Login
User name:
Password:
Remember me 
Powered by BlogHarbor
Powered by BlogHarbor
Re: Progress on XQuery Update
by John Cowan
2) FWIU, the whole point about XQuery Update (which makes it different from the proposed XQueryP) is that none of the updates actually do anything until either upd:applyUpdates is done (in which case all trees are rebuilt and revalidated) or the topmost expression in the query completes. XQuery update is essentially a monadic extension to XQuery: what is manipulated are ordinary immutable objects which are interpreted *outside the XQuery engine* as commands to make changes. Transform expressions also apply updates, but they apply them to produce an entirely novel tree. This is like the Command pattern, but the commands are applied only after the engine completes. So you don't need a mutable tree, you need a novel data structure that represents mutations. 5) The obvious answer is: if the document was retrieved by HTTP GET, it should be updated by HTTP PUT, as that is the semantics of PUT. If the PUT doesn't work, then you have to decide what to do: throw an error or write a local file or what not.
Post comment:
  Receive comment notifications for this article
Subject: 
Comment: 
Comment verification:

Please enter the text you see inside the graphic to post your comment:
This blog does not allow anonymous comments. Please provide your username and password along with your comment.
Login information:
Username: 
Password: 
If you would like to post contact information on your comment, please enter your information into the optional fields below:
Contact information:
URL:  example: http://yourdomain.com