[XSL-LIST Mailing List Archive Home]
[xsl] The hard cocktail of sequence and (node-)set (Was: RE: RE: Postional predicates de-mystified)
Subject: [xsl] The hard cocktail of sequence and (node-)set (Was: RE: RE: Postional predicates de-mystified)|
From: Dimitre Novatchev <dnovatchev@xxxxxxxxx>
Date: Fri, 4 Jan 2002 08:35:41 -0800 (PST)
> > Curious about the shift from node-set to sequence then.
> > Was it because of all the discussion (confusion?) on the list
> > re sets?
> I think it started with the fact that XML Schema supports lists as a data
> type, for example IDREFS is a list of strings. XQuery wanted to support
> lists of nodes (without them, what is the result of a sort operation?), and
> so came the idea of a data model based on sequences. There was then
> considerable debate about whether the model should support sets and
> sequences as separate types, with an eventual consensus that operations that
> naturally produced sets (like path and union expressions) would be defined
> as producing a sequence in canonical order, as this kept the model simpler.
> The confusion about XPath 1.0 node-sets may have had an influence. It wasn't
> confined to this list either - I've heard people within W3C, people with a
> mathematical training who should know better - claim that XPath 1.0
> node-sets were really sequences all the time.
> Mike Kay
If not corrected this "feature" alone will become a classical example of bad design.
I have pointed out elsewhere some of the unfortunate results of mixing these two
very different datatypes into one:
Do You Yahoo!?
Send your FREE holiday greetings online!
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list