Call Level Interface Guide and Reference

Using User Defined Types (UDT)

In addition to the SQL data types (referred to as base SQL data types) defined in "Data Types and Data Conversion", new distinct types can be defined by the user. These user defined types (UDTs) share its internal representation with an existing type, but is considered to be a separate and incompatible type for most operations. These UDTs are created using the CREATE DISTINCT TYPE SQL statement.

UDTs help provide the strong typing control needed in object oriented programming by ensuring that only those functions and operators explicitly defined on a distinct type can be applied to its instances. Applications continue to work with C data types for application variables, and only need to consider the UDT types when constructing SQL statements.

This means:

• All SQL to C data type conversion rules that apply to the built-in type apply to the UDT.

• The UDT will have the same default C Type as the built-in type.

• SQLDescribeCol() will return the built-in type information. The user defined type name can be obtained by calling SQLColAttribute() with the input descriptor type set to SQL_DESC_DISTINCT_TYPE.

• SQL predicates that involve parameter markers must be explicitly cast to the UDT. This is required since the application can only deal with the built-in types, so before any operation can be performed using the parameter, it must be cast from the C built-in type to the UDT; otherwise an error will occur when the statement is prepared. Refer to "User Defined Types in Predicates" for more information.

For complete rules and a description of user defined types(UDT) refer to the SQL Reference.

[ DB2 List of Books | Search the DB2 Books ]