RE: [xml-dev] Question about XPath 2.0

Thanks, I think we need arithmetic expressions etc: for example,

book[..]/@price + book[..]/@price

it is interesting as to see what kinds of expressions are needed for
elements, things such as variable bindings etc seemed slightly on the
higher (rather not needed) side, I am just saying based on my intuitions,
I am sure there are good reasons as you point out... thanks and regards -
murali.

On Tue, 20 May 2003, Michael Kay wrote:

> XPath is an expression language over the data model. The data model is
> based on sequences. The core operator on sequences is a mapping operator
> that applies a function to each item in a sequence to produce a new
> sequence. This mapping operator is implemented in XPath in the form of
> the "for" expression. Without the ability to map sequences, XPath would
> be very limited in its ability to manipulate values in the data model.
>
> I'm not convinced that we have actually got the For expression right -
> it's designed as a subset of the XQuery FLWR expression, whereas
> arguably a mapping operator designed specifically for XPath would look
> rather different. But the functionality is definitely needed.
>
> >
> > My reasoning: I was under the impression that XPath will
> > provide expressions that will allow us to get nodes and
> > values from XML documents, I thought this will be serve as
> > the basis for most operations on XML. I felt XQuery/XSLT etc
> > can sit on top of XPath.
> >
>
> Not all XPath users will be using it within XSLT or XQuery, and in any
> case, in you don't want to have to keep calling back from XPath
> expressions to execute XSLT functions.
>
> Michael Kay
>