2

I really like the functions of OSMnx, thank you! Nevertheless sometimes OSM-data is limited, especially for human powered modes like walking and cycling. I am working on a project for Zuerich, Switzerland and I am lucky that the city has a full walking network, including sidewalks, steps, paths, etc. At the moment this file is downloable as ESRI shapefile, gpkg or GEOjson. It just contains the lines but it would be easy to add nodes to them because edges are drawn until the next intersection point with another edge. For the 3D-layer it is accounted for, in such a way that lines on a different height aren't intersecting. I was wondering if it is possible to import a shape-file and work with it in OSMnx?

  • Did anyone did this before?
  • What would be a first step?
  • Does it make sense to try to convert the shapefile to "osm xml" in order to import it like this?

I have tried to first import it in networkx (https://gis.stackexchange.com/questions/227909/from-shapefile-to-python-networkx) but the syntax for nodes and edges is so different. Thank you, I appreciate every hint.

The data can be found here: https://data.stadt-zuerich.ch/dataset/geo_fuss__und_velowegnetz/resource/80af45c0-835c-4e29-b99d-f8cd6dfefa11?inner_span=True (includes also bike-infrastructure which has to be removed)

jonas_2021
  • 23
  • 3
  • Yes, assuming your data have the necessary structure, using the graph_from_gdfs function https://osmnx.readthedocs.io/en/stable/osmnx.html#osmnx.utils_graph.graph_from_gdfs – gboeing Mar 15 '21 at 15:34
  • I found a good workaround: Just load a GEOjson (can be created easily from a shape) in josm (https://josm.openstreetmap.de/wiki/Download) and export it to an OSM file in this program. Worked for my purpose easily. – jonas_2021 Mar 17 '21 at 08:25

0 Answers0