> If we follow the approach above, the coordiantesequence dimension
> is available only as lons as we have data. What about metadata?
> We'd need some information in the GeometryDescriptor I believe.

The CoordinateReferenceSystem is already at hands length from the raw
geometry; and is recorded at the GeometryDescriptor level; so having
some extra information here; getDimension() would be consistent. I
have already given up "understanding" what a raw geometry means
without an GeometryDescriptor; I just liked having on bit of
descriptive data rather then two...

> Granted, CRS is available in both places, but I'm not seeing
> any sane path to add this extra, override information into it,
> without breaking its API and requiring quite some changes in
> the referencing subsystem at the same time. Too bad CRS
> does not have a mutable user data map...

Well we could add that; or a look aside buffer - oh wait their are all
kinds of maps to make sure CRS instances are not duplicated (the same
instances is returned from multiple code paths etc...) so a user map
is not going to save us ....

> Any bright idea here? (I know I already said no to this, yet I said no to the 
> only way I can imagine this implemented. If you have any smarter path, please 
> share).

I will think about it; but currently no bright ideas.
Dim idea: use a datastore supplied wrapper around
CoordinateReferenceSystem to communicate with you need to system code;
and have the referencing module "pop" the origional out of the wrapper
when it starts doing math transform fun

Jody

------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations 
Conference from O'Reilly Media. Velocity features a full day of 
expert-led, hands-on workshops and two days of sessions from industry 
leaders in dedicated Performance & Operations tracks. Use code vel09scf 
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to