[postgis-users] parsing postgis answers

[Sandro Santilli]

On Mon, Dec 05, 2011 at 03:41:02PM -0800, Paul Ramsey wrote:
> I'm thinking that we could leave the canonical output the same for
> st_extent while enhancing the output type to incude an srid, so it
> will become possible to do things like
> 
> geom && st_extent(..something..)
> 
> and have it evaluate, which would be a nice improvement, since the
> extent box could then cast correctly back into geometry

Yes, it would be an improvement. Anyway the _canonical_ representation
is what ends up in dumps, so in the (probably deprecated) case of
using our new box in the definition of one of your tables, it will loose
an important bit of information on dump/reload.

--strk; 

  ,------o-. 
  |   __/  |    Thank you for PostGIS-2.0 Topology !
  |  / 2.0 |    http://www.pledgebank.com/postgistopology
  `-o------'