Skip to content

README.md

Latest commit

 

History

History
69 lines (49 loc) · 1.94 KB

README.md

File metadata and controls

69 lines (49 loc) · 1.94 KB

Pyvarnam

     _     _
    /  \  /  \         |
    \_//  \_//  /\_____|
     _     |_   _ _ _ 
    ( )_| (|_) ()| | )O
                 _ _ _
                ()| | |
    ===================

Pyvarnam is a python wrapper to the libvarnam library.

Varnam is an open-source transliterator suited for Indian languages and the main component is a C library called libvarnam. Pyvarnam provides python bindings for libvarnam.

Installation

You need to install libvarnam for pyvarnam to work. You could refer libvarnam installation to find out how to install libvarnam.

You can install it from PyPi using

pip install pyvarnam

Or you can download the archive, un-tar/zip and cd into the directory you unzipped the archive into, and then run

 user@user-pc:~/unzipped-dir$ python setup.py install
Usage

To transliterate with pyvarnam, you just have to run the following.

from pyvarnam import varnam

var_lib = varnam()
var_lib.varnam_init_from_id("ml")
result = var_lib.varnam_transliterate("varnam")
if result is not None:
    print result[0][0]

As of now only some of the functions mentioned in libvarnam/api.h has been implemented. You could refer varnam.py to see which all functions are available.

Running the tests

A basic test suite has been implemented based on the transliteration tests in libvarnam. To run them, cd to the tests folder in the archive and run

python test.py

Currently pyvarnam has only been tested in Debian. If you find any bugs or would like to suggest some other changes, report the issue here or feel free to contact me.

License & Contact

This project is licensed under the MIT License. For more info, see LICENSE.txt Copyright (c) 2015 Sebin Thomas

You could contact me at me(at)sebin(dot)in or I'm @stc043 on twitter.