Re: Getting useful data from this XML

Normally people have
<foo> &lt;p&gt;... &lt;/p&gt; </foo>

and we tell them life would be better if they could instead have

<foo> <p>... </p> </foo>


but if they really can't change the input format d-o-e will get them out
of a hole.

But you have

<foo> &amp;lt;p&amp;gt;... &amp;lt;/p&amp;gt; </foo>

which means that it is _double_ escaped ie someone dug a big hole and
then put a trap at the bottom of it before throwing you in.


You could look to using an extension element (eg if you have saxon:parse
available you could parse it _twice_ to get back to having elements
there) but I would not do this using xslt I would pass the input through
sed or perl to normalise the input first, before using xslt on it.

David


________________________________________________________________________
This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________