geojson-pydantic 0.7.0

Pydantic data models for the GeoJSON spec.

geojson-pydantic

Pydantic models for GeoJSON.

Description

geojson_pydantic provides a suite of Pydantic models matching the GeoJSON specification rfc7946. Those models can be used for creating or validating geojson data.

Install

$ python -m pip install -U pip
$ python -m pip install geojson-pydantic

Or install from source:

$ python -m pip install -U pip
$ python -m pip install git+https://github.com/developmentseed/geojson-pydantic.git

Install with conda from conda-forge:

$ conda install -c conda-forge geojson-pydantic

Usage

from geojson_pydantic import Feature, FeatureCollection, Point

geojson_feature = {
    "type": "Feature",
    "geometry": {
        "type": "Point",
        "coordinates": [13.38272, 52.46385],
    },
    "properties": {
        "name": "jeff",
    },
}


feat = Feature(**geojson_feature)
assert feat.type == "Feature"
assert type(feat.geometry) == Point
assert feat.properties["name"] == "jeff"

fc = FeatureCollection(type="FeatureCollection", features=[geojson_feature, geojson_feature])
assert fc.type == "FeatureCollection"
assert len(fc) == 2
assert type(fc.features[0].geometry) == Point
assert fc.features[0].properties["name"] == "jeff"

Advanced usage

In geojson_pydantic we've implemented pydantic's Generic Models which allow the creation of more advanced models to validate either the geometry type or the properties.

In order to make use of this generic typing, there are two steps: first create a new model, then use that model to validate your data. To create a model using a Generic type, you HAVE TO pass Type definitions to the Feature model in form of Feature[Geometry Type, Properties Type]. Then pass your data to this constructor.

By default Feature and FeatureCollections are defined using geojson_pydantic.geometries.Geometry for the geometry and typing.Dict for the properties.

Here's an example where we want to validate that GeoJSON features have Polygon types, but don't do any specific property validation.

from typing import Dict

from geojson_pydantic import Feature, Polygon
from pydantic import BaseModel

geojson_feature = {
    "type": "Feature",
    "geometry": {
        "type": "Point",
        "coordinates": [13.38272, 52.46385],
    },
    "properties": {
        "name": "jeff",
    },
}

# Define a Feature model with Geometry as `Polygon` and Properties as `Dict`
MyPolygonFeatureModel = Feature[Polygon, Dict]

feat = MyPolygonFeatureModel(**geojson_feature)  # should raise Validation Error because `geojson_feature` is a point
>>> ValidationError: 3 validation errors for Feature[Polygon, Dict]
...
geometry -> type
  unexpected value; permitted: 'Polygon' (type=value_error.const; given=Point; permitted=['Polygon'])


geojson_feature = {
    "type": "Feature",
    "geometry": {
        "type": "Polygon",
        "coordinates": [
            [
                [13.38272, 52.46385],
                [13.42786, 52.46385],
                [13.42786, 52.48445],
                [13.38272, 52.48445],
                [13.38272, 52.46385],
            ]
        ],
    },
    "properties": {
        "name": "jeff",
    },
}

feat = MyPolygonFeatureModel(**geojson_feature)
assert type(feature.geometry) == Polygon

Or with optional geometry

from geojson_pydantic import Feature, Point
from typing import Optional

MyPointFeatureModel = Feature[Optional[Point], Dict]

assert MyPointFeatureModel(type="Feature", geometry=None, properties={}).geometry is None
assert MyPointFeatureModel(type="Feature", geometry=Point(coordinates=(0,0)), properties={}).geometry is not None

And now with constrained properties

from geojson_pydantic import Feature, Point
from pydantic import BaseModel, constr

# Define a Feature model with Geometry as `Point` and Properties as a constrained Model
class MyProps(BaseModel):
    name: constr(regex=r'^(drew|vincent)$')

MyPointFeatureModel = Feature[Point, MyProps]

geojson_feature = {
    "type": "Feature",
    "geometry": {
        "type": "Point",
        "coordinates": [13.38272, 52.46385],
    },
    "properties": {
        "name": "jeff",
    },
}

feat = MyPointFeatureModel(**geojson_feature)
>>> ValidationError: 1 validation error for Feature[Point, MyProps]
properties -> name
  string does not match regex "^(drew|vincent)$" (type=value_error.str.regex; pattern=^(drew|vincent)$)

geojson_feature["properties"]["name"] = "drew"
feat = MyPointFeatureModel(**geojson_feature)
assert feat.properties.name == "drew"

Enforced Keys

Starting with version 0.6.0, geojson-pydantic's classes will not define default keys such has type, geometry or properties. This is to make sure the library does well its first goal, which is validating GeoJSON object based on the specification

o A GeoJSON object has a member with the name "type".  The value of
  the member MUST be one of the GeoJSON types.

o A Feature object HAS a "type" member with the value "Feature".

o A Feature object HAS a member with the name "geometry". The value
of the geometry member SHALL be either a Geometry object as
defined above or, in the case that the Feature is unlocated, a
JSON null value.

o A Feature object HAS a member with the name "properties". The
value of the properties member is an object (any JSON object or a
JSON null value).

from geojson_pydantic import Point

## Before 0.6
Point(coordinates=(0,0))
>> Point(type='Point', coordinates=(0.0, 0.0), bbox=None)

## After 0.6
Point(coordinates=(0,0))
>> ValidationError: 1 validation error for Point
   type
      field required (type=value_error.missing)

Point(type="Point", coordinates=(0,0))
>> Point(type='Point', coordinates=(0.0, 0.0), bbox=None)

Contributing

See CONTRIBUTING.md.

Changes

See CHANGES.md.

Authors

Initial implementation by @geospatial-jeff; taken liberally from https://github.com/arturo-ai/stac-pydantic/

See contributors for a listing of individual contributors.

License

See LICENSE