Avro Phonetic

A high-performance Avro Phonetic transliteration library for PHP. Convert Banglish (Romanized Bengali) to Bengali script with blazing-fast Trie-based pattern matching and context-sensitive rule engine.

ami banglay gan gai → আমি বাংলায় গান গাই

✨ Features

• 🚀 Trie-Based Pattern Matching — O(m) lookup time instead of O(n×m) linear search
• 🎯 Context-Sensitive Rules — Intelligent conversion based on surrounding characters
• 🔌 Laravel Ready — Auto-discovery, Facade, and config publishing out of the box
• 📦 Standalone Compatible — Works with any PHP project, no framework required
• 🧪 Fully Tested — 180+ tests covering all functionality
• ⚡ Memory Efficient — Singleton pattern with lazy-loaded grammar
• 🔧 PHP 5.6+ — Compatible with PHP 5.6 through 8.x

🎯 What Makes This Unique?

Trie Data Structure for Fast Lookups

Unlike traditional implementations that linearly scan through hundreds of patterns, we use a Trie (prefix tree) data structure. This provides:

• O(m) lookup time where m is the pattern length (vs O(n×m) for linear search)
• Efficient longest-match finding — automatically finds the longest matching pattern
• Minimal memory overhead — shared prefixes are stored only once

Context-Sensitive Rule Engine

Our rule engine understands context! The same input can produce different outputs based on what comes before or after:

Avro::to('rri');   // ঋ (at word start)
Avro::to('krri');  // কৃ (after consonant)

Avro::to('OI');    // ঐ (standalone)
Avro::to('kOI');   // কৈ (after consonant)

Supported rule scopes:

• vowel / !vowel — Check if prefix/suffix is a vowel
• consonant / !consonant — Check if prefix/suffix is a consonant
• punctuation / !punctuation — Check for word boundaries
• exact / !exact — Match specific characters

📦 Installation

composer require eru/avro-phonetic

🚀 Quick Start

Basic Usage

use Eru\AvroPhonetic\Avro;

// Static method
echo Avro::to('ami banglay gan gai');
// Output: আমি বাংলায় গান গাই

// Instance method
$avro = new Avro();
echo $avro->parse('amar sonar bangla');
// Output: আমার সনার বাংলা

Helper Functions

// Quick conversion
echo avro('bangladesh');  // বাংলাদেশ
echo bangla('dhaka');     // ঢাকা

// Get instance for chaining
$result = avro()->parse('tumi kemon acho');

Laravel Usage

The package auto-registers with Laravel 5.5+. No additional setup required!

Using the Facade:

use Eru\AvroPhonetic\Facades\Avro;

// In your controller
public function store(Request $request)
{
    $bengaliText = Avro::to($request->input('text'));
    // ...
}

In Blade templates:

<p>{{ avro('ami tomake bhalobashi') }}</p>
<p>{{ bangla('sundor bangladesh') }}</p>

Publish configuration (optional):

php artisan vendor:publish --tag=avro-phonetic-config

This creates config/avro-phonetic.php where you can customize the grammar file path.

📖 API Reference

Avro Class

use Eru\AvroPhonetic\Avro;

// Static conversion
Avro::to(string $text): string

// Instance methods
$avro = new Avro();
$avro->parse(string $text): string
$avro->convert(string $text): string  // Alias for parse()

// Factory methods
Avro::getInstance(): Avro                        // Singleton
Avro::withGrammar(array $grammar): Avro          // Custom grammar
Avro::fromGrammarFile(string $path): Avro        // Load from file

// Access parser
$avro->getParser(): PhoneticParser

Helper Functions

// Convert text or get instance
avro(?string $text = null): Avro|string

// Always converts text
bangla(string $text): string

⚙️ Advanced Usage

Custom Grammar

You can provide your own grammar rules:

$customGrammar = [
    'patterns' => [
        ['find' => 'ph', 'replace' => 'ফ', 'rules' => []],
        // ... more patterns
    ],
    'vowel' => 'aeiou',
    'consonant' => 'bcdfghjklmnpqrstvwxyz',
    'number' => '1234567890',
    'casesensitive' => 'oiudgjnrstyz',
];

$avro = Avro::withGrammar($customGrammar);
echo $avro->parse('phone'); // ফনে

Load Grammar from File

$avro = Avro::fromGrammarFile('/path/to/custom-grammar.json');

Direct Parser Access

For performance-critical applications, access the parser directly:

use Eru\AvroPhonetic\PhoneticParser;

$grammar = json_decode(file_get_contents('grammar.json'), true);
$parser = new PhoneticParser($grammar);

// Parse multiple texts
foreach ($texts as $text) {
    $results[] = $parser->parse($text);
}

🏎️ Performance

Benchmarks on typical hardware (PHP 8.x):

| Operation | Time | |-----------|------| | Single word conversion | ~0.05ms | | Sentence (10 words) | ~0.1ms | | Paragraph (100 words) | ~0.8ms | | Trie initialization | ~15ms (one-time) |

The Trie is built once and reused via singleton pattern, making subsequent conversions extremely fast.

🧪 Testing

# Run all tests
composer test

# Or directly with PHPUnit
./vendor/bin/phpunit

# Run specific test suite
./vendor/bin/phpunit --testsuite Parser
./vendor/bin/phpunit --testsuite Performance

# With coverage report
./vendor/bin/phpunit --coverage-html coverage

📁 Project Structure

avro-phonetic/
├── src/
│   ├── Avro.php                    # Main entry point
│   ├── PhoneticParser.php          # Core parser with rule engine
│   ├── Trie.php                    # Trie data structure
│   ├── TrieNode.php                # Trie node class
│   ├── AvroPhoneticServiceProvider.php  # Laravel service provider
│   ├── Facades/
│   │   └── Avro.php                # Laravel facade
│   └── helpers.php                 # Global helper functions
├── config/
│   └── avro-phonetic.php           # Laravel config
├── resources/
│   └── grammar.json                # Avro phonetic rules
└── tests/
    ├── AvroTest.php
    ├── PhoneticParserTest.php
    ├── RuleEngineTest.php
    ├── TrieTest.php
    └── ...

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Write tests for your changes
4. Ensure all tests pass (composer test)
5. Commit your changes (git commit -m 'Add amazing feature')
6. Push to the branch (git push origin feature/amazing-feature)
7. Open a Pull Request

📜 License

This package is open-sourced software licensed under the GPL-3.0 License.

🙏 Credits

• Based on the Avro Phonetic keyboard layout
• Rule engine ported from pyAvroPhonetic by Kaustav Das Modak
• Developed by Erfan Ahmed Siam

Made with ❤️ for the Bengali language