Re: exslt:node-set differs between Xalan-J and XSLTC

On 21.10.2010 16:12, Mukul Gandhi wrote:
> Hi Julian,
>
> On Mon, Oct 18, 2010 at 8:28 PM, Julian Reschke<[email protected]>  wrote:
>>   <xsl:variable name="t">
>>     <root>
>>       <x>1</x>
>>       <x>2</x>
>>       <x>3</x>
>>     </root>
>>   </xsl:variable>
>>
>>   <xsl:variable name="tns" select="exslt:node-set($t)/root/x"/>    [1]
>
>>   <xsl:call-template name="foo">
>>     <xsl:with-param name="t" select="$tns"/>
>>     <xsl:with-param name="i" select="'1'"/>
>>   </xsl:call-template>
>
>> <xsl:template name="foo">
>>   <xsl:param name="t"/>
>>   <xsl:param name="i"/>
>>   <xsl:variable name="tns" select="exslt:node-set($t)"/>       [2]
>
> Since at line [1], you've already constructed a node-set then what's
> the need of doing again node-set($t) at [2] ?

At some point I got an error and thought that passing a nodeset as a 
parameter to a template would turn it into a RTF. (That code was more 
complex, so I'll have to investigate).

> At line [2] I believe, $t is already a node-set and must be accessed
> simply like $t and not node-set($t).
>
> Thought I haven't cross checked what's Xalan's behavior when we do for example:
>
> xx:node-set(xx:node-set($rtf))  which is what I understand you've
> intended to mean with your syntax. And this logically looks wrong to
> me.

My expectation is that applying node-set() to a node set should be 
harmless, but

   <http://www.exslt.org/exsl/functions/node-set/index.html>

indeed doesn't say -- but I'm ready to say it's the only sane thing to do...

Best regards, Julian