Re: Implementing " and ' in literals

> I want to instigate change (for XPath 2.0,
> for what it's worth)

Ah  but that isn't what you said in your message, you said you'd just
added it as a feature to a (presumably xslt 1.0) implementation :-)

> Should I try and stump up the thousands of dollars needed to join
> the w3c,

No, just plug away making suggestions on this list, it is I know
monitored by the Working Group. I went from `complaining on public
www-math list' to `invited member of W3C Math WG' by some indeterminate
process:-)

> There aren't _any_ conforming implementations of XPath 

But there is a complete difference between unintentional bugs
and personal interpretations on underspecified grey areas,
and an intentional incompatibility in a perfectly well specified area.

> It's a shame XPath has no namespace mechanism to
> make adding grammar extensions simple.

You can add extension functions from a different namespace, can't you?
Which isn't strictly a change of grammar, but would allow

fastnet:quoted-string(.......)

or whatever you want the syntax to be.

Or you have an xpath (and xpointer and xslt) 2.0 with a different
string literal syntax which includes quoting.

Given you only need this to get a string literal that has both " and '
and in xslt it's easy to get such a string by
<xsl:variable name="x">this ' and this "</xsl:variable>
and then coercing the rtf variable to a string.

It seems not worth being incompatible just for this, but if anyway
an xpath 2.0 was coming out, it could I suppose be something to be
considered. (I should stress I'm not on the xsl group, so whether or not
I think its a good idea is immaterial)

David


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list