[postgis] 'real' contains, touches, overlaps etc.

[Manuel Weindorf]

Hi Martin,

thanks for the hints. As I looked at the details I have to agree completely 
with you. I'm looking forward to see what you've done in JTS and how it can 
be ported to C/C++.

Manuel

> You're right, Manuel, conformance to the OGC spec requires relationship
> operators that use the DE-9IM model (that is, that understand the
> difference between exterior, interior and boundary).    There's also some
> other subtle and not-so-subtle issues that need to be addressed in a
> production-quality spatial library:
>
> *	Using a finite-precision numerical representation (eg floating point
> or integers) can cause a naive implementation to fail badly on certain
> kinds of input.  This can be avoided by using robust geometric algorithms.
> *	Some kind of spatial indexing or other efficient spatial data
> structure need to be implemented if the code is to be usable on inputs with
> anything more than a small number of points.
> *	The OGC spec has some subtle with how the definition of the spatial
> data types interacts with the relationship operators.  A good
> implementation should specify how it handles these issues.
>
> JTS is designed to address these issues.  If you're interested in reading
> the design specification, which elaborates on these issues and other
> aspects of JTS, you can find it at 
> http://www.vividsolutions.com/jts/jtshome.htm


-- 
Manuel Weindorf (weindorf at ipf.uni-karlsruhe.de)
Institut fuer Photogrammetrie und Fernerkundung, Universitaet Karlsruhe
Postfach 6980 D-76128 Karlsruhe, Tel. +49 721 608-6010 Fax +49 721 608-8450
http://www-ipf.bau-verm.uni-karlsruhe.de

------------------------ Yahoo! Groups Sponsor ---------------------~-->
Secure your servers with 128-bit SSL encryption! Grab your copy of VeriSign's FREE Guide: "Securing Your Web Site for Business." Get it Now!
http://us.click.yahoo.com/n7RbFC/zhwCAA/yigFAA/PhFolB/TM
---------------------------------------------------------------------~->

To unsubscribe from this group, send an email to:
postgis-unsubscribe at yahoogroups.com

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/