Anke,

Yes, the PrecisionModel is integrated into some geometrical operations

of JTS. As I said in an earlier post, the PM is used in all operations

which compute a new geometry. The operations ensure that new geometry

meets the PM which is supplied with the input geometries (or

independently, in some cases). This is useful, since sometimes JTS is

used to process geometry which comes from systems which have a more

restricted PM than double-precision. By using an appropriate PM, JTS

will compute new geometry which can be stored back into those systems

with no problems of topology collapse.

JTS currently does not support using a tolerance for the spatial

predicates. This could be a useful extension - I'd certainly be

interested in exploring this possibility, especially if there was a

funded project to support it.

Anke Trittenbach wrote:

> Is the PrecisionModell integrated in the geometrical operations of JTS? If this is the case, for what is the PrecisionModell used in these operations? For which use is the PrecisionModell in general exactly?

>

> Many have also suggested that I could work with a tolerance. Thereby the renewed queries could be answered with true whether an educated intersection lies on one of the lines. Why does JTS not use such a tolerance?

>

>

--

Martin Davis

Senior Technical Architect

Refractions Research, Inc.

(250) 383-3022

_______________________________________________

jts-devel mailing list

[hidden email]
http://lists.refractions.net/mailman/listinfo/jts-devel