"""
Language Detection Module
Detects the language of input text

Developer: RSK World
Website: https://rskworld.in
Email: help@rskworld.in
Phone: +91 93305 39277
Year: 2026
"""

from langdetect import detect, detect_langs, LangDetectException
import re

class LanguageDetector:
    """
    Language detection class
    Developer: RSK World - https://rskworld.in
    """
    
    def __init__(self):
        """Initialize language detector"""
        pass
    
    def detect_language(self, text):
        """
        Detect the primary language of text
        
        Args:
            text (str): Input text
            
        Returns:
            dict: Language detection results
        """
        # Clean text for better detection
        cleaned_text = re.sub(r'[^\w\s]', '', text)
        cleaned_text = ' '.join(cleaned_text.split())
        
        if len(cleaned_text) < 3:
            return {
                'language': 'unknown',
                'confidence': 0.0,
                'all_languages': []
            }
        
        try:
            # Detect primary language
            primary_lang = detect(cleaned_text)
            
            # Get all language probabilities
            lang_probabilities = detect_langs(cleaned_text)
            
            all_languages = [
                {
                    'language': lang.lang,
                    'probability': lang.prob
                }
                for lang in lang_probabilities[:5]  # Top 5
            ]
            
            # Get confidence for primary language
            confidence = next(
                (lang.prob for lang in lang_probabilities if lang.lang == primary_lang),
                0.0
            )
            
            return {
                'language': primary_lang,
                'confidence': confidence,
                'all_languages': all_languages
            }
        except LangDetectException:
            return {
                'language': 'unknown',
                'confidence': 0.0,
                'all_languages': []
            }
        except Exception as e:
            print(f"Error in language detection: {e}")
            return {
                'language': 'unknown',
                'confidence': 0.0,
                'all_languages': []
            }
    
    def is_english(self, text):
        """
        Check if text is in English
        
        Args:
            text (str): Input text
            
        Returns:
            bool: True if English, False otherwise
        """
        result = self.detect_language(text)
        return result['language'] == 'en' and result['confidence'] > 0.5