A
word of background: A DOM (document object model) style parser reads your XML and generates a tree of objects. I find working with the tree more complex than I need for simple stuff like reading a configuration file, but it is very good if you want to modify the data and write it back to new XML or use some other advanced features.
A SAX style parser calls a "handler" object of your choice every time it finds something interesting in the XML, like the start of a tag, some characters, the end of the tag, etc. It is relatively easy to write your own handler to collect the characters SAX sends you into records as it goes along. Since SAX doesn't build a tree in memory, it can use less memory. But it can't rewrite the tree to a new XML file.
Here is some work I did with SAX parsing. Make it your second tutorial on SAX as it assumes you did well with your first.
http://www.surfscranton.com/architecture/ReusableSAXParsers.htm
