W3C home > Mailing lists > Public > xsl-editors@w3.org > January to March 2002

Re: Sort keys and selecting nodes

From: Michael Kay <michael.h.kay@ntlworld.com>
Date: Mon, 25 Mar 2002 08:52:07 -0500 (EST)
To: <xsl-editors@w3.org>
Message-ID: <000001c1d404$4b467170$6501a8c0@pcukmka>
This is a belated reply to Jeni Tennison's message:
http://lists.w3.org/Archives/Public/xsl-editors/2002JanMar/0067.html
which made two important points:

(a) I don't think that it's particularly clear ... what
happens if the select attribute [of xsl:sort] evaluates to a sequence of
more than
one simple typed value or to a node or node sequence. Possibly this is
because the sort key is described as being converted to the target
data type as if by the XPath cast expression, rather than in terms of
a required type and the basic conversion rules.

(b) Another point here is that I think that banning the data-type
attribute from taking QNames other than those naming the primitive
types of XML Schema is a source of backwards incompatibility, but it
isn't listed as such in Appendix J.1.1.

On (a), it's definitely a gap in the spec that we don't way what happens if
the sort key evaluates to a non-singleton sequence. I don't know what we'll
decide on this, but we need to say something. XQuery have decided against
allowing their sort keys to be sequence-valued, we may choose to do the
same.

On (b), I don't think this is a serious incompatibility, because the
semantics of specifying a QName as the data-type were never defined in the
1.0 specification, but I will add it to J.1.1 nevertheless.



Michael Kay
Software AG
home: Michael.H.Kay@ntlworld.com
work: Michael.Kay@softwareag.com
Received on Monday, 25 March 2002 09:14:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:52 GMT