Skip to main content

Async client for the Overpass API

Project description

A client for the Overpass API, a read-only API that serves up custom selected parts of OpenStreetMap data.

The Overpass API is optimized for data consumers that need a few elements within a glimpse or up to roughly 10 million elements in some minutes, both selected by search criteria like location, type of objects, tag properties, proximity, or combinations of them. To make use of it, you should familiarize yourself with Overpass QL, the query language used to select the elements that you want.

Contents

See also


Features

  • Asynchronous requests using aiohttp
  • Parallel queries within rate limits
  • Fault tolerance through a (customizable) retry strategy
  • Extras
    • Typed elements that simplify browsing result sets
    • Shapely geometries for manipulation and analysis
    • GeoJSON exports
    • Simplified querying and processing of public transportation routes

Design Goals

  • A small and stable set of core functionality.
  • Good defaults for queries and retrying.
  • Room for extensions that simplify querying and/or processing of spatial data in specific problem domains.
  • Sensible and spec-compliant GeoJSON exports for all objects that represent spatial features.
  • Detailed documentation that supplements learning about OSM and the Overpass API.

Usage

There are three basic steps to fetch the spatial data you need:

  1. Formulate a query

    • Either write your own custom query, f.e. Query("node(5369192667); out;"),
    • or use one of the Query subclasses, f.e. SingleRouteQuery(relation_id=1643324).
  2. Call the Overpass API

    • Prepare your client with client = Client(user_agent=...).
    • Use await client.run_query(query) to fetch the result set.
  3. Collect results

    • Either access the raw result dictionaries with query.result_set,
    • or use a collector, f.e. collect_elements(query) to get a list of typed Elements.
    • Collectors are often specific to queries - collect_routes requires a RouteQuery, for instance.

Example: looking up a building in Hamburg

a) Results as Dictionaries

You may use the .result_set property to get a list of all query results without any extra processing:

from aio_overpass import Client, Query

query = Query('way["addr:housename"=Elbphilharmonie]; out geom;')

client = Client()

await client.run_query(query)

query.result_set
[
      {
          "type": "way",
          "id": 24981342,
          # ...
          "tags": {
              "addr:city": "Hamburg",
              "addr:country": "DE",
              "addr:housename": "Elbphilharmonie",
              # ...
          },
      }
]

b) Results as Objects

This will give you a user-friendly Python interface for nodes, ways, and relations. Here we use the .tags property:

from aio_overpass.element import collect_elements

elems = collect_elements(query)

elems[0].tags
{
    "addr:city": "Hamburg",
    "addr:country": "DE",
    "addr:housename": "Elbphilharmonie",
    # ...
}

c) Results as GeoJSON

The processed elements can also easily be converted to GeoJSON:

import json

json.dumps(elems[0].geojson, indent=4)
{
    "type": "Feature",
    "geometry": {
        "type": "Polygon",
        "coordinates": [
            [
                [
                    9.9832434,
                    53.5415472
                ],
                ...
            ]
        ]
    },
    "properties": {
        "id": 24981342,
        "type": "way",
        "tags": {
            "addr:city": "Hamburg",
            "addr:country": "DE",
            "addr:housename": "Elbphilharmonie",
            ...
        },
        ...
    },
    "bbox": [
        9.9832434,
        53.540877,
        9.9849674
        53.5416212,
    ]
}

Choosing Extras

This library can be installed with a number of optional extras.

  • Install no extras, if you're fine with dict result sets.

  • Install the shapely extra, if you would like the convenience of typed OSM elements. It is also useful if you are interested in elements' geometries, and either already use Shapely, or want a simple way to export GeoJSON.

    • This includes the pt module to make it easier to interact with public transportation routes. Something seemingly trivial like listing the stops of a route can have unexpected pitfalls, since stops can have multiple route members, and may have a range of different tags and roles. This submodule will clean up the relation data for you.
  • Install the networkx extra to enable the pt_ordered module, if you want a route's path as a simple line from A to B. It is hard to do this consistently, mainly because ways are not always ordered, and stop positions might be missing. You can benefit from this submodule if you wish to

    • render a route's path between any two stops
    • measure the route's travelled distance between any two stops
    • validate the order of ways in the relation
    • check if the route relation has gaps
  • Install the joblib extra to speed up pt_ordered.collect_ordered_routes(), which can benefit greatly from parallelization.


Coordinates

  • Geographic point locations are expressed by latitude (lat) and longitude (lon) coordinates.
    • Latitude is given as an angle that ranges from –90° at the south pole to 90° at the north pole, with 0° at the Equator.
    • Longitude is given as an angle ranging from 0° at the Prime Meridian (the line that divides the globe into Eastern and Western hemispheres), to +180° eastward and −180° westward.
    • lat/lon values are floats that are exactly those degrees, just without the ° sign.
  • This might help you remember which coordinate is which:
    • If you think of a world map, usually it’s a rectangle.
    • The long side (the largest side) is the longitude.
    • Longitude is the x-axis, and latitude is the y-axis.
  • Be wary of coordinate order:
    • The Overpass API explicitly names the coordinates: { "lat": 50.2726005, "lon": 10.9521885 }
    • Shapely geometries returned by this library use lat/lon order, which is the order stated by ISO 6709, and seems like the most common order.
    • GeoJSON, on the other hand, uses lon/lat order.
  • OpenStreetMap uses the WGS84 spatial reference system used by the Global Positioning System (GPS).
  • OpenStreetMap node coordinates have seven decimal places, which gives them centimetric precision. However, the position accuracy of GPS data is only about 10m. A reasonable display accuracy could be five places, which is precise to 1.1 metres at the equator.
  • Spatial features that cross the 180th meridian are problematic, since you go from longitude 180.0 to -180.0. Such features usually have their geometries split up, like the area of Russia.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aio_overpass-0.13.0.tar.gz (53.8 kB view hashes)

Uploaded Source

Built Distribution

aio_overpass-0.13.0-py3-none-any.whl (54.6 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page