Re: structure of extracted index page

[Dan Mueth <dan eazel com>]

On 26 Apr 2001, Gregory Leblanc wrote:

> > Mary is creating a new DTD to describe indexing information which is used
> > internally to ScrollKeeper, and possibly to a help browser.  It is not
> > meant to be a part of a document, so it isn't necessary or even
> > appropriate to use DocBook.
> 
> In what way is the DocBook markup incomplete for this.  Put another way,
> why do we need another DTD?

DocBook allows you to specify two types of indexing information:
 (1) indexterm's which are used to generate an index
 (2) indexentry's which are used to specify formatted index entries
     in the index

It does not describe the intermediate information one gets when one reads
in all the indexterms and tries to generate an index from it.  

You could almost imagine using the indexterms to generate a bunch of
indexentry's and then using a DTD which resembles the index and indexentry
part of DocBook.  I don't know if indexentry's will allow you to specify
the links into the documents and between various indexentry's the way we
need to.  Even if they do, we would be specifying a new DTD which
resembles part of DocBook.  We would want to throw out as much of DocBook
(extra tags, attributes, children we don't need) as possible so that
we keep things as simple as possible.  Remember - we will not be
converting this index to HTML.  We will be using it directly in
applications.  For example, your index may become a tree widget or a
searchable data structure.

I posted a bit of XML in response to Mary's post onto the
scrollkeeper-devel list and not on gnome-doc-list to keep down the noise
here.  But since people on this list are interested, you can see a bit
more here:

http://www.geocrawler.com/lists/3/SourceForge/6429/0/5630864/

Dan