|
Fundamental to all information systems is the need to search through a large quantity of data, in order to find a subset that satisfies the user’s query. The distinguishing characteristics of geographical data retrieval is that it is expressed in terms of spatial locations and spatial relationships.
As shown before in general in the introduction, spatial queries may be either location-based (geometry-based) or phenomenon-based (attribute-based), or a combination of the two (see B-AN: Spatial Queries).
This type of request selects features or records geographic features that satisfy a statement expressing a set of conditions that forms the basis for the retrieval. The expression considers only conditions for the attributes describing the features. In this case the required result may be generated from the intersection of several layers corresponding to particular thematically specific phenomena.
The need to access data specified in terms of geometry (points, lines and polygons) and of spatial relationships between
them, has introduced the requirement for specialized storage and data-search procedures. This is because, in such a case we
need to be able to retrieve records based on some spatial properties, which are not stored explicitly in the database.
In a relatively simple case, known as a range search, the query may request all data or particular classes that are inside
a rectangular spatial window defined by ranges of coordinates in two dimensions. Stored geometric objects may actually lie
within the ranges, i.e. be entirely inside, in which case they can be retrieved as a whole. Alternatively they may overlap
the range, in which case the overlapping objects may need to be clipped at the boundary of the ranges of the search region
to find the part that are inside.
Queries that include topological relationships between phenomena may make use of stored topological relations. Commonly used procedures are those to test whether a point, a line or polygon is located inside a specified polygon. Other related procedures test whether geometric objects are coincident or adjacent with each other.