XML decomposition annotations

Annotated XML schema decomposition relies on annotations added to XML schema documents. These decomposition annotations function as mappings between the elements or attributes of the XML document to their target tables and columns in the database. Decomposition processing refers to these annotations to determine how to decompose an XML document.

The XML decomposition annotations belong to the http://www.ibm.com/xmlns/prod/db2/xdb1 namespace and are identified by the "db2-xdb" prefix throughout the documentation. You can select your own prefix; however, if you do, you must bind your prefix to the following namespace: http://www.ibm.com/xmlns/prod/db2/xdb1. The decomposition process recognizes only annotations that are under this namespace at the time the XML schema is enabled for decomposition.

The decomposition annotations are recognized by the decomposition process only if they are added to element and attribute declarations, or as global annotations, in the schema document. They are either specified as attributes or as part of an <xs:annotation> child element of the element or attribute declaration. Annotations added to complex types, references, or other XML schema constructs are ignored.

Although these annotations exist in the XML schema documents, they do not affect the original structure of the schema document, nor do they participate in the validation of XML documents. They are referred to only by the XML decomposition process.

Two annotations that are core features of the decomposition process are: db2-xdb:rowSet and db2-xdb:column. These annotations specify the decomposed value's target table and column, respectively. These two annotations must be specified in order for the decomposition process to successfully complete. Other annotations are optional, but can be used for further control of how the decomposition process operates.