Data export of all German streets (including postal codes, localities and regional keys) from the OpenStreetMap project.
This dataset is an extract from the regional OpenStreetMap file for Germany germany-latest.osm.pbf. The following algorithm is used:
-
Read in all postal code areas (OSM relations), municipality areas (OSM relations) and streets (OSM ways) together with their geometric information.
-
Include only ways which have one of the the following properties:
Name Value place
square
leisure
park
highway
primary
highway
secondary
highway
tertiary
highway
residential
highway
living_street
highway
road
highway
unclassified
highway
footway
highway
pedestrian
highway
track
(but only fortracktype=grade1
)highway
service
(but only forservice=alley
) -
For each OSM Street, determine the geometric centre and in which municipality and postcode area it is located.
-
Create a new street object (name, postal code, locality, regional key) from the combination of the three data objects (OSM street, OSM municipality area, OSM postal code area).
-
Shorten all street names with
strasse
orStrasse
tostr.
orStr.
. -
Ignore all streets with names matching the following regular expression:
^(\\?|\\+|-|_|\\(.*\\))$
. -
Export all street objects to CSV.
The data in the OpenStreetMap project is not perfect, but surprisingly well maintained for Germany.
-
A total of 120 municipality regional keys from the official GV100AD (Gemeindeleitdatei) are not represented. This is largely due to the fact that these are areas without buildings and proper roads.
-
Streets and their postal codes or their municipality keys are determined on the basis of geometrical comparisons. This procedure is not 100% perfect.
-
Street names, postal codes and municipality keys are subject to constant change. The OpenStreetMap project as a community project tries as best as possible to update its data regularly. Current changes in streets and postal codes are documented quarterly by Deutsche Post (see Deutsche Post Direkt), current changes in municipality assignment in the Gemeindeleitdatei of the Federal Statistical Office (Destatis).
Yes, that would be much appreciated. The best way to help is to post a response via the Issue Tracker and/or submit a Pull Request.