This is the mail archive of the
docbook-apps@lists.oasis-open.org
mailing list .
Re: [docbook-apps] Docbook XSL and RELAX NG documents
- From: Matthew Burgess <matthew at linuxfromscratch dot org>
- To: Mauritz Jeanson <mj at johanneberg dot com>
- Cc: davep at dpawson dot co dot uk,'Docbook-Apps' <docbook-apps at lists dot oasis-open dot org>
- Date: Thu, 16 Dec 2004 22:24:10 +0000
- Subject: Re: [docbook-apps] Docbook XSL and RELAX NG documents
- References: <200412162023.iBGKNskG236713@mailhost.bostream.com>
Mauritz Jeanson wrote:
-----Original Message-----
From: Matthew Burgess [mailto:matthew@linuxfromscratch.org]
Sent: den 16 december 2004 21:19
To: Mauritz Jeanson
Cc: davep@dpawson.co.uk; 'Docbook-Apps'
Subject: Re: [docbook-apps] Docbook XSL and RELAX NG documents
Mauritz Jeanson wrote:
If you change to a "null namespace" in the template, then
things work
better. That is, change
<xsl:element name="{local-name(.)}"
namespace="http://www.w3.org/1999/xhtml">
to
<xsl:element name="{local-name(.)}" namespace="">
Thanks for the suggestion, Mauritz. Unfortunately that
doesn't appear
to have any effect whatsoever here :(
Hmm... This might have something to do with the processor. It works with
Saxon, but I vaguely recall some issues with the implementation of
node-set() in xsltproc.
Many thanks! Indeed, saxon appears to get this right, where both
xsltproc and xalan fail.
Sorry to pester you further, but I'm now getting a bunch of warnings:
XRef to nonexistent id: ch-scripts-console
Error: no ID for constraint linkend: ch-scatter-maillists.
XRef to nonexistent id: ch-scatter-maillists
I've a feeling this too is namespace related, as the "id" attribute is
now namespace qualified, i.e. it's "xml:id". Any ideas on solving this
one? I took a quick look at xhtml/xref.xsl but couldn't see an
immediate way of fixing/hacking it.
I realise I may just be a victim of my own "jumping ahead of the gun"
here, but I was simply after an idea of how the DocBook processing
toolchains hold up with RELAX NG documents. Obviously they don't seem
to fair too well at the moment (just an observation, not a complaint!).
Thanks again,
Matt.