1.2.5 Notes on using large-object data

When using large-object data, be aware of the following restrictions:

Organization of this subsection
(1) Items in which large-object data is not allowed
(2) Scalar functions
(3) Data insertion and updating
(4) Using a locator

(1) Items in which large-object data is not allowed

Large-object data cannot be used in any of the following items:

Large object data specified in a definition as an attribute of an abstract data type can be specified in an index definition by using a plug-in function. Concatenation operations on large-object data can be used only on updated values in the SET clause of an UPDATE statement.

(2) Scalar functions

The only scalar functions in which large-object data can be specified are the LENGTH, SUBSTR, and POSITION functions.

(3) Data insertion and updating

A literal cannot be used in the INSERT statement or UPDATE statement to insert or update large-object data. The following items can be used to insert large-object data: an embedded variable, a ? parameter, an SQL variable, an SQL parameter, the SUBSTR scalar function, a function call, and NULL. The following items are used to update large-object data: a column specification, a component specification, an embedded variable, a ? parameter, an SQL variable, an SQL parameter, a concatenation operation, the SUBSTR scalar function, a function call, a subquery, and NULL.

(4) Using a locator

When handling large object data in a UAP, the use of a locator allows you to process the SQL that handles large object data without storing the data itself on a client. For a description of locators, see 1.13 Locator.