I agree with David on this...but....I also agree with Fabio. As long as spatial is not ubiquitous (ie: only supported by a few databases) then it should remain outside of core. However, once *most* databases have spatial support then data-types like geometry and geography become first-class citizens like varchar and int. At that point in time I believe that NH Spatial should move into core. The question is, how do we decide when that time is?
On Apr 25, 5:59 am, David Pfeffer <[email protected]> wrote: > Thanks for the link. That project unfortunately is a LINQ to Objects > implementation of GIS querying, however. What's really lacking is support > for querying said spatial data from a database, such as in a where clause. > These queries rely on the database's native spatial indexes, and are > substantially faster than in-memory manipulation of geospatial data. > > > > > > On Sat, Apr 24, 2010 at 1:53 PM, Fabio Maulo <[email protected]> wrote: > > On Sat, Apr 24, 2010 at 2:37 PM, David Pfeffer <[email protected]>wrote: > > >> IMO, the NHibernate core should support everything that the database > >> engine supports, right out of the box. If the consensus is that this isn't > >> a > >> good policy, then I'll go about extending the core with extensibility > >> points > >> for Linq. (I don't think Spatial is going to go anywhere without full Linq > >> support.) > > >http://linqtogeo.codeplex.com/ > > > -- > > Fabio Maulo > > -- > Subscription > settings:http://groups.google.com/group/nhibernate-development/subscribe?hl=en
