On 7 November 2011 15:07, Stefan Schwarzer <[email protected]> wrote: > Hi there, > > I am somewhat fighting with the projection of my file. > > I have a file for Morocco, which, after ArcMap, has the following projection: > > Projected Coordinate System: zone1 > Projection: Lambert_Conformal_Conic > False_Easting: 500000.00000000 > False_Northing: 300000.00000000 > Central_Meridian: -5.40000000 > Standard_Parallel_1: 34.86600000 > Standard_Parallel_2: 31.72500000 > Scale_Factor: 1.00000000 > Latitude_Of_Origin: 33.30000000 > Linear Unit: Meter > Geographic Coordinate System: GCS_Merchich_Degree > Datum: D_Merchich > > After looking around, it seems to be the EPSG 26191 > (http://spatialreference.org/ref/epsg/26191/). > > Now, trying to import it with: > > shp2pgsql -s EPSG:26191 -I -D morocco.shp morocco > morocco.sql > psql -U xxxx my_database < morocco.sql > > > I can see in the SQL file that the SRID is set to -1. If I change this to > 26191, and try to import it then, it gives me an error message: > > ERROR: new row for relation "morocco" violates check constraint > "enforce_srid_the_geom" >
Hi, Can you explicitly drop the table before re-running the SQL ? > > So, I changed the SRID in the geometry_columns manually. > > But when trying to map the layer, I receive the error messsage: > > ERROR: Operation on two geometries with different SRIDs > > It is not enough to update the geometry_columns. You also have to update internal geometries SRID: update mytable set geom = set_srid(geom). (looking at the manual, I also see: updateGeometrySrid, that manages all operations (metadata, constraints and data, did not know this one :) ) > My Mapserver.map file show only SQLs with 26191, for projection of main > document, as well as SQLs such as: > > DATA 'the_geom FROM morocco USING UNIQUE gid USING srid=26191' > > > Extent and units of the mapserver file have been changed to the proper > extent/units. > > So, I am confused about where the problem comes from. It would be great if > someone could give me a hint where the error lies. > > Thanks a lot! > Nicolas _______________________________________________ postgis-users mailing list [email protected] http://postgis.refractions.net/mailman/listinfo/postgis-users
