorpius AT what DOT netzero DOT com
On BF-committers in September someone mentioned an interest in coding up
half edge to replace the current data structure used by Blender. Apparently
hashing has been used by other implementers trying to get around similar
limitations. However, no real interest was expressed on the list, and I
could find no further email messages regarding the subject.
See this thread for more info
http://www.blender.org/pipermail/bf-committers/2003-November/004550.html
Also there was a development of an alternative structure with Blender.
However, the author seems to have decided not to contribute it as best as I
can tell.
http://www.blender.org/pipermail/bf-committers/2003-June/003238.html
Here is a quad-edge implementation in C++
http://www-2.cs.cmu.edu/afs/andrew/scs/cs/15-463/pub/src/a2/quadedge.html
They don't say what license the code is under.
K3d uses a split edge B-Rep
http://k3d.sourceforge.net/cgi-bin/wiki?MeshStructure
and I've also found an LGPL library for a winged B-REP in C++
https://sourceforge.net/projects/breplibrary/
Okay,
that is all for now, G'Night
Tom M.
LetterRip
_________________________________________________________________
FREE pop-up blocking with the new MSN Toolbar get it now!
http://toolbar.msn.com/go/onm00200415ave/direct/01/
More information about the Bf-python
mailing list