I couldn't understand questions 1 & 2, so I haven't answered them.
| 3/ If i understand the event-based philosophy, an XML parser do not
| need to know something about DOM objects (no "new Element()", "new
| Comment()" called within the code of the parser !) ?
That's right. A sensible way to deal with this might be to take a
layered approach and give the parser just a simple SAX-like event
interface with callbacks for all the different events.
Then, you might make a special client to the event interface that uses
the event callbacks to build a DOM tree.
Your event interface would probably have more events than SAX, so a
SAX driver would probably use another special event client that only
echoed some of the events to the SAX handlers.
-- "These are, as I began, cumbersome ways / to kill a man. Simpler, direct, and much more neat / is to see that he is living somewhere in the middle / of the twentieth century, and leave him there." -- Edwin Brockhttp://www.stud.ifi.uio.no/~larsga/ http://birk105.studby.uio.no/