django/docs/ref/contrib/gis/serializers.txt

==================
GeoJSON Serializer
==================

.. versionadded:: 1.8

.. module:: django.contrib.gis.serializers.geojson
   :synopsis: Serialization of GeoDjango models in the GeoJSON format.

GeoDjango provides a specific serializer for the `GeoJSON`__ format. See
:doc:`/topics/serialization` for more information on serialization.

The GDAL library is required if any of the serialized geometries need
coordinate transformations (that is if the geometry's spatial reference system
differs from the ``srid`` serializer option).

.. versionchanged:: 1.9

    The GeoJSON serializer no longer needs GDAL if all geometries are in the
    same coordinate system as the ``srid`` serializer option.

__ http://geojson.org/

The ``geojson`` serializer is not meant for round-tripping data, as it has no
deserializer equivalent. For example, you cannot use :djadmin:`loaddata` to
reload the output produced by this serializer. If you plan to reload the
outputted data, use the plain :ref:`json serializer <serialization-formats-json>`
instead.

In addition to the options of the ``json`` serializer, the ``geojson``
serializer accepts the following additional option when it is called by
``serializers.serialize()``:

* ``geometry_field``: A string containing the name of a geometry field to use
  for the ``geometry`` key of the GeoJSON feature. This is only needed when you
  have a model with more than one geometry field and you don't want to use the
  first defined geometry field (by default, the first geometry field is picked).

* ``srid``: The SRID to use for the ``geometry`` content. Defaults to 4326
  (WGS 84).

The :ref:`fields <subset-of-fields>` option can be used to limit fields that
will be present in the ``properties`` key, as it works with all other
serializers.

Example::

    from django.core.serializers import serialize
    from my_app.models import City

    serialize('geojson', City.objects.all(),
              geometry_field='point',
              fields=('name',))

Would output::

    {
      'type': 'FeatureCollection',
      'crs': {
        'type': 'name',
        'properties': {'name': 'EPSG:4326'}
      },
      'features': [
        {
          'type': 'Feature',
          'geometry': {
            'type': 'Point',
            'coordinates': [-87.650175, 41.850385]
          },
          'properties': {
            'name': 'Chicago'
          }
        }
      ]
    }