Click-based argument and option decorators for Python GIS command line programs
Python
Latest commit 892bd5b Nov 7, 2016 @sgillies sgillies committed on GitHub Merge pull request #14 from mapbox/geoms-as-features
Geometry sequence as features

README.rst

cligj

https://coveralls.io/repos/mapbox/cligj/badge.png?branch=master

Common arguments and options for GeoJSON processing commands, using Click.

cligj is for Python developers who create command line interfaces for geospatial data. cligj allows you to quickly build consistent, well-tested and interoperable CLIs for handling GeoJSON.

Arguments

files_in_arg Multiple files

files_inout_arg Multiple files, last of which is an output file.

features_in_arg GeoJSON Features input which accepts multiple representations of GeoJSON features and returns the input data as an iterable of GeoJSON Feature-like dictionaries

Options

verbose_opt

quiet_opt

format_opt

JSON formatting options

indent_opt

compact_opt

Coordinate precision option

precision_opt

Geographic (default), projected, or Mercator switch

projection_geographic_opt

projection_projected_opt

projection_mercator_opt

Feature collection or feature sequence switch

sequence_opt

use_rs_opt

GeoJSON output mode option

geojson_type_collection_opt

geojson_type_feature_opt

def geojson_type_bbox_opt

Example

Here's an example of a command that writes out GeoJSON features as a collection or, optionally, a sequence of individual features. Since most software that reads and writes GeoJSON expects a text containing a single feature collection, that's the default, and a LF-delimited sequence of texts containing one GeoJSON feature each is a feature that is turned on using the --sequence option. To write sequences of feature texts that conform to the JSON Text Sequences proposed standard (and might contain pretty-printed JSON) with the ASCII Record Separator (0x1e) as a delimiter, use the --rs option

import click
import cligj
import json

def process_features(features):
    for feature in features:
        # TODO process feature here
        yield feature

@click.command()
@cligj.features_in_arg
@cligj.sequence_opt
@cligj.use_rs_opt
def pass_features(features, sequence, use_rs):
    if sequence:
        for feature in process_features(features):
            if use_rs:
                click.echo(u'\x1e', nl=False)
            click.echo(json.dumps(feature))
    else:
        click.echo(json.dumps(
            {'type': 'FeatureCollection',
             'features': list(process_features(features))}))

On the command line, the generated help text explains the usage

Usage: pass_features [OPTIONS] FEATURES...

Options:
--sequence / --no-sequence  Write a LF-delimited sequence of texts
                            containing individual objects or write a single
                            JSON text containing a feature collection object
                            (the default).
--rs / --no-rs              Use RS (0x1E) as a prefix for individual texts
                            in a sequence as per http://tools.ietf.org/html
                            /draft-ietf-json-text-sequence-13 (default is
                            False).
--help                      Show this message and exit.

And can be used like this

$ cat data.geojson
{'type': 'FeatureCollection', 'features': [{'type': 'Feature', 'id': '1'}, {'type': 'Feature', 'id': '2'}]}

$ pass_features data.geojson
{'type': 'FeatureCollection', 'features': [{'type': 'Feature', 'id': '1'}, {'type': 'Feature', 'id': '2'}]}

$ cat data.geojson | pass_features
{'type': 'FeatureCollection', 'features': [{'type': 'Feature', 'id': '1'}, {'type': 'Feature', 'id': '2'}]}

$ cat data.geojson | pass_features --sequence
{'type': 'Feature', 'id': '1'}
{'type': 'Feature', 'id': '2'}

$ cat data.geojson | pass_features --sequence --rs
^^{'type': 'Feature', 'id': '1'}
^^{'type': 'Feature', 'id': '2'}

In this example, ^^ represents 0x1e.