I am trying to process a .osm file in python with ogr, but I have problems accessing the lat and lon attributes of the point layer - they are always null.
I want to do spatial filtering to find the closest neighbor in a certain distance, e.g. like:
sql_lyr = osmData.ExecuteSQL("SELECT * FROM points WHERE lat BETWEEN ... AND ... AND long BETWEEN ... AND ...")
(Inspired by the answer to this question: Python: how to speed up spatial search (nearest point)? )
When I call DumpReadable on a feature in the point layer, I get the following:
OGRFeature(points):392633 osm_id (String) = 392633 name (String) = (null) ... lon (String) = (null) lat (String) = (null) all_tags (String) = (null) POINT (16.3540923 48.2041024)
So I see that the lat and lon are read in for the POINT Geometry, but I have no idea how to use this information in a SQL query.
Is there any way I can access the lat and lon via the POINT Geometry in a query, or can I do any further configuration to the OSMdriver so that lat and lon are read in as fields?
I modified osmconf.ini to enable all_tags, to report all nodes, I tried enabling all common attributes and I also tried adding lat and long to the list of keys like this (even though I think this is pointless, because lat long are attributes of the node tag itself and not really keys...):
# keys to report as OGR fields attributes=name,barrier,highway,ref,address,is_in,place,man_made,lon,lat