Wow,

Thanks for the swift, comprehensive reply.

I tried your javascript trick (?) and I got:

----
<?xml-stylesheet version="1.0" type="text/xsl" href="page.xslt"?>

<result />
----

However I don't get:

<?xml version='1.0' encoding="utf-8" ?>

I assume that's FireFox 'trying to render the xml' :)

I'm downloading saxon now...

Having solved that mystery I now find myself with another one which 
perhaps would be better as a separate thread. I'll mention it here just 
in case it's another obvious case of ignorance..:

I'd hoped that the output of the transform would also be transformed 
into html by FireFox (which is the purpose of page.xslt):

------------------
<?xml version='1.0' encoding="utf-8" ?>
<?xml-stylesheet version="1.0" type="text/xsl" href="page.xslt"?>

<result />
------------------

Can browsers perform this two stage transform?

Thanks again for your time,

Martin

Martin Honnen wrote:
> Martin Welch wrote:
> 
>> Now I don't know if this is significant but I'm using firefox 2.x on 
>> WinXP to open pi.xml and looking at the results using WebDeveloper 
>> 1.1.4 which gives me an option to view generated source.
>>
>> I've searched the newsgroups for two days now with no result.
>>
>> Could anyone give me a clue where I might be going wrong? Can anyone 
>> suggest a better (free) tool to help 'debug' my feeble attempts?
> 
> I have tried your sample with Firefox 2.0, to check the result document 
> I have typed
> javascript: alert(new XMLSerializer().serializeToString(document)); void 0
> in the location bar (as one line) and hit enter.
> Rest assured that the processing instruction is part of the result 
> document so the XSLT transformation is working fine in Firefox.
> You can also simply use Tools -> DOM Inspector to see the nodes in the 
> result document, it also shows the processing instruction node.
> 
> No idea about WebDeveloper, check whether they have a mailing list or 
> forum to ask why it does not show that processing instruction.
> 
> And generally, unless you are specifically developing for Firefox, I 
> think learning XSLT and transforming XML to XML is better done with a 
> command line XSLT processor such as (for instance) Saxon 6 
> <URL:http://saxon.sourceforge.net/saxon6.5.5/index.html> as that way you 
> see the result document in the console window or in a file editor. A 
> browser tries to render the XML and that way you see only the text node 
> content and not the XML.
>