[xquery-talk] momentary recursion dyslexia

[Howard Katz]

Thank you, David. That clarifies matters. 

For those interested in such things, the October, 2004 working draft was the
first WD to specify that "the last step in a path is allowed to return a
sequence of atomic values." Prior to that version, path expressions were
only allowed to return (possibly empty) sequences of nodes. Since the Mark
Logic content server is based on the earlier May, 2003 WD, its "not a node"
error report is consistent with the earlier spec. 

Galax claims to be compatible with the October, 2004 WD but nevertheless
reports that an atomic value in the leaf position is an error, which appears
to be inconsistent with that claim.

Howard

 > -----Original Message-----
 > From: talk-bounces at xquery.com 
 > [mailto:talk-bounces at xquery.com] On Behalf Of David Carlisle
 > Sent: November 27, 2005 1:50 PM
 > To: howardk at fatdog.com
 > Cc: talk at xquery.com
 > Subject: Re: [xquery-talk] momentary recursion dyslexia
 > 
 > 
 > >     <a/>/fn:name(.)
 > > 
 > > Can someone tell me why this is problematic for these two engines?
 > > Howard 
 > 
 > Check which draft of xquery they implement. The ability for the last
 > step in a path to return an atomic value rather than a node 
 > was added a
 > couple of drafts or so ago, but some implementations may 
 > still implement
 > early drafts.
 > 
 > David
 > 
 > _____________________________________________________________
___________
 > This e-mail has been scanned for all viruses by Star. The
 > service is powered by MessageLabs. For more information on a 
 > proactive
 > anti-virus service working around the clock, around the globe, visit:
 > http://www.star.net.uk
 > _____________________________________________________________
___________
 > _______________________________________________
 > talk at xquery.com
 > http://xquery.com/mailman/listinfo/talk
 >