[postgis-devel] dumppoints.sql regression test relies on special configuration

[strk]

Alright, this seems related to the change
in ST_NumGeometries, as the _ST_DumpPoints plpgsql function
makes use of that.

In particular, it relies on a NULL return from NumGeometries
to figure if a geometry is a collection or not:

  -- Special case (MULTI* OR GEOMETRYCOLLECTION) : iterate and return the DumpPoints of the geometries
  SELECT ST_NumGeometries(the_geom) INTO nb_geom;

  IF (nb_geom IS NOT NULL) THEN

Doh...

--strk;

On Fri, Mar 12, 2010 at 03:58:04PM +0100, strk wrote:
> On Fri, Mar 12, 2010 at 02:54:42PM +0100, strk wrote:
> > Errors on make check:
> > 
> > ! ERROR:  stack depth limit exceeded
> > ! HINT:  Increase the configuration parameter "max_stack_depth", after ensuring the platform's stack
> > depth limit is adequate.
> > 
> > This is running the dumppoints.sql
> 
> Tried raising max_stack_depth from the default of 2MB up to 7MB
> with same results.
> 
> Setting to 8MB resulted in a silent startup failure (pg_ctl
> start doesn't complain but bakcend doesn't come up).
> Can anyone confirm such behaviour ?
> 
> Environment informations:
> 
>  PostgreSQL 8.3.8 on i686-pc-linux-gnu, compiled by GCC gcc (SUSE Linux) 4.3.2 [gcc-4_3-branch revision 141291]
>  Postgis 2.0.0SVN - 2010-03-12 13:28:32
>    GEOS: 3.3.0-CAPI-1.6.1
>    PROJ: Rel. 4.7.1, 23 September 2009
> 
> CPU is about still at 100% during run.
> 
> --strk; 
> 
>   ()   Free GIS & Flash consultant/developer
>   /\   http://strk.keybit.net/services.html
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel

-- 

  ()   Free GIS & Flash consultant/developer
  /\   http://strk.keybit.net/services.html