esy-osmfilter

esy-osmfilter is a Python library to read and filter OpenStreetMap data files in the Protocol Buffers (PBF) format and export them to a Python dictionary and/or JSON file.

Install

foo@bar:~$ pip install esy-osmfilter

Usage

In the following example the prefilter of esy-osmfilter is used to extract all pipelines from Liechtenstein. The easiest way to run esy-osmfilter is to download the sample.py from https://gitlab.com/dlr-ve-esy/esy-osmfilter and run it. It is quite similar to the more detailed desription below.

In this example, we start by importing all necessary libraries and methods.

>>> import configparser, contextlib
>>> import os, sys
>>> from esy.osmfilter import osm_colors as CC
>>> from esy.osmfilter import run_filter 
>>> from esy.osmfilter import Node, Way, Relation

Thereafter, the IO filepaths are defined, assuming the osm-pbf file is already downloaded.

>>> PBF_inputfile = os.path.join(os.getcwd(),
...                              'tests/input/liechtenstein-191101.osm.pbf')
>>> JSON_outputfile = os.path.join(os.getcwd(),
...                              'tests/output/LI/liechtenstein-191101.json')

Alternatively, you could also make use of urllib library to retrieve a OSM file:

>>> import urllib.request
>>> if not os.path.exists('tests/input/liechtenstein-191101.osm.pbf'):
...    filename, headers = urllib.request.urlretrieve(
...        'https://download.geofabrik.de/europe/liechtenstein-191101.osm.pbf',
...        filename='liechtenstein-191101.osm.pbf'
...    )
...    PBF_inputfile = filename

In the next step, a prefilter for all pipeline objects is defined. With the prefilter, we accept all way-items that have "man_made" as key and "pipeline" as value in their taglist. The white and black filter are left empty for the moment.

>>> prefilter   = {Node: {}, Way: {"man_made":["pipeline",],}, Relation: {}}
>>> whitefilter = []
>>> blackfilter = []

The run_filter function will allow to filter for OSM items from a pbf-file. We confirm the prefilter phase by setting the boolean variable NewPreFilterData=True.

>>> [Data,_]=run_filter('noname',
...                     PBF_inputfile, 
...                     JSON_outputfile, 
...                     prefilter,
...                     whitefilter, 
...                     blackfilter, 
...                     NewPreFilterData=True, 
...                     CreateElements=False, 
...                     LoadElements=False,
...                     verbose=True)

The prefilter returns the filter results to the Data dictionary. This means all OSM way-items with the tag "man_made"="pipeline" are stored there. But not enough, additionally, all referenced node items of these pipelines are stored there too.

>>> len(Data['Node'])
13
>>> len(Data['Relation'])
0
>>> len(Data['Way'])
2

In this example, we have only found two pipelines and their correspondent 13 nodes.

PLEASE NOTICE:
You can also set "man_made":True to accept items independently of a key value.

In the next step we use run_filter to load the Data dictionary and specify the main filtering results. In this example, we use the blackfilter to exclude possible pipelines substations from our prefiltering results.

>>> blackfilter = [("pipeline","substation"),]

We further only accept the drain pipelines that have the really great name "Wäschgräbli".

>>> whitefilter =[(("waterway","drain"),("name","Wäschgräble")), ]

We initiate the mainfilter phase by setting CreateElements=True.

>>> [_,Elements]=run_filter('funny-waterway-pipelines',
...                            PBF_inputfile, 
...                            JSON_outputfile, 
...                            prefilter,
...                            whitefilter, 
...                            blackfilter, 
...                            NewPreFilterData=False, 
...                            CreateElements=True, 
...                            LoadElements=False,
...                            verbose=True)
>>> len(Elements['funny-waterway-pipelines']['Node'])
0
>>> len(Elements['funny-waterway-pipelines']['Relation'])
0
>>> len(Elements['funny-waterway-pipelines']['Way'])
1

We see, that there is only one way-item left in the Elements dictionary, the other has been filtered out. There are no referenced nodes (or relation members) of the remaining way-item passed to the Elements dictionary.

However, these are still accessible in the Data dictionary.

Esy-osmfilter comes with an export function for GeoJSON files (not implemented for relations yet) which will make thinks a lot easier:

>>> from esy.osmfilter import export_geojson
>>> export_geojson(Elements['funny-waterway-pipelines']['Way'],Data,
... filename='test.geojson',jsontype='Line')

To visualize the output-file just open http://geojson.io and drag it on the screen.

For more details, jump to the documentation.