osmpath
Shortest paths using OpenStreetMap. In-process; efficient for batch queries. Ideal for city-scale data analysis.
>>> from osmpath import OSMPathPlanner
>>> def is_bikeable(way):
... nogo = {'motorway', 'motorway_link', 'footway', 'service'}
... return ("highway" in way.tags) and (way.tags["highway"] not in nogo)
>>> osm = OSMPathPlanner.from_osm("path/to/sf.osm", way_filter=is_bikeable, verbose=True)
>>> # specify more than one origin/destination pair
>>> orig = [[-122.45893333, 37.80118333],
... [-122.41249 , 37.79199167],
... [-122.41979667, 37.77634 ]]
>>> dest = [[-122.41249667, 37.791975 ],
... [-122.41979667, 37.77634 ],
... [-122.41767833, 37.77632 ]]
>>> paths = list( osm.get_shortest_paths( orig, dest ) )
>>> len(paths)
3
>>> paths[0] # each edge in a path is an OSM way_id and an inclusive range of node indices.
[[219171857, [0, 6]],
[219171857, [6, 7]],
[529868660, [0, 1]],
...
>>> osm.get_path_geom( paths[0] )
array([[-122.459098 , 37.8018337],
[-122.459182 , 37.8018022],
[-122.4592759, 37.801769 ],
[-122.4593613, 37.8017409],
...
Authors
osmpath was written by Brandon Martin-Anderson <badhill@gmail.com>_.