Hi,
I am sorry for responding after such a long time, I was too busy with my
work.
First of all, xindy is a replacement for MakeIndex, it offers features not
present in MakeIndex. However, it has its own syntax for input files. The
input has to be preprocessed first. For this purpose texindy is used, it
automatically loads necesary modules to process the latex idx files by
xindy. MakeIndex was able to sort in English and German only, xindy allows
to define sort rules for any alphabet or any language.
Since I use TeX Live on Linux, I do not know whether xindy comes with MiKTeX
or whether it is necessary to install it manually. If you have xindy, you
have to install the Hindi + Marathi module. You just unzip
install-xindy-devanagari.zip to a proper directory. It will probably be
texmf\xindy or texmf\xindy\modules or something like that. As written on my
web page, the xindy directory structure has changed durng years sou you have
to find xindy directory containing subdirectories base, lang, and tex. It
might be necessary to refresh FNDB from the MiKTeX menu. At least it makes
no harm if you do it.
Now you process the document containing \index{...} macros which will
generate the .idx file. Then you process the .idx file by texindy.
1. If you use xelatex and type the text directly in UTF-8, it is the simpler
way. Years ago Omega TeX was invented, it works directly in Unicode. XeTeX
works in unicode too, so you specify that the input markup is omega. You
have to specify the language and the name of the input file. If you sort in
Hindi, the command line may look as:
texindy -I omega -L hindi filename.idx
The output will be sent to filename.ind which will be read by the next
xelatex run. You may load other modules with the -M switch. You will
probably like -M page-ranges so that the index displays 1--5 instead of 1,
2, 3, 4, 5. There are two ways or writing characters with nuktas. for
instance, za can be written directly since it has its codepoint in unicode,
or as ja followed by nukta combining character. If you type it as ja+nukta,
it will not be sorted properly. for this purpose you can load -M dvngnukta
which will normalize these characters. you can load several modules, so the
command line may look:
texindy -I omega -L hindi -M page-ranges -M dvngnukta filename.idx
You will do the same for marathi, only dvngnukta makes no sense because
AFAIK characters with nuktas are not use in Marathi.
2. If you use Velthuis Devanagari, the command line will be more complex.
The input markup will be latex and the encoding wi be velthuis, ie you need
options -I latex -C velthuis. You cannot write the .ind file directly, you
have to generate an intermediate file and run it through the devnag
preprocessor. You therefore specify the output file name with -o
filename.inx and after running texindy you run (both file names are
mandatory)
devnag filename.inx filename.ind
Of course, it may only work if the preprocessor instractions are added to
the filename.inx. This is acheved by specifying a proper module with the -M
switch. Four modules are available, namely:
velthuis-dn-latex-lhindi
velthuis-dn-latex-modernhindi
velthuis-dnnum-latex-lhindi
velthuis-dnnum-latex-modernhindi
The differences are subtle. As you can judge from the names, they insert
either @hindi or @modernhindi. the modules containing dnnum in their names
add also \dnnum, thus you will see devanagari numerals in the output. Of
course, you may like to use -M page-ranges too. Thus the command line may
look as:
texindy -I latex -C velthuis -L hindi -M velthuis-dnnum-latex-modernhindi -M
page-ranges -o filename.inx filename.idx
3. If you use XeLaTeX but enter the text in the Velthuis transliteration
using \fontspec[Script=Devanagari,Mapping=Velthuis]{Snaskrit 2003}, you have
a big problem. This does not work so far, I will support it when I have
time.
The title of the index is taken from the \indexname macro. This is defined
for Hindi in Polyglossia as well as in babel. Unfortunatelly, the babel
module is not included automatically, the manual for the velthuis package
explains how to add it manually. There is no Marathi support in Babel and
Polyglossia. I do not know Marathi, so I am not able to translate the
captions. I can try to find Marathi synsets on
http://www.cfilt.iitb.ac.in/wordnet/webhwn/wn.php but I do not like to o it
without knowing even the basics of the language. if you can supply the
captions, I will write the Marathi modules quickly.
Regards,
2011/7/11 Rahul Athale <[email protected]>
>
> Hi,
>
> I just came across Xindy modules for Hindi and Marathi. I have not used
> Xindy earlier. I am using MiKTeX 2.8 on Windows Vista. Can you please give
> some guidelines on how to use the modules created for Hindi and Marathi?
>
> Thank you so much.
>
> Best Regards,
> Rahul.
>
>
> <http://sigads.rediff.com/RealMedia/ads/click_nx.ads/www.rediffmail.com/signatureline.htm@Middle?>
> Treat yourself at a restaurant, spa, resort and much more with *Rediff
> Deal ho
> jaye!<http://track.rediff.com/click?url=___http://dealhojaye.rediff.com?sc_cid=mailsignature___&cmp=signature&lnk=rediffmailsignature&newservice=deals>
> *
> _______________________________________________
> Devnag-general mailing list
> [email protected]
> http://lists.sarovar.org/cgi-bin/mailman/listinfo/devnag-general
>
--
Zdeněk Wagner
http://hroch486.icpf.cas.cz/wagner/
http://icebearsoft.euweb.cz
_______________________________________________
Devnag-general mailing list
[email protected]
http://lists.sarovar.org/cgi-bin/mailman/listinfo/devnag-general
