> '"pull" model parser' - does this mean optimizing based on the schema? No. A Sax-style parser is where you parse and catch events... which means you have to set up a state machine in your ContentHandler... a task which tends to be not very fun for most casual developers. A pull model parser would be where you would ask to parse up to the next node, handle the event, and then continue parsing, perhaps all in a single loop. I'm not certain of the exact API, but the model is easier, if maybe not quite as performant, for many tasks. Ted may be able to tell you more about what the API should look like, since he was thinking of this well over a year ago. I think the API would be something along the lines of a NodeIterator? Note they have XMLTextReader and XMLNodeReader as concrete implementations of this interface... which means you should be able to use the same code over both an XML stream and a DOM tree. I think. > ditto for "data-friendly DOM" - does this mean a schema-centric DOM? > or designed for large datasets? No. It means an API that is set up to be friendly for table-shaped data, as in an export from ODBC rowsets, rather than irregular data as you might find in a document. At least 50% of XML used in business transactions is likely to be table-shaped-regular, so this makes sense. For the database crowd. Not sure of details. > How is XmlNavigator different from the selectNodes() method, which has been > in MSXML for over 2 years? It's an itterator object. In one sense it's doing the same thing. But I think select nodes was returning a NodeList, or some such. An XmlNavigator provides cursor style access to the data. The major methods on it seem to be: void Select(String xpath); bool MoveNextSelected(); void MoveChildren(XmlNavigator other); Don't take what I'm saying as gospel. I'm just learning about this stuff, and have somewhat incomplete data. -scott "Ed Staub" one.net> cc: (bcc: Scott Boag/CAM/Lotus) Subject: RE: Microsoft XML stuff 07/12/2000 09:29 PM Please respond to general Scott, Thanks for the report from the front. Can you elaborate on these? I might have understood, or I might not have... '"pull" model parser' - does this mean optimizing based on the schema? ditto for "data-friendly DOM" - does this mean a schema-centric DOM? or designed for large datasets? How is XmlNavigator different from the selectNodes() method, which has been in MSXML for over 2 years? -Ed -----Original Message----- From: Scott Boag/CAM/Lotus [mailto:Scott_Boag@lotus.com] Sent: Wednesday, July 12, 2000 3:46 PM To: general@xml.apache.org Subject: Microsoft XML stuff BTW, as I sit here in the Microsoft Developers Conference, they are talking about a "pull" model parser. Something to think about. They call this an XmlReader. They are also talking about a data-friendly DOM -- i.e. a DOM that is friendly and specialized for data-shaped XML, i.e. the kind of XML you might get from a database. They also have an XmlNavigator, which is basically a NodeIterator that is a "lens" onto the XML tree, using XPath. This is similar or exactly what I am doing in the XPath module in Xalan2. Their latest XSLT architecture looks suspiciously like TrAX. Probably a good thing. Note that they are defining a SAX (like) api for their parser. -scott --------------------------------------------------------------------- In case of troubles, e-mail: webmaster@xml.apache.org To unsubscribe, e-mail: general-unsubscribe@xml.apache.org For additional commands, e-mail: general-help@xml.apache.org --------------------------------------------------------------------- In case of troubles, e-mail: webmaster@xml.apache.org To unsubscribe, e-mail: general-unsubscribe@xml.apache.org For additional commands, e-mail: general-help@xml.apache.org