[postgis-devel] isvalid(geom), but st_ashexewkb(geom, 'NDR')::geometry fails

[David Fuhry]

That makes sense; I guess I've updated from an older postgis on this system.

I do wish that isvalid() would return false for this geometry. 
Intuitively, any geometry which cannot be parsed, should not be valid.

db=> select postgis_full_version();
                                postgis_full_version 

----------------------------------------------------------------------------------
  POSTGIS="1.3.3" GEOS="3.0.0-CAPI-1.4.1" PROJ="Rel. 4.6.0, 21 Dec 2007" 
USE_STATS
(1 row)

Thanks,

Dave

Kevin Neufeld wrote:
> I was going to say that it looks like a bug in the WKB parser, but then 
> I noticed that this is fixed in newer PostGIS versions.
> 
> In my 1.1.6 version, I can produce the same thing - it's clearly not 
> symmetric.
> 
> postgis=# select 
> astext('0106000020C90E00000100000001030000000100000000000000'::geometry);
>          astext
> ------------------------
>  MULTIPOLYGON(( EMPTY))
> (1 row)
> 
> postgis=# select astext('MULTIPOLYGON(( EMPTY))');
> ERROR:  parse error - invalid geometry
> 
> -- Kevin

 > David,
 >
 > Well it is all kind of bizarre.  Which build exactly are you using.  Are
 > you using an old 1.3.4 trunk version?
 >
 > If I do any of the following
 >
 > SELECT
 > ST_AsText('0106000020C90E00000100000001030000000100000000000000');
 > SELECT '0106000020C90E00000100000001030000000100000000000000'::geometry;
 > SELECT
 > ST_IsValid('0106000020C90E00000100000001030000000100000000000000');
 >
 > I get the below error on my 1.3.3 or 1.3.4SVN (I have to check my
 > vintage)
 > ERROR:  geometry requires more points
 >
 > ST_IsValid doesn't return anything for me because apparently its so
 > bogus that it errors out before it can even attempt to run validity
 > checks.
 >
 > What does your SELECT postgis_full_version(); return
 >
 > There is not enough info given by postgis_version(); for me to tell what
 > you are running.
 >
 > Thanks,
 > Regina