Docs Menu
Docs Home
/ /
Atlas Device SDKs
/ / /

Geospatial - Node.js SDK

On this page

  • Geospatial Data Types
  • GeoPoint
  • GeoCircle
  • GeoBox
  • GeoPolygon
  • Persist Geospatial Data
  • Create a GeoJSON-Compatible Class
  • Use the Embedded Class
  • Query Geospatial Data

New in version 12.0.0.

Geospatial data, or "geodata", specifies points and geometric objects on the Earth's surface. With the geodata types, you can create queries that check whether a given point is contained within a shape. For example, you can find all coffee shops within 15 km of a specified point.

Changed in version 12.3.0: Geospatial data supported in Atlas Device Sync

Realm Node.js SDK v12.3.0 and later adds support for geospatial data in Atlas Device Sync. This allows you to subscribe to geospatial queries in a synced realm. If you try to subscribe to a geospatial query with an older version of the SDK, you will receive a server error with a compensating write. For more information about managing your Sync subscriptions, refer to Manage Sync Subscriptions - Node.js SDK.

For more information on querying geospatial data with Device Sync, refer to Geospatial Data in the App Services documentation.

The Node.js SDK supports geospatial queries using the following data types:

  • GeoPoint

  • GeoCircle

  • GeoBox

  • GeoPolygon

The SDK provides these geospatial data types to simplify querying geospatial data. You cannot persist these data types directly.

For information on how to persist geospatial data, refer to the Persist Geospatial Data section on this page.

A GeoPoint defines a specific location on the Earth's surface. All of the geospatial data types use GeoPoints to define their location.

GeoPoint can be one of three types:

  • An object with:

    • latitude: a number value

    • longitude: a number value

    • altitude: an optional number value

  • CanonicalGeoPoint: an interface that satisfies the GeoJSON specifications for a point

  • GeoPosition: an array with longitude, latitude, and an optional altitude

A GeoPoint is used only as a building block of the other shapes: GeoCircle, GeoBox, and GeoPolygon. These shapes, and the GeoPoint type, are used in queries, not for persistance.

To save geospatial data to realm, refer to Persist Geospatial Data.

A GeoCircle defines a circle on the Earth's surface. You define a GeoCircle by providing:

  • A GeoPoint for the center of the circle

  • A number for the distance (radius) of the circle

The radius distance uses radians as the unit of measure. The SDK provides the methods kmToRadians and miToRadians to convert kilometers or miles to radians.

The following code shows two examples of creating a circle:

Two GeoCircles
click to enlarge

A GeoBox defines a rectangle on the Earth's surface. You define the rectangle by specifying the bottom left (southwest) corner and the top right (northeast) corner. A GeoBox behaves in the same way as the corresponding GeoPolygon.

The following example creates 2 boxes:

2 GeoBoxes
click to enlarge

A GeoPolygon defines a polygon on the Earth's surface. Because a polygon is a closed shape, you must provide a minimum of 4 points: 3 points to define the polygon's shape, and a fourth to close the shape.


The fourth point in a polygon must be the same as the first point.

You can also exclude areas within a polygon by defining one or more "holes". A hole is another polygon whose bounds fit completely within the outer polygon. The following example creates 3 polygons: one is a basic polygon with 5 points, one is the same polygon with a single hole, and the third is the same polygon with two holes:

3 GeoPolygons
click to enlarge


Cannot Persist Geospatial Data Types

Currently, you can only persist geospatial data. Geospatial data types cannot be persisted directly. For example, you can't declare a property that is of type GeoBox.

These types can only be used as arguments for geospatial queries.

If you want to persist geospatial data, it must conform to the GeoJSON spec.

To create a class that conforms to the GeoJSON spec, you:

  1. Create an embedded realm object. For more information about embedded objects, refer to Embedded Objects - Node.js SDK.

  2. At a minimum, add the two fields required by the GeoJSON spec:

    • A field of type double[] that maps to a "coordinates" (case sensitive) property in the realm schema.

    • A field of type string that maps to a "type" property. The value of this field must be "Point".

To simplify geodata persistance, you can define a model that implements CanonicalGeoPoint, which already has the correct shape. The following example shows an embedded class named MyGeoPoint that is used to persist geospatial data:

You then use the custom MyGeoPoint class in your realm model, as shown in the following example:

You add instances of your class to the realm just like any other Realm model. However, in this example, because the MyGeoPoint class does not extend Realm.Object, we must specify MyGeoPoint.schema when opening the realm:

The following image shows the results of creating these two company objects.

2 GeoPoints
click to enlarge

To query against geospatial data, you can use the geoWithin operator with RQL. The geoWithin operator takes the "coordinates" property of an embedded object that defines the point we're querying, and one of the geospatial shapes to check if that point is contained within the shape.


The format for querying geospatial data is the same, regardless of the shape of the geodata region.

The following examples show querying against various shapes to return a list of companies within the shape:


Querying a GeoCircle example.
click to enlarge


Querying a GeoBox example.


Querying a GeoPolygon example.
click to enlarge
← Embedded Objects - Node.js SDK