nepali-roman 0.45

Convert Nepali text to roman English

nepali-roman

Package to convert Nepali text to romanized English.

Many a time, while working with data, either from government sector or from private sector, we encounter Nepali names of VDCs, municipalities, local level units, districts and places. While merging such data with another datasheet consisting names of units in English, we need to romanize the Nepali names to English before using some text similarity algorithms to do string matching.

This package helps in romanizing Nepali names using rudimentary script. It is worth mentioning that romanization of Nepali text used in our daily life is inconsistent. For example, the popular romanization of "नेपाल" is "Nepal". It might seem as the correct romanization but if we have to distinguish between "नेपाल" and "नेपल", we can't do that with this romanization scheme. So, this package uses unique and standard romanization scheme such that no two Nepali words have same romanization.

Installation

nepali-roman package is available for Python 3 and can be installed using pip.

First make sure pip is installed.

Then, the package can be installed using this command.

pip install nepali-roman

Usage

Import the nepali_roman module using the following command.

import nepali_roman as nr

The nepali_roman module has two functions: romanize_text and romanize.

romanize_text

This function can be used to romanize the Nepali text to English.

Syntax:

>>> nr.romanize_text(nepali_text)

Example:

>>> import nepali_roman as nr
>>> nr.romanize_text("नगरपालिका")
    nagarapaalikaa

romanize

This function takes Nepali text file as an input and saves the romanized text in specified output file.

Syntax:

>>> nr.romanize(input_file_path, output_file_path)

Example:

>>> import nepali_roman as nr
>>> nr.romanize('nepali.txt', 'romanized_english.txt')
# this takes Nepali text file nepali.txt and stores romanized English in the file romanized_english.txt

Contributions

The package is licenced with The MIT License (MIT) about which details can be found in the LICENSE file. As the package is open sourced and requires many improvements and extensions, any contributions are welcome. Any other feedback of any sort are highly welcome.