Have not managed to get Rini on the email list - she did add the following comment:
> Thanks Jody. > I think, the way the encoder works, it parses the schema, and checks for each > property and attribute in the object. If it's there, it gets encoded. > As for geometries, AbstractGeometryTypeBinding.getProperty() gets the CRS > from the userData if the property name is "srsName". > The feature/attribute type is not available from that point, but maybe it's > because I don't have enough knowledge.. > Anyway I'll take a look again on Monday and will post a discussion on the > mailing list, if I can't find a smarter way to do this. Rini I don't think it is a lack of knowlege; I think it is a lack of the encoder taking advantage of the feature model (one of the thing your crew is pushing out into the geotools library). Jody On 16/07/2010, at 1:54 PM, Jody Garnett wrote: > Reni an outstanding patch for GEOT-2707 - and while the patch is clear > I would like to take discussion to the devel list as it touches on a > few cross module questions. > > Reni's patch correctly updates a CoordinateReferenceSystem object > placed in a JTS Geometry userData. > > In the feature module we have GeometryAttribute which holds both a > Geometry and provides access to the CRS. It is an implementation > detail if this is actually stored in the geometry user data; or if it > consults the feature type an reports back the correct value. > > So my question is this - why is Reni's patch required? What code is > checking the geometry user data - and can we update it to check the > GeometryAttribute instead. > > Reni your patch may very well still be needed; but I would like to > nail down an API contract for the handling of CRS. The convention of > stroing the CRS in the Geometry userData is just that a convention. > > Jody ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
