Translib

Translib

A common task that indian language developers face is converting input in various forms (transliterated roman text, a bytestream encoded according to a well-known character encoding standard, etc) to a sequence of correctly positioned glyphs from a font.

The Translib library is intended to encapsulate this functionality into a library that applications can use.

Executive Summary

Rationale

The problem of text layout for indian language scripts seems to have been partially solved many times in the past. The earliest transliteration schemes (e.g. ITRANS), converted transliterated US-ASCII text to glyph sequences for a specific fonts. The ICU project from IBM offers facilities to render Unicode encoded characters. The GNOME project, as another example, offers the Pango renderer, capable of rendering Unicode encoded text.

However, there seems to be a lack of a small library, available under a liberal license, that can handle multiple kinds of input streams, and which can work with multiple font technologies.

Background Information

The Pango project is worth studying.. The information at the Unicode consortium website is essential reading. Simpler transliteration schemes for a few languages (and usually specific to a few fonts) like Varamozhi (Malayalam) and ITRANS (many languages, but uses TeX for rendering) exist, and can be studied.

The Central Institute of Indian Languages is considered a good source of Indian language information.

Skill Set Required

Competence in the C programming language. Expertise with Indian scripts.

Other Resource Requirements

None.

Time Estimate

3--4 months of concentrated effort.

Status

Being handled by: K Nagarajan <k_nagarajan[@]users.sourceforge.net>.

Comments

This project needs linguistic expertise especially for the languages in use in North-East India. Please join up.

Copyright © 2001--2009 The Indic-Computing Project. Contact: jkoshy

View document revision history

Site Search