RE: [xsl] Trying to understand root-less or document-node-less nodes

["Michael Kay" <mike@xxxxxxxxxxxx>]

> I stumbled across this ('this' is explained later) when I 
> realized that
> 
>    <xsl:variable name="root">
>         <test1 />
>         <test2 />
>     </xsl:variable>
> 
> can be queried by an xpath using the simple $root/* axis. In other
> words: children of $root. And that this felt non-analogous to 
> the following:
> 
>     <xsl:function name="my:rootless">
>         <test1 />
>         <test2 />
>     </xsl:function>

Yes, it's not analogous.

I would recommend always using an "as" attribute with xsl:variable,
especially when you don't use the select attribute. Think of the (above)
syntax without an "as" attribute as a legacy shorthand for

     <xsl:variable name="root" as="document-node()">
       <xsl:document>
         <test1 />
         <test2 />
       </xsl:document>
     </xsl:variable>

If it weren't for backwards compatibility, I'm sure we would have made
xsl:variable behave the same way as xsl:function, that is in this case, to
return a sequence of two parentless elements - which is what you get if you
write <xsl:variable as="element()*">.

Michael Kay
http://www.saxonica.com/