Thanks Michael.

I'm trying to do this using Oracle so I haven't got access to Saxon. I'm just
trying to find out if I can call an equivalent Oracle command but Google keeps
pointing me to
http://www.oracle.com/XSL/Transform/java/oracle.xml.parser.v2.Extensions which
doesn't seem to exist any more. I really wish they'd stop messing about with
their site.

-----Original Message-----
From: Michael Kay [mailto:mike@xxxxxxxxxxxx]
Sent: 27 April 2005 11:02
To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
Subject: RE:  Getting useful data from this XML

They call this an XML feed?

I struggled with this at first, until I realized that they have not only
committed the crime of escaping the nested markup so that < is written as
&lt;, they have actually double-escaped it so that it is written as &amp;lt;
This must be the most perverse XML document I have seen for a long time.
Using disable-output-escaping will compensate for a single level of
unnecessary escaping in the input, but it won't compensate for two levels.

If your project allows you to use Saxon, I would suggest

        <p>1) <xsl:value-of select="saxon:parse(concat('&lt;d&gt;',
@description, '&lt;/d&gt;'))"
                              disable-output-escaping="yes"
xmlns:saxon="http://saxon.sf.net/"/></p>

which does the trick. Horrible input sometimes demands a horrible stylesheet.

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