[postgis-users] Possible issue with Geos on PostgreSQL 8.0 NativeWindows

TECHER Jean David davidtecher at yahoo.fr
Wed Jan 12 08:20:20 PST 2005


hi Paul

since it is me that gave to Magnus the command in order to compiling postgis
on windows U will find this command here

http://postgresqlwin32.free.fr/Tutorial_PostGIS/pg800/compil_info.txt

that's right I didn't give him the instruction 'LDFLAGS=-lstdc++' but I
thought as MCA said that it wasn't worth to adding this???
----------------------------------------------------------------
TECHER Jean David
Responsable Informatique 01MAP
e-mail: davidtecher at yahoo.fr
Bureau: 04 67 45 60 27
Portable: 06 85 37 36 75
site perso : http://techer.pascal.free.fr/postgis/
site pro: http://www.01map.com/download/
K-S:"The greatest trick the devil pulled off was convincing people he didn't
exist"
------------------------------------------------------------
----- Original Message -----
From: "Paul Ramsey" <pramsey at refractions.net>
To: "PostGIS Users Discussion" <postgis-users at postgis.refractions.net>
Sent: Wednesday, January 12, 2005 4:51 PM
Subject: Re: [postgis-users] Possible issue with Geos on PostgreSQL 8.0
NativeWindows


> Doh. This sounds like the original problem we had with GEOS, which is
> that the C++ exceptions are interpretted as signals by the PostgreSQL,
> in particular a signal to die, die, die. The problem appeared to be
> quietly "going away", I think as more modern gcc versions actually
> handled the exception more gracefully.  Our original workaround was to
> compile postgresql with libstdc++ linked in at the top of the DLL list
> (even ahead of libc) so that exceptions would be caught by a runtime
> library that knew what to do with them. In the absence of a windows
> native compiler that will do nice things for us, we might have to
> revert to doing our own windows native compilation with the libstdc++
> hack again in order to catch the exceptions.
>
> P
>
> On Wednesday, January 12, 2005, at 06:40 AM, Obe, Regina DND\MMIS wrote:
>
> >
> > I was doing a trial port of my data from Cygwin PostGreSQL 7.4
> > Postgis 0.8
> >
> > To Native Windows PostGreSQL  8.0.0 RC4, postgis 0.9.
> >
> > 2 issues -  these could be caused by something wrong with my
> > reconfiguration
> > since there are so many variables that have changed so not sure if
> > they are
> > problems or not.  I just wanted to know if others have run into similar
> > issues.
> >
> > When I was testing the geom_union function on one of my tables
> >
> > On the cygwin install I get the following behavior
> >
> > SELECT geomunion(the_geom) from neighborhood;
> >
> >
> > results in Geos throwing the below error
> >
> > ERROR:  GEOS union() threw an error!
> >
> > NOTICE:  TopologyException: side location conflict (780343,2.95405e+06)
> > ---------------------------------------------------------
> >
> > select geomunion(the_geom) from neighborhoods where name
> > IN('Dorchester',
> > 'Charlestown');
> >
> > Results in
> > SRID=102686;MULTIPOLYGON(((778278.699002388  .........
> >
> > -----------------------------------------------------------------------
> > -----
> > --------------------
> >
> > Same Tests on the: Native Windows PostGreSQL  8.0.0 RC4, postgis 0.9.
> > (I'm
> > using the PostGIS that came packaged with 8.0.0 RC4 official release)
> > --------------------------------------------------------
> >
> > SELECT geomunion(the_geom) from neighborhood
> >
> > Result: The postgres connection crashes without throwing an error so I
> > need
> > to reconnect.
> >
> > -----------------------------------------------------------------------
> > -----
> > -----
> > select geomunion(the_geom) from neighborhoods where name
> > IN('Dorchester',
> > 'Charlestown');
> >
> > Result - No SRID shows
> > SRID=-1;MULTIPOLYGON(((778278.699002388
> >
> >
> > Other functions such as collect for example return an SRID on both
> > installs
> > and I checked the table and the geometry_columns table and both
> > register an
> > SRID.
> >
> > I tried reloading the table from the original shape file using the
> > newer
> > shp2pgsql and the same error still stands.
> >
> >
> >
> >
> > _______________________________________________
> > postgis-users mailing list
> > postgis-users at postgis.refractions.net
> > http://postgis.refractions.net/mailman/listinfo/postgis-users
> >
>       Paul Ramsey
>       Refractions Research
>       Email: pramsey at refractions.net
>       Phone: (250) 885-0632
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users





More information about the postgis-users mailing list