C# in a Nutshell, Second Edition

Name

XmlReader

Synopsis

This class is a simple reader for XML documents. XmlReader provides a non-cached, forward-only navigation through an XML data stream. It does not provide validation, nor does it expand general entities. Two derived classes provide these features: XmlTextReader and XmlValidatingReader.

The XmlReader class parses XML in a streaming-based approach (exemplified by the SAX specification). This means the XML parser presents “interesting pieces” (elements, attributes, namespace declarations, and so forth) in a linear order. Within XmlReader, this ordering of nodes is done using successive calls to the Read( ) method. An XmlReader is not positioned on a node at first—an initial call to Read( ) is required to move to the root node of a document. Subsequent calls to Read( ) move the reader sequentially through the nodes. The NodeType property tells you which type of node the reader is currently positioned on, returning values from the XmlNodeType enumeration. A special node-type value for XmlReader is EndElement. As Read( ) moves through the stream, it can be positioned on an element’s end tag after it has stepped through the element’s children. This is not a real node, in the DOM sense, but is required for XmlReader to parse XML data properly. The Skip( ) method steps through data node by node. A call to Skip( ) moves the reader to the next real node, disregarding the current node’s children.

XML documents can also be parsed in a tree-based approach, using the XmlDocument ...