Atomization

Atomization is the process of converting a sequence of items into a sequence of atomic values. Atomization is used by expressions whenever a sequence of atomic values is required.

Each item in a sequence is converted to an atomic value by applying the following rules:

• If the item is an atomic value, then the atomic value is returned.
• If the item is a node, its typed value is returned. The typed value of a node is a sequence of zero or more atomic values that can be extracted from the node. If the node has no typed value, then an error is returned. If an XML document is validated with a schema, DB2® for i does not keep the type annotation for each node. The data is always stored as untyped. The typed value of an untyped element or attribute is the string value as an instance of xs:untypedAtomic.

Implicit atomization of a sequence produces the same result as invoking the fn:data function explicitly on a sequence.

For example, the following sequence contains a combination of nodes and atomic values:

("Some text", <anElement>More text</anElement>, 1001)

Applying atomization to this sequence results in the following sequence of atomic values:

("Some text", "More text", 1001)

The following XPath expressions use atomization to convert items into atomic values:

• Arithmetic expressions
• Comparison expressions
• Function calls with arguments whose expected types are atomic