Justin Deoliveira wrote: > I prefer the first. Taking apart the structure of a GmlObjectId is a > hack. However... I don't see isIdentifiedApplicable() as being much > better since the datastore sill probably not know how to take apart teh > identifier either. You see the problem however; it speaks to the getGmlObject method being a DataStore concern rather than a FeatureStore concern (just like transactions). I will also note that you could use aggregation here; .... DataStore.getGmlObjectStore() returns non null if and only iff the operation is supported. You may find this approach easier to hack given the cascade of wrappers Andrea mentioned. > Most certainly it will just turn the thing to a > string and do a look up. And since it does not throw IOException it > means it cannot even do that. I dont see much use for it. Just returning > null from getGmlObject is just as meaningul. > I am thinking of the FeatureID case where "Road.123455" format is very common. For many (most?) implementations they will be using a similar trick to figure out what identifier goes with which feature source. > Object getGmlObject( GmlObjectId id, Hints hints ) throws IOException; > Sounds good; lets update the proposal page. Jody
------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Geotools-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geotools-devel
