Re: [xsl] getting type information in xslt 2.0

Subject: Re: [xsl] getting type information in xslt 2.0
From: "Dimitre Novatchev" <dnovatchev@xxxxxxxxx>
Date: Wed, 25 Oct 2006 11:51:59 -0700
I understand the examples and the idea.

Anyhow, having to parse for nested sequences seems weird. Also, this
approach sacrifices static type checking, as every function now has
just one argument -- of type item*.

Dimitre Novatchev
Truly great madness cannot be achieved without significant intelligence.
To invent, you need a good imagination and a pile of junk
You've achieved success in your field when you don't know whether what
you're doing is work or play

On 10/25/06, David Carlisle <davidc@xxxxxxxxx> wrote:

> the examples you provided will not run in the case of a function that > has 2 or more arguments that must be sequences.

yes you need to be able to model sequences of sequences to do that. I started to do that (see: <xsl:variable name="q" as="element()"><f:q/></xsl:variable> which then isn't used, but I realised it was midnight and I ought to stop playing and go to bed so posted what I had:-) Note that that is a locally generated element (so with a distinct identity) rather than the global variable with empty elements used elsewhere.

If you model sequences of sequences by using XML element nesting then
you lose type information, another way is to model
as something like

where q' q'' and q''' are three distinct element nodes with name q (each
appearing twice in the sequence, denoting the start and end of a

Given such a structure, with a bit of effort you can handle nested
sequences, as you look at the next item, if it's not the special quote
marker then you use the item, if it is a q then you grab the sequence up
to the next (only) other item that is "is" equal to that item.


Current Thread