Skip to main content

Query a Spatial Index

  • Documents that contain spatial data can be queried by spatial queries that employ geographical criteria.
    You have two options:

    • Dynamic spatial query
      Either make a dynamic spatial query on a collection (see how to make a spatial query).
      An auto-index will be created by the server.

    • Spatial index query
      Or, index your documents' spatial data in a static-index (see indexing spatial data)
      and then make a spatial query on this index ( described in this article ).

  • A few examples of querying a spatial index are provided below.
    A spatial query performed on a static-index is similar to the dynamic spatial query.
    Find all spatial API methods listed here.

  • Examples in this page:

Search by radius

  • Query the spatial index:

  • Use the WithinRadius method to search for all documents containing spatial data that is located
    within the specified distance from the given center point.

// Define a spatial query on index 'Events_ByNameAndCoordinates'
List<Event> employeesWithinRadius = session
.Query<Event, Events_ByNameAndCoordinates>()
// Call 'Spatial' method
.Spatial(
// Pass the spatial index-field containing the spatial data
"Coordinates",
// Set the geographical area in which to search for matching documents
// Call 'WithinRadius', pass the radius and the center points coordinates
criteria => criteria.WithinRadius(20, 47.623473, -122.3060097))
.ToList();

// The query returns all matching Event entities
// that are located within 20 kilometers radius
// from point (47.623473 latitude, -122.3060097 longitude).

Search by shape

  • Query the spatial index:
    Use the RelatesToShape method to search for all documents containing spatial data that is located
    in the specified relation to the given shape.

  • The shape in the query is specified as either a circle or a polygon in a WKT format.
    See polygon rules here.

  • The relation to the shape can be one of: Within, Contains, Disjoint, Intersects.

  • See more usage examples in the dynamic search by shape query.

// Define a spatial query on index 'EventsWithWKT_ByNameAndWKT'
List<EventWithWKT> employeesWithinShape = session
.Query<EventWithWKT, EventsWithWKT_ByNameAndWKT>()
// Call 'Spatial' method
.Spatial(
// Pass the spatial index-field containing the spatial data
"WKT",
// Set the geographical search criteria, call 'RelatesToShape'
criteria => criteria.RelatesToShape(
// Specify the WKT string
shapeWkt: @"POLYGON ((
-118.6527948 32.7114894,
-95.8040242 37.5929338,
-102.8344151 53.3349629,
-127.5286633 48.3485664,
-129.4620208 38.0786067,
-118.7406746 32.7853769,
-118.6527948 32.7114894
))",
// Specify the relation between the WKT shape and the documents spatial data
relation: SpatialRelation.Within))
.ToList();

// The query returns all matching Event entities
// that are located within the specified polygon.
  • Note:
    The index in the above example indexes a WKT string in the spatial index-field.
    However, you can query by shape also on spatial data that is indexed as lat/lng coordinates.

Sort results

  • Query the spatial index:
    Use OrderByDistance or OrderByDistanceDescending to sort the results by distance from a given point.

  • By default, distance in RavenDB measured in kilometers.
    The distance can be rounded to a specific range.

// Define a spatial query on index 'Events_ByNameAndCoordinates'
List<Event> employeesSortedByDistance = session
.Query<Event, Events_ByNameAndCoordinates>()
// Filter results by geographical criteria
.Spatial(
"Coordinates",
criteria => criteria.WithinRadius(20, 47.623473, -122.3060097))
// Sort results, call 'OrderByDistance'
.OrderByDistance(
// Pass the spatial index-field containing the spatial data
"Coordinates",
// Sort the results by their distance from this point:
47.623473, -122.3060097)
.ToList();

// Return all matching Event entities located within 20 kilometers radius
// from point (47.623473 latitude, -122.3060097 longitude).

// Sort the results by their distance from a specified point,
// the closest results will be listed first.