[Qgis-developer] adding graduated symbol classification algorithms

[Carson Farmer]

Hi Devs, attached is a diff that adds 3 new modes to the graduated
symbol renderer, including Jenks' Natural Breaks Optimisation
algorithm, a standard deviation classification algorithm, and the
'pretty' algorithm from the R statistical programming language. An
explanation of the algorithms is provided as comments in-code, and you
can also have a look here: http://www.carsonfarmer.com/?p=761 for the
initial Python version, along with a brief overview of the algorithms
etc. I'd really appreciate any comments, suggestions, or fixes, as I'm
not really a C++ programmer. I'm sure there are a few spots where
things could be sped up, especially in the Jenks algorithm. Right now
I'm using a random sample to speed things up, but I'm sure others more
familiar with C++ could find other speed improvements. I don't really
want to commit this directly, again mainly because I'm not really
confident in my C++ abilities, but if it would make it easier, I can
also add the patch as an enhancement ticket?

Thanks,

Carson

-- 
Carson J. Q. Farmer
ISSP Doctoral Fellow
National Centre for Geocomputation
National University of Ireland, Maynooth,
http://www.carsonfarmer.com/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: jenks_class.diff
Type: application/octet-stream
Size: 13852 bytes
Desc: not available
Url : http://lists.osgeo.org/pipermail/qgis-developer/attachments/20100924/ed39aa44/jenks_class-0001.obj