Just to clarify a bit my position. If we were coming up with a new api I would be for having false (non-reliable) fids as the default. But it is my understanding that most client code operates under the assumption that datastores report reliable fids, and those that don't are sort of "broken". So flipping that assumption seems a serious break to backward compatibility.
Also as written we will need to update all datastores to return true. when they return a repeatable fid, which is the majority of them afaik. The patch only deals with shapefile and the old postgis. Returning true by default would mean we only need to update the cases where this is not so, as far as i know this amounts to database tables with no key set. -Justin David Winslow wrote: > Hi all, > > I put together a patch for the capabilities improvement described in > http://jira.codehaus.org/browse/GEOT-2257 . It's attached to the > issue. One problem that jdeolive brought up during a quick review was > the question of what the default implementation should return. Justin > advocated having it return true, since consistency is part of what > feature ids are supposed to *mean*; whereas I thought it would be better > to have it return false since a false positive is more likely to result > in undefined behavior than a false negative. > > The other point that Justin brought up was one of style, I named the new > method 'hasReliableFIDs()' instead of the more consistent > 'isReliableFidSupported()'. I'm currently making this change. > > Does anyone else care to weigh in? > > David Winslow > > > ------------------------------------------------------------------------------ > _______________________________________________ > Geotools-devel mailing list > Geotools-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/geotools-devel -- Justin Deoliveira OpenGeo - http://opengeo.org Enterprise support for open source geospatial. ------------------------------------------------------------------------------ _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
