** These are legacy fonts, based on a no-longer-current character set.
** They are not Unicode-compliant. For Unicode-compliant fonts that
** contain all the characters needed to represent Indian-language text,
** consider using the IndUni fonts instead. These are available at
** http://bombay.indology.info.

The principal files in this directory are TeX virtual font files
applying the CSX encoding to various commonly-used fonts: (1) the
standard PostScript fonts Times (ptm*, rptm*), Helvetica (phv*,
rphv*) and Courier (pcr*, rpcr*); (2) the Ghostscript lookalikes for
Palatino (ppl*, rppl*) and New Century Schoolbook (pnc*, rpnc*); and
(3) the standard Computer Modern text fonts (cmr_csx*, cmti_csx*,
cmsl_csx*, cmbx_csx*, cmbxti_csx*, cmbxsl_csx*). (The Computer
Modern fonts do not have the less, greater, cent, sterling or yen
characters, or the two "guillemots" [recte guillemets].) If you
install these fonts, you will be able to print CSX text via TeX
without need for any conversion. The fonts have been generated by
means of a program (vpl2vpl) which can be used to produce
arbitrarily accented TeX virtual fonts for any PostScript or
Metafont font; the program itself may be found at
http://bombay.indology.info/software/programs/accfonts/, together
with the definition file CSX.def used to create the CSX fonts.

Unlike the previous versions of these fonts (created with the now
obsolete program csxvpl), the present versions are non-recursive and
can therefore be renamed by the user. The only name that must not be
used is that of the underlying "real" font (e.g. rptmr in the case
of ptmr, cmr10 in the case of cmr_csx10).

It is possible to use CSX on Unix workstations:
http://bombay.indology.info/software/fonts/csx_unix/ contains some
information, fonts and configuration files to assist with this.

A PostScript version of a document by Dominik Wujastyk giving basic
information about CSX is included as CSX.ps. In addition to the
character encodings listed there, my fonts include three extra
characters:

a macron tilde is located at decimal 171 = octal 253 = hex AB;
i macron tilde is located at decimal 172 = octal 254 = hex AC;
u macron tilde is located at decimal 216 = octal 330 = hex D8.

The file genmacs.tex is my "general macros" file. It is not necessary
to place this entire file at the head of any document using the CSX
fonts, but it *is* necessary for the initial section setting up
\lccode and \uccode to be somewhere where TeX can see it, and you
should also include the sections defining \l and \L. If you use
dvips, and have therefore redefined \H and \., you should remove
those redefinitions as the accents in question are now back in their
proper positions. (If you do use my genmacs.tex file, you will find
that \. now produces a subscript dot, \: a superscript dot: I find
this more natural, but no one is under any compulsion to agree with
me.)

Certain characters, to which TeX assigns one position while CSX
assigns another, appear twice --- thus the German es-zet appears both
at decimal 25 (= octal 31 = hex 19) and at decimal 225 (= octal 341 =
hex E1).

These files are made available on the assumption that anyone who
wishes to use them will already have a working knowledge of TeX. If
you encounter problems with the files, please let me know. If you
encounter problems in getting the files to work on your TeX system,
please consult your local expert.

In addition to the virtual fonts and supporting files for CSX, this
directory also contains a file csx.def, provided by Juergen Hanneder
<hanneder@muenchen.org>. This allows the CSX encoding to be used in
LaTeX in conjunction with the standard Computer Modern fonts.
Juergen writes as follows:

> csx.def provides a way to use the Classical Sanskrit Extended input encoding 
> with standard LaTeX and with the standard fonts supplied with TeX. 
> It is similar to the .tcp files in emTeX, with the exception that the 
> information is not kept in the format file, but can be loaded in the preamble
> of a document. This is done by:
>
>       \usepackage[csx]{inputenc}
>
> The file includes only the most common Sanskrit accents, not the whole
> CSX, and is done rather pragmatically. I hope to complete it as soon
> as I find time. If anyone feels prompted to do the rest, please send
> me a copy.

John D. Smith
http://bombay.oriental.cam.ac.uk/index.html
