Gloss, (yet) another way of making glossaries
=============================================
(c) 1998 Jose Luis Diaz


Gloss is a package which allows the creation of glossaries using BibTeX.

Several approaches have been tried for automating the creation of
glossaries. Some of these approaches developped its own tool for the
sorting and formatting of the glossary (glotex, glosstex), some other
tried to use an existing tool (typically Makeindex).

This new approach uses BibTeX for the sorting and formatting of the
glossary. That is, the user writes a database of terms and definitions
using a file format much like the bibliographic databases. Then he
inserts in the latex document a command (\gloss{key}) for selecting
which entries he wants to appear in the glossary. These keys are
written in an auxiliar file when latex is run on the document. Then,
running bibtex these entries are extracted from the database and
collected in alphabetical order in a file. The next run of latex read
this file and inserts it in the appropiate point of the document,
typesetting the desired glossary. 

The process is much like the mechanism for including the bibliographic
references. This approach has several advantages but also some
disadvantages. The obvious advantages are:

 * There are a lot of tools for managing bibtex databases.
 * It is possible to write different .bst (bibliographic styles) for
generating differents styles of glossaries.
 * There exists a version of BibTeX (called bibtex-8) which is able to
manage alphabets with 8-bit encoding. This is the main raison for
which I developped the package, since the other approaches I tried did
not manage properly the latin1 encoding.

The main disadvantage is that BibTeX assumes some extensions for the
file names, and don't allows to specify different ones. For example,
the file with the citations has to be named '*.aux', the file with the
database has to be named '*.bib' and the file which itself writes
always is named '*.bbl'. In order to use in the same document a
glossary and a bibliography, the solution adopted for Gloss was to
generate the glossary citations in a file named '*.glo.aux', store the
definitions in a database named '*.gdf.bib' and use for the formatted
glossary file the name '*.glo.bbl'. The use of the 'double extension'
in the file names disallows the use of this package to MS-DOS users.

INSTALLATION:
------------

Run:

 latex gloss.ins

And you'll obtain the files:

  gloss.sty        (the package)
  gloss.bst        (the "bibliographic" style which allows BibTeX to
                   create the glossary. Thanks to Oren Patashnik)
  sample.tex       (a sample document)
  sample.gdf.bib   (a glossary database example)


You have to copy the .sty file to a directory where LaTeX can find it,
and the .bst to a directory where bibtex can find it.

DOCUMENTATION AND EXAMPLES:
--------------------------
Run:

  latex gloss.dtx 

(twice) for obtaining the documentation.

The documentation and the samples are written in spanish. The only
english documentation at the moment is the 00readme file that you are
reading. In addition, the file gloss.dtx uses the latin1 encoding, and
this will cause trouble in the generation of the sample files on most
of TeX implementations, because TeX tends to write the 8-bit characters
in "hexadecimal form", that is, for example, instead of "ñ" it
outputs "^^f1". Since the sample files are created by TeX (when you
run "latex gloss.ins"), they will be full of these funny sequences. If
you want to "recode" these files for obtaining "ñ" instead of "^^f1",
you can use the utility texto8.awk distributed with this package.
The use is (assuming a unix-like operating system):

$ gawk -f texto8.awk < sample.tex > s.tex
$ gawk -f texto8.awk < sample.gdf.bib > s.gdf.bib
$ mv s.tex sample.tex
$ mv s.gdf.bib sample.gdf.bib

(Obviously, you'll need the GNU tool "gawk")

Then, for typesetting the sample, you can type:

$ latex sample
$ bibtex8 -c 88591lat sample.glo
$ latex sample
$ latex sample

If you don't have bibtex8 you can use bibtex, but then the glossary
entries will not be correctly ordered (because the sample files use
8-bit encoding).

This software is still in beta-testing phase. If you find a bug or
want to make any comment, you're welcome at:

jdiaz@etsiig.uniovi.es

Enjoy it!


PD: Sorry for my english ;)