As osm2po was a more reliable way to convert an .osm I went for that solution to load my date in postgres.
In the workshop however the author points out to check the result after osm2pgrouting. The following tables must be created:
"classes"
"geometry_columns"
"nodes"
"spatial_ref_sys"
"types"
"vertices_tmp"
"vertices_tmp_id_seq"
"ways"
After the sql (generated from osm2po) has run, only the tables
"geometry_columns" and "spatial_ref_sys" are created.
How do I fix this difference?
Is osm2po incomplete?
Did something go wrong?
Can anyone point me in a better direction?
Thanks alot!
Best Answer
geometry_columnsandspatial_ref_sysare created and populated by PostGIS automatically in case you add a geometry column. The other tables are more or less temporary, used for finding links via assign_vertex_id. osm2po does all this segmenting-stuff for you. All you need is the resulting table which is created by osm2po. In addition osm2po does not rely on assign_vertex_id and uses the original links from osm instead.