Selon Ted Spradley <tsprad...@snoogems.com>: When OGR creates a PG table, the name of the primary key is 'OGC_FID'. So, you just have to replace 'oid' by 'OGC_FID' in your example.
> > Hi, > > I have used ogr2ogr to load data from shapefile countyp020.shp into > db=project1, following Tyler Mitchell's 'Web Mapping Illustrated'. > > > ogr2ogr -f "PostgreSQL" PG:dbname=project1 countyp020.shp > > Tyler explains that: to optimize for Mapserver we can create an index on the > unique ID for the table. If we use a simple DATA statement in a map file: > DATA wkb_geometry from countyp020, Mapserver will use PostgreSQL's internal > object ID. The oid column doesn't appear in the column listing for a table, > but we can use it in a SELECT statement. > > I am assuming: > > SELECT oid FROM countyp020 LIMIT 1; > > should return one value for the column 'oid'. Instead the return is: > > ERROR: column "oid" does not exist > LINE 1: SELECT oid FROM countyp020 LIMIT 1; > ^ > > So, when preparing the table for performance, the CREATE INDEX statement > fails. > > project1=# CREATE INDEX idx_countyp020 ON countyp020 (oid); > ERROR: column "oid" does not exist > > When the ogr2ogr command above is given, I am assuming that ogr2ogr creates > the SQL CREATE TABLE, and INSERT statements to hold the data from the > shapefile. And PostgreSQL, then does the table creation. Is the 'oid' > column created by default or do we somehow need to instruct ogr2ogr to add > 'oid' as a column? > > Thank you, > Ted S. > > > > -- > View this message in context: > http://n2.nabble.com/ogr2ogr-creation-of-internal-object-id-oid-column-in-pgsql-table-tp3744541p3744541.html > Sent from the GDAL - Dev mailing list archive at Nabble.com. > _______________________________________________ > gdal-dev mailing list > gdal-dev@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/gdal-dev > _______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
