Singlish

A comprehensive Sinhala to Singlish (romanized Sinhala) transliterator for Node.js.

Overview

This package converts Sinhala Unicode text (U+0D80 – U+0DFF) to romanized Singlish text, making it easier to read and type Sinhala using Latin characters.

Features

• Comprehensive character support: Handles all Sinhala vowels, consonants, vowel signs, and special characters
• Legacy font support: Decodes text typed in old Sinhala fonts (FM-Abhaya, Amalee, DL-Manel, kaputadotcom, Thibus-Sinhala) to Unicode or Singlish
• Accurate transliteration: Properly handles consonant clusters, gemination, and complex linguistic features
• Zero-Width Joiner handling: Automatically removes invisible formatting characters
• Flexible options: Configurable output with lowercase and space preservation options
• Type-safe: Handles invalid inputs gracefully
• Well-tested: 150 test cases covering edge cases
• No dependencies: Lightweight and standalone

Installation

npm install @suhasdissa/singlish

Usage

Basic Usage

The package supports both ES modules and CommonJS via named imports:

// ES modules
import { unicodeToSinglish } from '@suhasdissa/singlish';

// CommonJS
const { unicodeToSinglish } = require('@suhasdissa/singlish');

// Sinhala Unicode -> Singlish
const result = unicodeToSinglish('සිංහල');
console.log(result); // Output: "sinhala"

// More examples
console.log(unicodeToSinglish('ආයුබෝවන්')); // "aayuboowan"
console.log(unicodeToSinglish('කොළඹ'));      // "kolamba"
console.log(unicodeToSinglish('ශ්‍රී ලංකා')); // "shrii lankaa"

Note: singlish is a deprecated alias for unicodeToSinglish, kept for backward compatibility. New code should use unicodeToSinglish.

Legacy Font Conversion

Much older Sinhala content was typed in legacy fonts (FM-Abhaya, DL-Manel, etc.) that store text as repurposed ASCII/Latin-1 bytes rather than Sinhala Unicode. Use fontToUnicode to decode such text to Sinhala Unicode, or fontToSinglish to go straight to Singlish:

import { fontToUnicode, fontToSinglish, supportedFonts } from '@suhasdissa/singlish';

fontToSinglish('isxy,', 'FM-Abhaya'); // "sinhala"
fontToUnicode('isxy,', 'FM-Abhaya');  // "සිංහල"

console.log(supportedFonts);
// ['FM-Abhaya', 'Amalee', 'DL-Manel', 'kaputadotcom', 'Thibus-Sinhala']

Font names are matched case-insensitively. Passing an unsupported font name throws an error.

With Options

import { unicodeToSinglish } from '@suhasdissa/singlish';

// Convert to lowercase
const lowercase = unicodeToSinglish('සිංහල', { lowercase: true });
console.log(lowercase); // "sinhala"

// Control space preservation
const text = 'මම  කන්නේ  බත්'; // Multiple spaces
const result = unicodeToSinglish(text, { preserveSpaces: false });
console.log(result); // "mama kannee bath" (normalized spaces)

Detailed Transliteration

For debugging or analysis purposes, you can get a character-by-character breakdown:

import { singlishDetailed } from '@suhasdissa/singlish';

const detailed = singlishDetailed('සිංහල');
console.log(detailed);
// {
//   result: "sinhala",
//   breakdown: [
//     { original: "සි", transliterated: "si", position: 0 },
//     { original: "ං", transliterated: "n", position: 2 },
//     { original: "හ", transliterated: "ha", position: 3 },
//     { original: "ල", transliterated: "la", position: 4 }
//   ]
// }

API Reference

unicodeToSinglish(text, options)

Transliterates Sinhala Unicode text to Singlish.

Parameters:

• text (string): The Sinhala text to transliterate
• options (object, optional):
  • preserveSpaces (boolean, default: true): Whether to preserve multiple spaces
  • lowercase (boolean, default: false): Whether to convert output to lowercase

Returns: string - The romanized Singlish text

Example:

unicodeToSinglish('ආයුබෝවන්', { lowercase: true })
// Returns: "aayuboowan"

fontToUnicode(text, font)

Decodes text typed in a legacy Sinhala font to Sinhala Unicode.

Parameters:

• text (string): The legacy-font text
• font (string): One of supportedFonts (matched case-insensitively)

Returns: string - The Sinhala Unicode text (empty string for non-string input)

Throws: if font is not a supported font name.

Example:

fontToUnicode('isxy,', 'FM-Abhaya')
// Returns: "සිංහල"

fontToSinglish(text, font, options)

Converts legacy Sinhala font text directly to Singlish. Equivalent to unicodeToSinglish(fontToUnicode(text, font), options).

Parameters:

• text (string): The legacy-font text
• font (string): One of supportedFonts (matched case-insensitively)
• options (object, optional): Same options as unicodeToSinglish

Returns: string - The romanized Singlish text

Example:

fontToSinglish('isxy,', 'FM-Abhaya')
// Returns: "sinhala"

supportedFonts

An array of the supported legacy font names: ['FM-Abhaya', 'Amalee', 'DL-Manel', 'kaputadotcom', 'Thibus-Sinhala'].

singlish(text, options)

Deprecated alias for unicodeToSinglish, kept for backward compatibility.

singlishDetailed(text)

Provides detailed transliteration with character-by-character breakdown.

Parameters:

• text (string): The Sinhala text to transliterate

Returns: object - An object containing:

• result (string): The transliterated text
• breakdown (array): Array of objects with original, transliterated, and position properties

Example:

const details = singlishDetailed('මම');
// Returns:
// {
//   result: "mama",
//   breakdown: [
//     { original: "ම", transliterated: "ma", position: 0 },
//     { original: "ම", transliterated: "ma", position: 1 }
//   ]
// }

Transliteration Mapping

Vowels (Independent)

| Sinhala | Singlish | Unicode | |---------|----------|---------| | අ | a | U+0D85 | | ආ | aa | U+0D86 | | ඇ | ae | U+0D87 | | ඈ | ae | U+0D88 | | ඉ | i | U+0D89 | | ඊ | ii | U+0D8A | | උ | u | U+0D8B | | ඌ | uu | U+0D8C | | එ | e | U+0D91 | | ඒ | ee | U+0D92 | | ඔ | o | U+0D94 | | ඕ | oo | U+0D95 |

Consonants (with inherent 'a')

| Sinhala | Singlish | Unicode | |---------|----------|---------| | ක | ka | U+0D9A | | ග | ga | U+0D9C | | ච | cha | U+0DA0 | | ජ | ja | U+0DA2 | | ට | ta | U+0DA7 | | ඩ | da | U+0DA9 | | ත | tha | U+0DAD | | ද | dha | U+0DAF | | න | na | U+0DB1 | | ප | pa | U+0DB4 | | බ | ba | U+0DB6 | | ම | ma | U+0DB8 | | ය | ya | U+0DBA | | ර | ra | U+0DBB | | ල | la | U+0DBD | | ව | wa | U+0DC0 | | ස | sa | U+0DC3 | | හ | ha | U+0DC4 |

Vowel Signs (Dependent)

| Sinhala | Singlish | Unicode | |---------|----------|---------| | ා | aa | U+0DCF | | ැ | ae | U+0DD0 | | ෑ | ae | U+0DD1 | | ි | i | U+0DD2 | | ී | ii | U+0DD3 | | ු | u | U+0DD4 | | ූ | uu | U+0DD6 | | ෙ | e | U+0DD9 | | ේ | ee | U+0DDA | | ො | o | U+0DDC | | ෝ | oo | U+0DDD |

Special Characters

| Sinhala | Singlish | Description | Unicode | |---------|----------|-------------|---------| | ං | n | Anusvara | U+0D82 | | ඃ | h | Visarga | U+0D83 | | ් | - | Halanta (removes inherent vowel) | U+0DCA |

Examples

Common Words

unicodeToSinglish('අම්මා')      // "ammaa" (mother)
unicodeToSinglish('තාත්තා')    // "thaaththaa" (father)
unicodeToSinglish('ගෙදර')      // "gedhara" (home)
unicodeToSinglish('පාසල')      // "paasala" (school)

Sentences

unicodeToSinglish('සුභ දවසක්')        // "subha dhawasak" (good day)
unicodeToSinglish('මම කන්නේ බත්')     // "mama kannee bath" (I eat rice)

Mixed Content

The transliterator preserves non-Sinhala characters:

unicodeToSinglish('Hello මගේ නම John')
// "Hello magee nama John"

unicodeToSinglish('මම 10 යි')
// "mama 10 yi"

How It Works

1. Input Processing: Removes Zero-Width Joiner (ZWJ) and other invisible formatting characters
2. Character Analysis: Processes text character by character, identifying vowels, consonants, vowel signs, and special characters
3. Smart Handling:
   • Consonants with halanta (්) remove the inherent 'a' sound
   • Consonants with vowel signs replace the inherent 'a' with the appropriate vowel
   • Standalone consonants retain the inherent 'a'
4. Post-Processing: Cleans up any duplicate characters and normalizes spacing

Testing

The package includes 150 test cases covering:

• Independent vowels
• Consonants with inherent vowels
• Consonants with halanta (virama)
• Consonants with vowel signs
• Special characters (anusvara, visarga)
• Common Sinhala words
• Sentences and phrases
• Consonant clusters
• Edge cases (empty strings, null values, non-Sinhala text)
• Real-world examples
• Unicode edge cases
• Legacy font conversion (FM-Abhaya, Amalee, DL-Manel, kaputadotcom, Thibus-Sinhala)

Run tests with:

npm test

Limitations

• The transliteration is phonetic and may not always match exact pronunciation
• Some consonant clusters produce multiple consonant letters in sequence (e.g., බුද්ධ → "budhdhha")
• The package focuses on standard Sinhala; archaic or rarely-used characters may not be fully supported

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.

License

GPL version 3.0

Author

Suhas Dissanayake [email protected]

Keywords

sinhala, singlish, transliteration, romanization, sri lanka, unicode, language