Table of contents.

["Dave Pawson" <dave.pawson@gmail.com>]

I've just had a complaint about my odf to daisy conversion.
The source of the problem is the table of contents.

In our annex we have,
 1 Background and Overview	4
 1.1 Introduction	4
 1.2 What is Accessibility?	4
 1.2.1 Types/Categories of Access	4
 1.2.2 Types of Disabilities, and How they are Addressed	5
 1.3 Importance of the Accessibility of the ODF Application	9



at the xml level this is, quite frankly, a mess.
            <text:index-body>
               <text:p text:style-name="P12">
                  <text:a xlink:type="simple" xlink:href="#1
Background and Overview|outline">
                     <text:s/>1 Background and Overview<text:tab/>4</text:a>
               </text:p>
               <text:p text:style-name="P13">
                  <text:a xlink:type="simple" xlink:href="#1.1
Introduction|outline">
                     <text:s/>1.1 Introduction<text:tab/>4</text:a>
               </text:p>
               <text:p text:style-name="P13">
                  <text:a xlink:type="simple" xlink:href="#1.2 What is
Accessibility?|outline">
                     <text:s/>1.2 What is Accessibility?<text:tab/>4</text:a>
               </text:p>

1. It's an index (I'm guessing we formatted it as an index, rather
than as a toc?)
2. There is a mismatch between the xlink:href attribute content and
the 'hot' text (see the 1.1 entry above.
3. This is a manual insert. If I ask Open Office to 'update all',
indicating I want a new
toc generating, the toc is really quite different.

End game, for XML usage of ODF. For me.

1. To be able to generate a table of contents using valid ID values
for the cross references.
e.g. no white space in the value.
2. To be able to 'easily' locate the items which have been indicated
as belonging in the toc. This is less easy. Easier if automatically
done.

I think this is a main TC list issue, but the one that bothers me is
the ease with which
we (as users) managed to mess this up?

Before I go to the TC with this, are there any comments from our group?
 E.g. suggested steer, options, other implementation views etc.

What is reasonable to expect of a user, what must the implementors and
users of XML do?
Where should the pain be, how best to share it?



regards








-- 
Dave Pawson
XSLT XSL-FO FAQ.
http://www.dpawson.co.uk