<item>...</item>

How much impact does removing the assumption that every document has a
root node of '<item/>' have on the system?  Some things I can think of
right away:

 - documents get an @id attribute inserted into the root element

 - pubDate may not be valid in non-<item/> documents (trivial to fix,
   just check the rootElement name before adding)

 - searching via Xpath might be less intuitive (e.g., 
   http://www.domain.com/WK/blog///item wouldn't return all documents)

 - this might affect indexing by dbxml, but I have no idea whether this
   is true or not

I see that it's perfectly possible to POST non-<item/> documents to the
site, thought they'll get a <pubDate/> element added to them and an @id
in the root element.  It looks to me though as if any xmlns declarations
are being stripped.  Additionally, it looks like Xpath isn't happy with
much to do with a namespace when searching.  Hm.

Rick
-- 
 http://www.rickbradley.com    MUPRN: 142
                       |  away for a
   random email haiku  |  delivery that would have
                       |  no return address.


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf