There is a python implementation of the
tree generation algorithm of Jason Weber
and Joseph Penn now. You can find an early
(but working) alpha release on the sourceforge server:
http://sourceforge.net/projects/arbaro/

It's ad simple command line program eating
XML parameter files and shitting povray files
with lot of cones, spheres and discs.

I added parameter files for some trees and
herbs:
  - aspen
  - tupelo
  - palm
  - shave-grass
  - rush
  and some other

Complex trees like the weeping willow and the
black oak examples of the Weber/Penn paper
need too much memory and time for my
computer. Don't know, if this is a problem of
python or my basic knowledge of it. (I learned it
especially to make this software) May be I have to port 
the software to C++ when the implementation 
is complete and working well.

I didn't implement pruning and other special
features yet. Furthermore I need help to
understand the Weber/Penn paper in some
aspects. It's quite unclear, uncomplete and
misleading for me in several places. Does
someone like to discuss some geometrical 
problems with me?

Thank you for any hints and help
    Wolfram.

-- 
Wolfram Diestel <wol### [at] stelojdeFORIGU>

Post a reply to this message

in news:3f4bcf98@news.povray.org Wolfram Diestel wrote:

> There is a python implementation of the
> tree generation algorithm of Jason Weber
> and Joseph Penn now.

Wolfram,

What version of python are you using? With 2.2.1 I get the following 
error:

D:\Download\arbaro_0_3>arbaro.py <rush.xml >rush.pov
Reading params from stdin...
Traceback (most recent call last):
  File "D:\Download\arbaro_0_3\arbaro.py", line 80, in ?
    p.ParseFile(infile)
xml.parsers.expat.ExpatError: no element found: line 1, column 0


By the way, do you know this one:
http://members.chello.nl/~l.vandenheuvel2/TReal/
It's a program based on the same paper.


Ingo

Post a reply to this message