xpointer problem with refentry

[Donna Robinson <donna@terpsichore.ws>]

Hi,

We have various xml files in the build tree, some of which have command-line 
options listed in them.  It would be great if I could create a manpage by 
including the various lists thus:

<refentry id="valgrind">
... ...

<refsect1 id="options">
<title>Options</title>

<xi:include href="hg-manual.xml" xpointer="xpointer(id('hg-cmd-options'))"
            xmlns:xi="http://www.w3.org/2001/XInclude"; />
<xi:include href="cl-manual.xml" xpointer="xpointer(id('cl-cmd-options'))"
            xmlns:xi="http://www.w3.org/2001/XInclude"; />
... ...

</refsect1>

However, I cannot get this to work.  If I generate the refentry as html, then 
everything is just peachy.  But when I do this:
  xsltproc -o ./man/ /path/to/manpages/docbook.xsl man-page.xml
I get this:
  No template matches xi:include in refsect1.

And in the output file:
.SH "OPTIONS"
.PP
This program follows the usual GNU command line syntax, with long options 
starting with two dashes ('\-\-'). A summary of the most important options is 
included below.
<xi:include></xi:include>.SH "SEE ALSO"

I've tried every permutation of xpointer syntax that I can find on the web, 
but have found no solution.  The other syntax that generates html with no 
complaints is:
<xi:include href="cl-manual.xml" xpointer="element(cmd-options)"
            xmlns:xi="http://www.w3.org/2001/XInclude"; />

According to this:
  http://www.sagehill.net/docbookxsl/ModularDoc.html#XincludeSelect

this should work (and it does for html):
<xi:include href="cl-manual.xml" xpointer="cl-cmd-options"
            xmlns:xi="http://www.w3.org/2001/XInclude"; />

but no joy for a manpage even though all *.xml files have this:
<?xml version="1.0"?>
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
  "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd";>

I've also tried this approach:
  http://www.sagehill.net/docbookxsl/RefentryToMan.html
but I don't think I understand enough to make it work.

What's the simplest way around this problem?

Donna

-- 
There is not a shred of evidence to suggest 
that life should be taken seriously