[postgis-devel] Rant on SRID -1 vs 0

[maplabs at light42.com]

I was around during this discussion.. 
(most respect to SWoodbridge and some empathy for the frustration.. but..) 

2.0 was clearly a major change, with the binary format of geometry 
first and foremost.. 
If there was any time to make a somewhat  arbitrary change like the 
"unassigned" value, 
2.0 was arguably the best time to do it.. 

Personally, the zero value makes more sense to me, and I think I said 
so at that time also
$0.02

--
Brian M Hamlin
OSGeo California Chapter

On Mon, 15 Jul 2013 17:10:51 -0400, Stephen Woodbridge 
<woodbri at swoodbridge.com> wrote:
I need to vent some frustration with the the change from srid=-1 to
> srid=0 for the undefined projection because this totally breaks 
> scripts and makes it very hard to test things like pgrouting that 
> need to work on both 1.5, 2.0 and 2.1. 
>
> For example:
>
> select addgeometrycolumn('unnoded', 'the_geom', 0, 'LINESTRING', 2);
>
> does not work on  POSTGIS="1.5.3" GEOS="3.3.8-CAPI-1.7.8" PROJ="Rel. 
> 4.8.0, 6 March 2012" LIBXML="2.7.8" USE_STATS
>
> And this is just the first of the problems I'm trying to track down. 
>
> The second is the change in behavior of st_startpoint() with multi 
> geometries and the change of st_geometryn() with simple geometry. 
> This forces every function that uses these to check the postgis 
> version conditionally implement access. 
>
>
> Grummble, gummble, gummble, off to find some kind of solution(s). 
>
> -Steve
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
>
>