This is a package for hyphenating transliterated Sanskrit texts in LaTeX.
I wrote it because there is no support for Sanskrit in babel, and because polyglossia, which does include Sanskrit hyphenation patterns, cannot currently hyphenate Sanskrit texts in LuaTeX, for technical reasons. Compared to the polyglossia Sanskrit patterns---which were originally written by Yves Codet---, mine introduce two novelties.
Firstly, they work correctly with all precomposed characters. By contrast, some precomposed characters---liquids, for instance---are missing from the polyglossia patterns, such that kṛtvam with 'ṛ' = U+1E5B (LATIN SMALL LETTER R WITH DOT BELOW) is not hyphenated as kṛ-tvam, as it should be, but rather as kṛtvam, without any hyphen. I made sure that my patterns work with both composed and decomposed characters. However, they are designed to deal correctly only with text that is encoded in one of the four Unicode normalization forms. Mixing several normalization forms might yield incorrect results. Since (correct) texts are supposed to be encoded in a single given normalization form, this shouldn't be an issue.
Furthermore, the hyphenation policy is different. Patterns in polyglossia systematically keep together clusters of consonants, such that vikṣiptam is hyphenated as vi-kṣi-ptam. By contrast, I allow breaks inside runs of consonants in certain circumstances. The term vikṣiptam, for instance, is hyphenated as vi-kṣip-tam. To determine acceptable break points inside consonants clusters, I used Elisa Freschi's heuristic, which is to treat as syllable starters runs of consonants followed by a vowel that can occur at the beginning of a word. I used Monier-William's dictionary, as distributed by the University of Cologne, to extract the necessary data.
The result is far from being satisfactory enough, but is still preferable, I believe, to the way polyglossia segments Sanskrit texts. Much improvements are still needed.
Copy sanskrit-hyphenation.fmtand sanskrit-hyphenation.sty to the directory
your document resides in (or in some other appropriate location). Then, to
compile a document named my_document.tex, use a command like the following:
xelatex -fmt sanskrit-hyphenation.fmt my_document.tex
Like for XeLaTeX, except that the file sanskrit-hyphenation.fmt should not be
used. Just copy the file sanskrit-hyphenation.sty in an appropriate location,
and compile your document as usual.
The package defines a single environment hyphenatesanskrit. Use it like so:
\documentclass{article}
\usepackage{sanskrit-hyphenation}
\begin{document}
\begin{hyphenatesanskrit}
kṛtvam
\end{hyphenatesanskrit}
\end{document}
To save some typing, you might want to define a command in the document preamble:
\newcommand\sanskritword[1]{\begin{hyphenatesanskrit}#1\end{hyphenatesanskrit}}
Then, somewhere in the document:
\sanskritword{kṛtvam}