Extensibility Enhancements

The following new features are designed to improve the extensibility of Dynamic Server.

Enhanced HDR Support for Extensibility Features

High-Availability Data Replication (HDR) now supports replication of the following extended objects:

• All built-in and extended data types.
• User-defined routines.
• R-tree and functional indexes
• TimeSeries DataBlade module

User-defined data types (UDTs) must be logged and must reside in a single database server. Data types with out-of-row data are replicated if the data are stored in an sbspace or in a different table on the same database server.

HDR does not replicate data stored in operating system files nor in persistent (that is, non-temporary) external files. HDR also does not replicate memory objects that are associated with user-defined routines.

To use user-defined data types, user-defined routines, or DataBlade modules with HDR, you must install the user-defined data types, user-defined routines, or DataBlade modules on both the HDR primary and secondary database servers. Register the user-defined data types, user-defined routines, or DataBlade modules only on the HDR primary database server.

For more information see the IBM Informix Dynamic Server Administrator's Guide.

Using an Iterator Function in the FROM Clause of a SELECT Statement

An iterator function can now be specified in the FROM clause of the SELECT statement. (An iterator function is a user-defined function that returns to its calling context more than once, each time returning a value.)

You can query the returned result set of an iterator UDR using a virtual table-interface. You can then manipulate the iterator result set in a number of ways, such as by using the WHERE clause to filter the result set; by joining the UDR result set with other table scans; by running GROUP BY, aggregation, and ORDER BY operations; and so on.

For information on writing iterators, see the IBM Informix User-Defined Routines and Data Types Developer's Guide. For information on using iterators in the FROM clause of a SELECT statement syntax, see the IBM Informix Guide to SQL: Syntax.

Enhanced CREATE FUNCTION and CREATE PROCEDURE Syntax

Several new features improve the functionality of user-defined functions.

Multiple SLVs in the WHERE Clause of SELECT, UPDATE, and INSERT Statements

Because a user-defined function now can return more than one OUT parameter, DML (data manipulation language) statements that use the returned values from function calls as statement-local variables (SLVs) within queries or subqueries can now support multiple SLVs.

For more information on OUT parameters, see Multiple OUT Parameters.

For more information on SLVs, see the IBM Informix Guide to SQL: Syntax.

Declaring Names for Returned Values of an SPL UDR

Releases of Dynamic Server earlier than Version 9.4 support user-defined functions written in the SPL language that return one or more values of specified data types. In this release, the RETURNS (or RETURNING clause) of an SPL function can also declare a name for each returned value. This feature can make it easier for SPL functions to pass column headings to SELECT statements.

For more information, see the IBM Informix User-Defined Routines and Data Types Developer's Guide.