So, the FGDB driver right now flattens out the GDB hierarchy a little (as far as I can tell there is not *too* much to flatted out) which could maybe (?) result in duplicate layer names... I opened up Michaels test database and printed out the layer names,
basetable_2 basetable base_table1 test_lines more_test_lines test_points test_areas Some of those are Tables and the others are Feature Classes stored within a Feature Dataset. I'm guessing that there is a unique name enforcement mechanism within particular levels of the FGDB, but not sure what would happen if we had something like this: Tables: george, ringo Feature Classes: john Feature Datasets: paul (Feature Classes: john, ringo) So the fully qualified names would all be unique... \george \ringo \john \paul\john \paul\ringo but the layer names would not. A couple things could happen: we could mangle the fully-qualified names so they remain fully qualified in their OGR form (paul_john, paul_ringo) or we could test and see if it's even possible to create an FGDB that has non-unique leaf node names. Paul _______________________________________________ gdal-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/gdal-dev
