GeoDjango & GeoJSON¶

django-d3-bridge integrates with GeoDjango for geographic visualizations.

Auto-Serialization¶

Pass a GeoQuerySet directly — the geometry field is auto-detected and serialized to GeoJSON:

from d3_bridge import ChoroplethMap

chart = ChoroplethMap(
    geodata=Region.objects.filter(country="CI"),
    value_field="population",
    title="Population by Region",
)

This automatically:

Finds the geometry field on the model
Serializes all non-geometry fields as GeoJSON properties
Builds a FeatureCollection with id, geometry, and properties

Controlling Fields¶

Limit which fields are included in properties:

chart = ChoroplethMap(
    geodata=Region.objects.all(),
    value_field="gdp",
    properties=["name", "gdp", "population"],  # only these in tooltip
)

Specify Geometry Field¶

If your model has multiple geometry fields:

chart = ChoroplethMap(
    geodata=Parcel.objects.all(),
    value_field="risk_score",
    geometry_field="polygon",  # use this specific geometry
)

Raw GeoJSON¶

You can also pass pre-built GeoJSON:

import json

with open("boundaries.geojson") as f:
    geojson = json.load(f)

chart = ChoroplethMap(geodata=geojson, value_field="score")

GeoJSON via API (Polling)¶

For large datasets, serve GeoJSON via an endpoint:

from d3_bridge.views import ChartDataView

urlpatterns = [
    path("api/geo/regions/", ChartDataView.as_view(
        queryset=Region.objects.all(),
        geojson=True,
        geometry_field="geom",
    )),
]

chart = ChoroplethMap(
    geodata=Region.objects.all(),
    value_field="score",
    poll_url="/api/geo/regions/",
    poll_interval=60,
)

Projections¶

Projection	Best for
`mercator`	General purpose (default)
`naturalEarth`	World maps
`albers`	USA/continental
`equirectangular`	Simple lat/lon grid
`orthographic`	Globe view

chart = ChoroplethMap(
    geodata=data,
    value_field="value",
    projection="naturalEarth",
)

Centering on a Region¶

chart = ChoroplethMap(
    geodata=data,
    value_field="value",
    center=[-5.5, 7.5],  # [longitude, latitude]
    scale=3000,
)

Without center, the map auto-fits to the data extent.