[scribus-dev] Design for indexing feature.

[Andreas Vox]

> -----Original Message-----
> From: ale rimoldi [mailto:ale.comp_06 at xox.ch]
> Sent: Donnerstag, 3. Juli 2014 11:35
> To: scribus-dev at lists.scribus.net
> Subject: Re: [scribus-dev] Design for indexing feature.
> 
> hi john
> 
> > A spec was requested. I provided my ideas of how to provide a new
> > indexing system with minimum effort and maximum accuracy using the TeX
> > program makeindex as the processing engine. My paper is published
> > here:
> > http://wexfordpress.com/phase2.pdf
> >
> > I would appreciate it if the developers would read it and tell me if
> > it is usable as a spec for the purpose intended. I don't program in
> > C++ and in any case I don't have a roadmap to the Scribus program
> > code. So this is the best I can do.
> 
> thanks for your proposal!
> 
> if i understand you correctly, you're suggesting that we should use the
> tex/latex indexing system from scribus.
> 
> personally, i'd prefer a solution that -- while possibly using external
libraries --
> does not pull in a huge package like latex...
> but for the time being, going through latex could also be a way of doing
(in
> the future we could have a simpler internal index or an external one
through
> latex)

You can use makeindex without LaTeX. It just converts a raw index file to a
formatted index.
Using a style file it's possible to change the input and output formats.

I'd prefer to use its successor Xindy, though. Xindy supports international
languages (incl. Klingon)
and is more flexible.

The interesting part is how to get the raw index from Scribus. We could use
something like characters styles
or we could use escape chars like {thingy} and {dog|pets|animals} or
>thingy< and  >dog|pets|animals<.
It might also be interesting to combine it with a word frequency counting
tool for identifying possible index terms and uses.

Reading the formatted index shouldn't be more difficult than importing text.

/Andreas