public interface OMEntityReference
extends OMNode

Represents an unexpanded entity reference in an XML document.

Different XML APIs and object models handle entity references fairly differently:

• In DOM, the way entity references in an XML document are processed depends on the setting for expandEntityReferences. If this property is set to true (default), then the parser will expand entity references and the resulting DOM tree simply contains the nodes resulting from this expansion. If this property is set to false, then the parser will still expand entity references, but the resulting DOM tree will contain EntityReference nodes, the children of which represent the nodes resulting from this expansion. Note that since an entity declaration may contain markup, the children of an EntityReference node may have a type other than Node.TEXT_NODE. Application code not interested in entity references will generally set expandEntityReferences to true in order to avoid the additional programming logic required to process EntityReference nodes.
• In SAX, the parser will always expand entity references and report the events resulting from this expansion to the ContentHandler. In addition to that, if a LexicalHandler is registered, then the parser will report the start and end of the expansion using LexicalHandler.startEntity(String) and LexicalHandler.endEntity(String). This means that the processing of entity references in SAX is similar to DOM with expandEntityReferences set to false. Note that in SAX there is no corresponding configuration property. This makes sense because an application not interested in entity references can simply ignore the LexicalHandler.startEntity(String) and LexicalHandler.endEntity(String) events or not register a LexicalHandler at all.
• In StAX, the way entity references are processed depends on the setting for the XMLInputFactory.IS_REPLACING_ENTITY_REFERENCES property. If this property is set to true (default), then the parser will expand entity references and report only the events resulting from that expansion. If the property is set to false, then the parser no longer expands entity references. Instead, it will report each entity reference using a single XMLStreamConstants.ENTITY_REFERENCE event. XMLStreamReader.getText() can then be used to get the replacement value for the entity. Note that this replacement value may contain unparsed markup. One can see that the way StAX reports entity references is significantly different than DOM or SAX.

Axiom models entity references in the same way as StAX: the node corresponding to an (unexpanded) entity reference only stores the name of the entity as well as the replacement value.