[scribus] Index for large book

[John Culleton]

On Monday 04 October 2010 10:27:56 Gregory Pittman wrote:
> On 10/04/2010 04:08 AM, JLuc wrote:
> > Hello
> >
> > Is there any tool to build the index of a 400+ pages book ?
> >
> > Since scribus isnt friendly with documents that have 50+
> > linked-text-frames pages,
> > i plan to spread the book into 12 chapter files,
> > the index will be close to the end of the last file.
>
> I'll beat John Culleton to the punch and point out that TeX has
> this built-in, but Scribus does not.
>
> Thinking about it, though, I wonder if some modification of the
> Find function might accomplish this, by searching for a particular
> Paragraph Style that might be unique various headers, then creating
> a list from that. It's not accessible from Scripter.
>
> There may be some variably complex workarounds too, which could
> benefit from having the original text done in OpenOffice.
>
> Greg
>
> _______________________________________________
> scribus mailing list
> scribus at lists.scribus.net
> http://lists.scribus.net/mailman/listinfo/scribus
I have a few hundred points to make about indexing.

1. Index entries are not always words or phrases taken from text. 
Concepts can be indexed, sometimes in wording not found in that exact 
form in the book itself.

2. Embedded indexing is a great convenience and prevents page number 
errors. However professional indexers usually have a separate program, 
such as Cindex, Macrex or Sky Index to create the index. Often the 
pros will take a paper printout of the book, go to a comfy chair, and 
highlight items to be indexed and make some entries in the margin of 
the page. Then this marked up copy is used with the separate program 
to enter the items and their page numbers.

3. TeX uses embedded markup to generate a file of entries and the page 
number of each. But then a separate program, Makeindex or more rarely 
Xindy, is run to summarize collate an sort all the index entries. The 
output of this separate program is sucked in to the TeX program in a 
future pass.  So the sequence is like this:
 ->Mark up the tex source file foo.tex with indexing tags.
->Run TeX
-> Run makeindex referencing the foo.idx file and creating the foo.ind 
file. 
-> Run TeX again. This time it will find the foo.ind file and print it 
as the index. 
I usually create a little script to do the steps above in sequence.

 3. Although Makeindex is designed to work with the output of a TeX 
run it is fairly trivial to put a graphic front end to the Makeindex 
program and run it independently of TeX. It thus becomes a very 
limited replacement for the programs Cindex etc. but without most of 
the bells and whistles. I did this using the venerable Tcl/Tk 
programming language. Please visit:
http://wexfordpress.com/forindexers.html
for more information. 

4. Here is what I do when I index for money.
a. Accept the customer's final pdf file as input, call it up in e.g. 
Acrobat Reader, and save it as a txt file.
b. Edit the txt file in Gvim, replacing all page break symbols with
the TeX tag: \vfil\eject. This is a single mass change in Gvim.
c. Establish the starting page number in the txt file with the tag
\pageno 6
(or whatever is appropriate.)
d. add the tag \input eplain
e. Insert indexing tags as appropriate.
f. Follow the step-by step listed in 3. above, repeating as necessary.

5. Scribus is a single-pass WYSIWYG system.  If indexing a Scribus-
produced book I would take the pdf output and proceed as in 4. above. 


-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html