Node identity

Each node has a unique identity. Two nodes are distinguishable even though their names and values might be the same. In contrast, atomic values do not have an identity.

Node identity is not the same as an ID-type attribute. An element in an XML document can be given an ID-type attribute by the document author. A node identity, however, is automatically assigned to every node by the system but is not directly visible to users.

Node identity is used to process the following types of expressions:

• Node comparisons. Node identity is used by the is operator to determine if two nodes have the same identity.
• Path expressions. Node identity is used by path expressions to eliminate duplicate nodes.
• Sequence expressions. Node identity is used by the union, intersect, or except operators to eliminate duplicate nodes.