nepali_unicoder 0.2.0

A Python package for Roman to Nepali (Devanagari) transliteration

Nepali Unicoder

A robust Python package for converting Romanized Nepali text and Preeti font text into Unicode Devanagari script. It uses a greedy matching algorithm for Roman transliteration and a two-phase conversion process for Preeti with contextual rules.

Read the Full Documentation for detailed usage guides, Preeti mapping references, and API details.

Features

• Accurate Transliteration: Uses a greedy matching algorithm to prioritize longer phonetic matches (e.g., 'kha' is matched before 'k' and 'h').
• Preeti Font Support: Full support for Preeti to Unicode conversion with 30+ contextual rules for accurate transformation.
• Smart Vowel Handling: Distinguishes between independent vowels (e.g., 'aa' -> 'आ') and vowel signs/matras (e.g., 'ka' -> 'क', 'kaa' -> 'का').
• Contextual Rules: Handles complex Devanagari rules like reph positioning, matra reordering, and special character combinations.
• Mixed Content Support: Allows keeping English words or specific text in Roman script using {} blocks.
• Customizable: Supports custom word-level overrides via word_maps.json.
• CLI Support: Can be used directly from the command line.

Installation

You can install the package locally:

pip install nepali-unicoder

Usage

Command Line Interface (CLI)

You can use the converter directly from the terminal:

# Using the installed command
nepali-unicoder "namaste"
# Output: नमस्ते

# Or using python module
python -m nepali_unicoder "namaste"
# Output: नमस्ते

# Pipe input
echo "mero naam sanjeev ho" | nepali-unicoder
# Output: मेरो नाम् सन्जीव् हो

Python API

from nepali_unicoder.convert import Converter

converter = Converter()

# Basic conversion
text = "namaste nepal"
print(converter.convert(text))
# Output: नमस्ते नेपाल

# Using 'as-is' blocks for English text
mixed_text = "mero naam {Sanjeev} ho"
print(converter.convert(mixed_text))
# Output: मेरो नाम् Sanjeev हो

Preeti Mode

Convert Preeti font text to Unicode with full support for contextual rules:

from nepali_unicoder.convert import Converter

# Create converter in Preeti mode
preeti_converter = Converter(mode="preeti")

# Basic conversion
preeti_text = "s{sf"  # Preeti characters
print(preeti_converter.convert(preeti_text))
# Output: र्कका

# The converter handles:
# - Reph positioning: { → र् (moves before consonant)
# - Matra reordering: l (ि) moves after consonant
# - Special m transformations
# - Vowel combinations
# - Literal brackets: { and } are treated as normal characters in Preeti mode

Preeti Character Examples

Preeti	Unicode	Description
s	क	Consonant ka
s{	र्क	Reph + ka (contextual)
sl	कि	ka + short i (reordered)
qm	क्र	Special m transformation
!@#	१२३	Nepali numbers
Ù / Ú	; / :	Literal punctuation
« / »	्र	Ra-foot (for ट, ठ, ड, ढ)
¿	रू	Combined ruu
å	द्व	Combined dva
ˆ	फ्	Half ph
ª	ङ	Consonant nga
æ / Æ	“ / ”	Curly quotes
¥	र्	Half ra
¶	ठ्ठ	Combined thth
§	ट्ट	Combined tt
£	घ्	Half gh
Ë / Í	ङ्ग / ङ्क	Combined nga-ga / nga-ka
‰	झ्	Half jh

CLI for Preeti

python -m nepali_unicoder --preeti "s{sf"
# Output: र्कका

Transliteration Rules

• Consonants: k -> क्, ka -> क, kh -> ख्, kha -> ख
• Vowels: a -> अ, aa -> आ, i -> इ, u -> उ
• Matras: ki -> कि, ko -> को
• Special: . -> ।, .. -> ॥
• Numbers: 0-9 -> ०-९ (Decimal points are preserved: 1.5 -> १.५)

Advanced Usage

Handling Complex Text

The converter handles mixed content gracefully. You can use {} to keep text as-is (e.g., for English words or code snippets).

text = "mero naam {Sanjeev} ho ra ma 12.5 barsa ko bhaye."
print(converter.convert(text))
# Output: मेरो नाम् Sanjeev हो र म १२.५ बर्स को भये।

Configuration

The package uses word_maps.json for custom word-level overrides, located in the src/nepali_unicoder directory.

1. word_maps.json: Defines custom word-level overrides. Use this for words that don't follow standard phonetic rules.

Example word_maps.json:

{
    "nepal": "नेपाल",
    "kathamandu": "काठमाडौँ"
}

Contribution

We welcome contributions! Here's how you can help:

1. Clone the repository:

   git clone https://github.com/realsanjeev/nepali_unicoder.git
   cd nepali_unicoder
   

2. Set up a virtual environment:

   python3 -m venv .venv
   source .venv/bin/activate
   pip install -e .
   

3. Run tests:

   python -m unittest discover tests
   

4. Submit a Pull Request: Create a new branch, make your changes, and submit a PR.

Development

To run tests:

python -m unittest discover tests

License

MIT