Re: [xsl] Adding namespaces to output

[Kevin Jones <kjones@xxxxxxxxxxx>]

On Monday 23 Dec 2002 7:18 pm, G. Ken Holman wrote:
> Close, but I can think of two problems with this, Kevin,
>

Cut..

> >You can get the same effect by using a temporary tree and then copying the
> >namespace out of it.
> >
> ><xsl:variable name="dummy">
> >         <xsl:element name="p:a" namespace="something"/>
> ></xsl:variable>
> >
> ><a>
> >         <xsl:copy-of select="$dummy/*/namespace::*[1]"/>
> ></a>
> >
> >You will need to use your processors nodeset function to access $dummy in
> > the copy-of.
>
> (1) if the user is using W3C standard XSLT 1.0 without extensions, a result
> tree fragment cannot be supplied to xsl:copy-of except by variable name
> only.

Yep. Thats why I suggested he use nodeset on the copy-of. For interest are 
there any prcoessors that do not have support a nodeset function yet?

>
> (2) XSLT 1.1 allowed this, but since the order of namespace nodes is
> arbitrary and possibly different with all processors, and there may be any
> number of ancestral namespace declarations, the namespace for 'p:' isn't
> necessarily the first and the following would not be subject to the order:
>
>     <xsl:copy-of select="$dummy/*/namespace::*[name(.)='p']"/>

I am exploiting the fact that I know exactly what the RTF looks like because I 
created it in the variable. Therefore I know the namspace node is the only 
one and its at position one of the first element. I don't think there is any 
way this can be wrong in a 1.0 conforming processor. Having said that your 
solution will clearly be more resiliant to anybody hacking the variable 
around.

Regards,
Kev.



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