[
Lists Home |
Date Index |
Thread Index
]
On Fri, 30 Aug 2002, Dare Obasanjo wrote:
> XQuery was originally based on XDuce's tree grammar system but has since
> evolved. XDuce was a poor basis for an XML type system considering that
> it didn't handle attributes or namespaces[0] nor does it's successor
> Xtatic based on the presentation I saw at Microsoft Research a few
> months ago.
>
> I suggest reading the recent crop of XQuery related specs released about
> two weeks to bring yourself up to speed. The formal semantics is
> slightly behind but given conversations I had with our standards reps a
> few minutes ago plus my impressions from talking to other working group
> members a few weeks ago, it will lean more towards named typing than
> tree grammars (which are complex to implement and also raise concerns
> about scalability).
>
> XQuery is NOT based on regular tree grammars.
I think you are not clear about regular tree grammars vs named typing.
There are different things --
XDuce is based on static subsumption for subtyping, as opposed to named
subtyping. XQuery has decided to support named subtyping. But named
subtyping and regular tree grammars are orthogonal, and do not affect each
other, form what I have seen. Your distinction that named typing implies
no regular tree grammars is not correct.
This support for named subtyping does not affect in any way the fact that
the type inferencing for XQuery is based on regular tree grammars. Type
inferencing, regular tree grammars and closure under operations forms the
XQuery core, and they go together.
I will check the following link out, but again, I will suggest this --
conclusion that named subtyping is not regular tree grammars is incorrect.
We have been developing constraint specification and subtyping frameworks
on top of RELAX NG, we will update the list soon, when we are satisfied.
> [0] http://xduce.sourceforge.net/papers.html
<warning>speaking for himself only</warning>
cheers and regards - murali.
>
> --
> PITHY WORDS OF WISDOM
> Old cars never die, they just smell that way.
>
> This posting is provided "AS IS" with no warranties, and confers no
> rights.
>
> >
> >
> > -----Original Message-----
> > From: Murali Mani [mailto:mani@CS.UCLA.EDU]
> > Sent: Friday, August 30, 2002 12:58 PM
> > To: Dare Obasanjo
> > Cc: Maciejewski, Thomas; xml-dev@lists.xml.org
> >
> >
> > On Fri, 30 Aug 2002, Dare Obasanjo wrote:
> >
> > > Two points
> > > 1.) XQuery's type system used to be based on regular tree grammars
> > > but this is no longer the case working group since they adopted W3C
> > > XML Schema as the basis of its type system. Now XQuery's
> > data model is
> > > primarily based on named typing where the named types are W3C XML
> > > Schema types.
> >
> > Their processing is based fully on regular tree grammars, the
> > restrictions such as 1-unambiguity and single type tree
> > grammar has no relationship to XQuery what so ever, to the
> > best of my knowledge.
> >
>
|
