geojsplit

A python implementation of the node package geojsplit: https://github.com/woodb/geojsplit

Installation

With poetry

For an introduction to poetry.

$ poetry add geojsplit

will add geojsplit to your current virtual environment and update your poetry.lock file. If you would like to contribute or develop geojsplit

$ git clone https://github.com/underchemist/geojsplit.git
$ cd geojsplit
$ poetry install

You may need some extra configuration to make poetry play nice with conda virtual environments

poetry config settings.virtualenvs.path <path_to_conda_install>/envs  # tell poetry where you virtualenvs are stored
poetry config settings.virtualenvs.create 0  # tell poetry not to try to create its own virtualenvs.

See https://github.com/sdispater/poetry/issues/105#issuecomment-498042062 for more info.

$ poetry config settings.virtualenvs.path $CONDA_ENV_PATH
$ poetry config settings.virtualenvs.create 0

With pip

Though geojsplit is developed using poetry (and as such does not have a setup.py), pep517 implementation in pip means we can install it directly

$ pip install geojsplit

Usage

Although both the library code and the command line tool of geojsplit are relatively simple, there are use cases for both. You may want to use the backend GeoJSONBatchStreamer class directly in order to do more sophisticated manipulations with GeoJOSN documents. As a command line tool geojsplit also works well as a preprocessing step for working with large GeoJSON documents i.e. for piping into GDAL’s ogr2ogr tool.

As a library

Once installed, geojsplit can be imported in like

from geojsplit import geojsplit

geojson = geojsplit.GeoJSONBatchStreamer("/path/to/some.geojson")

for feature_collection in geojson.stream():
    do_something(feature_collection)
    ...

If the /path/to/some.geojson does not exists, FileNotFound will be raised.

You can control how many features are streamed into a Feature Collection using the batch parameter (Default is 100).

>>> g = geojson.stream(batch=2)  # instatiate generator object
>>> data = next(g)
>>> print(data)
{"features": [{"geometry": {"coordinates": [[[-118.254638, 33.7843], [-118.254637,
33.784231], [-118.254556, 33.784232], [-118.254559, 33.784339], [-118.254669,
33.784338], [-118.254668, 33.7843], [-118.254638, 33.7843]]], "type": "Polygon"},
"properties": {}, "type": "Feature"}, {"geometry": {"coordinates": [[[-118.254414,
33.784255], [-118.254232, 33.784255], [-118.254232, 33.784355], [-118.254414,
33.784355], [-118.254414, 33.784255]]], "type": "Polygon"}, "properties": {}, "type":
"Feature"}], "type": "FeatureCollection"}
>>> print(len(data["features"]))
2

If your GeoJSON document has a different format or you want to iterate over different elements on your document, you can also pass a different value to the prefix keyword argument (Default is 'features.item'). This is an argument passed directly down to a ijson.items call, for more information see https://github.com/ICRAR/ijson.

As a command line tool

After installing you should have the geojsplit executable in your PATH.

$ geojsplit -h
usage: geojsplit [-h] [-l GEOMETRY_COUNT] [-a SUFFIX_LENGTH] [-o OUTPUT]
                [-n LIMIT] [-v] [-d] [--version]
                geojson

Split a geojson file into many geojson files.

positional arguments:
geojson               filename of geojson file to split

optional arguments:
-h, --help            show this help message and exit
-l GEOMETRY_COUNT, --geometry-count GEOMETRY_COUNT
                        the number of features to be distributed to each file.
-a SUFFIX_LENGTH, --suffix-length SUFFIX_LENGTH
                        number of characters in the suffix length for split
                        geojsons
-o OUTPUT, --output OUTPUT
                        output directory to save split geojsons
-n LIMIT, --limit LIMIT
                        limit number of split geojson file to at most LIMIT,
                        with GEOMETRY_COUNT number of features.
-v, --verbose         increase output verbosity
-d, --dry-run         see output without actually writing to file
--version             show geojsplit version number

By default splitted GeoJSON files are saved as filename_x<SUFFIX_LENGTH characters long>.geojson. Default SUFFIX_LENGTH is 4, meaning that 456976 unique files can be generated. If you need more use -a or --suffix-length to increase this value appropriately.

The --geometry-count flag corresponds to the batch keyword argument for GeoJSONBatchStreamer.stream method. Note that if GEOMETRY_COUNT does not divide equally into the number of features in the Feature Collection, the last batch of features will be < GEOMETRY_COUNT.

Finally, to only iterate over the the first n elements of a GeoJSON document, use --limit.