RE: XPath 2.0 little question on draft (`for' statement)

I think you are correct. (The two expressions are probably not equivalent
for all possible source documents, because one forces document order where
the other does not; but they are equivalent for cases where books do not
have books as descendants.)

I suspect that the reason the example was written that way is that the
author comes from a SQL background. Some people prefer to spell everything
out in "for" expressions rather than writing compact path expressions. It's
largely a matter of personal taste.

Michael Kay

> -----Original Message-----
> From: Bernard van Gastel [mailto:bernardg@sci.kun.nl]
> Sent: 19 November 2003 15:56
> To: public-qt-comments@w3.org
> Cc: is1@bitpowder.com
> Subject: XPath 2.0 little question on draft (`for' statement)
> 
> 
> 
> Hello XPath 2.0 Working Group,
> 
> In Section 3.7 of the XPath 2.0 draft of 22 August 2003 the
> following example is included.
> 
> 	for $a in distinct-values(//author)
> 	return ($a,
>         	for $b in //book[author = $a]
>         	return $b/title)
> 
> I think it can be replaced by the following sniplet.
> 
> 	for $a in distinct-values(//author)
> 	return ($a, //book[author = $a]/title)
> 
> Is this correct? Why is chosen for the first version (if the
> second version is correct) instead of the second? Thank you 
> in advance,
> 
> 	Bernard van Gastel
> 	- Student Computer Science,
> 		University of Nijmegen,
> 		The Netherlands
>