Introduction to spatial data

Spatial data is data that describes the position, shape, and orientation of objects in a defined space. Spatial data in SQL Anywhere is represented as 2D geometries in the form of points, curves (line strings and strings of circular arcs), and polygons. For example, the following image shows the state of Massachusetts, representing the union of polygons representing zip code regions.

Image showing Massachusetts broken into colorful polygons, with each polygon representing a zip code region.

Two common operations performed on spatial data are calculating the distance between geometries, and determining the union or intersection of multiple objects. These calculations are performed using predicates such as intersects, contains, and crosses.

Example of how spatial data might be used

Spatial data support in SQL Anywhere lets application developers associate spatial information with their data. For example, a table representing companies could store the location of the company as a point, or store the delivery area for the company as a polygon. This could be represented in SQL as:

CREATE TABLE Locations(
   ID INT,
   ManagerName CHAR(16),
   StoreName CHAR(16),
   Address ST_Point,
   DeliveryArea ST_Polygon )

The spatial data type ST_Point in the example represents a single point, and ST_Polygon represents an arbitrary polygon. With this schema, the application could show all company locations on a map, or find out if a company delivers to a particular address using a query similar to the following:

CREATE VARIABLE @pt ST_Point;
SET @pt = ST_Geometry::ST_GeomFromText( 'POINT(1 1)' );

SELECT * FROM Locations
WHERE DeliveryArea.ST_Contains( @pt ) = 1

SQL Anywhere provides storage and data management features for spatial data, allowing you to store information such as geographic locations, routing information, and shape data.

These information pieces are stored as points and various forms of polygons and lines in columns defined with a corresponding spatial data type (such as ST_Point and ST_Polygon). You use methods and constructors to access and manipulate the spatial data. SQL Anywhere also provides a set of SQL spatial functions designed for compatibility with other products.

Object-oriented properties of spatial data types

Sub-types are more specific than parent type.
Sub-types inherit methods of parent type.
Sub-types can be automatically converted to parent type.
Columns or variables can store sub-types. For example, a column of type ST_Geometry(SRID=4326) can store spatial values of any type.
Column or variable typed with a parent type can be cast to, or treated as, a sub-type.