Graham Hannington wrote:
(That is, the index terms in the <meta> tag are comma separated, and ":" is
used to indicate a primary/secondary index entry. Nothing fancier than
this: I don't currently do "tertiary" - third-level - index entries, or even
- though perhaps I should - "see" or "see also" entries.)
I'd be very grateful for any tips or pointers on where to start.
Having adapted the Docbook indexing code myself I can tell you that it
will probably be a challenge--it does some very subtle things that have
taken me a good bit of puzzling to get a basic understanding of,
although if wasn't hard to adapt the code to my own needs *as long as*
the index entry markup was similar to Docbook's in structure (which it
was in my case). The code uses XSLT keys in clever and non-obvious ways
to enable the collection, sorting, and processing of index groups. I
would say more about it here except that I'm in the process of
developing a training module on exactly this topic, so I have to focus
on that for now. But I will say that you can puzzle it out if you read
the stuff in Dave P's XSLT FAQ on the Munchian method and then work
carefully through the code (it will also help to expand the internal
parsed entities that are used to hold the expressions used for the keys).
The challenge in your case will be that the meta tags are not at all
like the structure of Docbook index entries.
One thing that might make your task easier is to do a two-step process:
1. Write a separate process that turns the meta tags into Docbook-style
index entries (this would be facilitated by using the EXLST functions
for string tokenization).
2. Process those new entries with the existing Docbook stuff.
Hope this helps.
Cheers,
E.
--
W. Eliot Kimber, eliot@xxxxxxxxxx
Consultant, ISOGEN International
1016 La Posada Dr., Suite 240
Austin, TX 78752 Phone: 512.656.4139
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list