What method of parsing does XMLTextReader use ?

Question

What method of parsing does XMLTextReader use ? I am confused. Sax or Dom ?

Answer 1

Neither, really - it's a pull parser. It neither builds up a complete object model in memory (DOM) nor does it parse the whole document, raising events that the client can react to (SAX).

Instead, the client controls the parsing, asking the parser to do a bit more reading, then checking its state and reacting accordingly, then asking it to read a bit more etc.

Answer 2

DOM parsers read the entire XML document and build a representation of it as a hierarchy of objects in memory.

SAX parsers read one XML token at a time. You supply callback methods that the parser should call when it encounters certain kinds of things in the document.

XMLTextReader reads one XML token at a time so it is closer to SAX but it doesn't provide callbacks.

Answer 3

First of all, SAX is a Java only parser api. The language-agnostic term for this is push parsing, and XmlTextReader doesn't use this technique.

DOM is a W3C standard, .NET does implement DOM Levels 1 and 2. However this functionality is implemented in the class XmlDocument, which can load a document using an XmlTextReader.

XmlTextReader parses XML using pull parsing (the Java equivalent of this is StAX) which is basically the inverse of SAX/push parsing. Instead of being notified when a new entity is read, you ask for the next entity in the document when you're ready for it (btw, it's possible to implement a SAX style parser on top of a pull parser, but not the other way around).