[postgis-users] geos result sets

[Martin Davis]

Actually, the precision model does not enter into the spatial
relationship computation in JTS/GEOS.  Spatial Relationships are always
calculated based on the precise value of the coordinates, whether that
value has metre precision or nanometre precison.  

The precision model is mainly used by constructive functions, to
determine the allowable precision to use in computing new coordinates.

JTS & JUMP provide a Precision Reduction function, which might handle
some situations like this.  However, the precision reduction is not
guaranteed to maintain topological correctness, and thus will not be
useful in many situations.  (It's a sledgehammer, where often a scalpel
is required).

Oracle Spatial does provide a notion of tolerance in its spatial
relationship operators.  However, I have never seen a description of the
actual semantics of the tolerance value, so I don't feel that I fully
understand what it does or how it could be implemented.  It would be an
interesting exercise to try and define a tolerance-based semantics for
spatial relationships. (If anyone is aware of such a thing please let me
know).  If this proved useful it would of course be nice to add it to
JTS/GEOS.

Martin Davis, Senior Technical Architect
Vivid Solutions Inc.
Suite #1A-2328 Government Street Victoria, B.C. V8T 5G5
Phone: (250) 385 6040 - Local 308 Fax: (250) 385 6046


> -----Original Message-----
> From: Paul Ramsey [mailto:pramsey at refractions.net] 
> Sent: Wednesday, February 25, 2004 3:50 PM
> To: PostGIS Users Discussion
> Subject: Re: [postgis-users] geos result sets
> 
> 
> What would be the result if we were using JTS/GEOS with a precision 
> model more coarse than the differences between the vertices? 
> IE, a 0.1 
> meter precision?
> 
> Paul Ramsey wrote:
> 
> > Thanks for the checks Martin. I think the standard that 
> ArcView works 
> > to
> > is "good enough for government work", or "plus or minus 
> something or 
> > other".
> > 
> > JTS/GEOS is so good it starts to beg the question -- how 
> good is *too*
> > good? :)
> > 
> > Martin Davis wrote:
> > 
> >> I thought I better leap to the defence of JTS/GEOS.  I've analyzed 
> >> the situation and (no surprise to me) found that the results from 
> >> GEOS are correct.  The Within=FALSE result is caused by a slight 
> >> difference in the vertices in the two coverages.
> > 
> > 
> > 
> 
> 
> -- 
>        __
>       /
>       | Paul Ramsey
>       | Refractions Research
>       | Email: pramsey at refractions.net
>       | Phone: (250) 885-0632
>       \_
> 
> _______________________________________________
> postgis-users mailing list postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users
>