index on id

[frodrigo]

Any plan to support an index on id ?

[vpikulik]

Sorry for the late answer. I think no. Best way to use this extension is to read all data from osm file to MATERIALIZED VIEW and create needed indexes on this view.

CREATE MATERIALIZED VIEW osm_data AS SELECT * FROM osm_foreign_table WITH DATA;

[frodrigo]

Ok. My point of view is, using this extension like is, is just an other way to import data into postgres. I think it would be interesting to have at left index and split osm data type in tables (node, way, relation). So we can benefit of pbf size with no import (size and process disadvantage) for read only on osm data.

(my live I working on Osmose-Qa tools and we massively import osm data for read only and drop database for 200 sub areas of world in loop, every day)

[vpikulik]

Thank for the your answer. You right that this extension was created as other way to import data to postgres. Your idea sounds interesting for me. But currently I have no much time, no idea how to navigate in osm pbf file fast. I will try to think how to do this, but no promises. I will try to think how to do this, but no promises.