[postgis-users] Problem with Geomunion after upgrade

[Paul Ramsey]

Well, it's not catastrophic, but it seems pretty obvious that an 
operation on a homogeneous set of geometries should return a result 
with the same srid as the inputs. The same thing should be true of the 
two-argument variants (intersection(), Union(), difference() and 
buffer()).
Paul

On 15-Nov-04, at 6:33 PM, Ethan Alpert wrote:

> So is this a bug? FWIW the work around for me was to enclose 
> geomunion, intersection, etc calls in setSRID() calls.
>   
> -e
>
>
> From: postgis-users-bounces at postgis.refractions.net on behalf of 
> strk at refractions.net
> Sent: Mon 11/15/2004 1:22 PM
> To: PostGIS Users Discussion
> Cc: Paul Ramsey
> Subject: Re: [postgis-users] Problem with Geomunion after upgrade
>
>
> On Mon, Nov 15, 2004 at 12:24:54PM -0700, Ethan Alpert wrote:
> >
> >
> > I upgraded to 0.9.0 (I'll have a post on what a PITA that was soon)
> >
> > Was it always the case that Geomunion returns and SRID=-1 ? I'm 
> doing a
> > geomunion of a table in which all geometries are SRID=4326. How come
> > geomunion now drops the SRID? If this has always been the case then 
> does
> > that mean prior to 0.9.0 constaints weren't being checked?
> >
> > -e
>
> I've checked and it seems that in no version of postgis did
> any overlay operation respected input geoms SRID.
> Does anyone remember anything different from this ?
>
> --strk;
>
>
> > _______________________________________________
> > postgis-users mailing list
> > postgis-users at postgis.refractions.net
> > http://postgis.refractions.net/mailman/listinfo/postgis-users
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users