[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
Re: [xml-dev] What class of grammars is an =?UTF-8?Q?XPath=3F?=
- From: rjelliffe <rjelliffe@allette.com.au>
- To: <xml-dev@lists.xml.org>
- Date: Fri, 28 Jan 2011 04:28:53 +1100
On Thu, 27 Jan 2011 14:19:41 +0000, Michael Kay <mike@saxonica.com>
wrote:
> Ah, you have interpreted the question quite differently from me (and
> perhaps correctly?) I thought it was looking at XPath as a grammar in
> which the sentences being described are document instances.
Yes. I did not make myself very clear.I blame the humidity. The syntax
of XPath is not my interest.
XPaths could be considered kinds of catapillar expressions, or indexed
grammars, or branching grammars. And probably some combination of them:
indexed catapillar pebble grammars? I am interested in whether anyone
has done any formal work recently on this. (I suppose if it does not
have an adequate formal characterization at the current state of the
art, it is in the position of SGML/DTDs in the 80s and 90s, waiting for
the right economic moment to become a carrot attractive to academia.)
That the language may be open is no problem: if the XPath is "/x"
then it will include simple <x/> or <x><a/></x> or anything with
top-level element x.
The particular expression that Michael brought up --
sum(//order/@value) > 826 -- certainly implies an open language (the
document must have at least one //order/@value). There is no necessary
difference between the grammar for data content and for elements, so
actually the language would consist of all documents that have at least
one //order/@value where all //order/@values contain numbers.
It is possible to
Cheers
Rick
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]