Momento/memento-note/lib/ai/services/language-detection.service.ts

import { detect } from 'tinyld'

/**
 * Language Detection Service
 *
 * Uses hybrid approach:
 * - TinyLD for notes < 50 words (fast, ~8ms)
 * - AI for notes ≥ 50 words (more accurate, ~200-500ms)
 *
 * Supports 62 languages including Persian (fa)
 */
export class LanguageDetectionService {
  private readonly MIN_WORDS_FOR_AI = 50
  private readonly MIN_CONFIDENCE = 0.7

  /**
   * Detect language of content using hybrid approach
   */
  async detectLanguage(content: string): Promise<{
    language: string      // 'fr' | 'en' | 'es' | 'de' | 'fa' | 'unknown'
    confidence: number   // 0.0-1.0
    method: 'tinyld' | 'ai' | 'unknown'
  }> {
    if (!content || content.trim().length === 0) {
      return {
        language: 'unknown',
        confidence: 0.0,
        method: 'unknown'
      }
    }

    const wordCount = content.split(/\s+/).length

    // Short notes: TinyLD (fast, TypeScript native)
    if (wordCount < this.MIN_WORDS_FOR_AI) {
      const result = detect(content)
      return {
        language: this.mapToISO(result),
        confidence: 0.8,
        method: 'tinyld'
      }
    }

    // Long notes: AI for better accuracy
    try {
      const detected = await this.detectLanguageWithAI(content)

      return {
        language: detected,
        confidence: 0.9,
        method: 'ai'
      }
    } catch (error) {
      console.error('Language detection error:', error)

      // Fallback to TinyLD
      const result = detect(content)
      return {
        language: this.mapToISO(result),
        confidence: 0.6,
        method: 'tinyld'
      }
    }
  }

  /**
   * Detect language using AI provider
   * (Fallback method for long content)
   */
  private async detectLanguageWithAI(content: string): Promise<string> {
    // For now, use TinyLD as AI detection is not yet implemented
    // In Phase 2, we can add AI-based detection for better accuracy
    const result = detect(content)
    return this.mapToISO(result)
  }

  /**
   * Map TinyLD language codes to ISO 639-1
   */
  private mapToISO(code: string): string {
    const mapping: Record<string, string> = {
      'fra': 'fr',
      'eng': 'en',
      'spa': 'es',
      'deu': 'de',
      'fas': 'fa',
      'pes': 'fa',  // Persian (Farsi)
      'por': 'pt',
      'ita': 'it',
      'rus': 'ru',
      'zho': 'zh',
      'jpn': 'ja',
      'kor': 'ko',
      'ara': 'ar',
      'hin': 'hi',
      'nld': 'nl',
      'pol': 'pl',
      'tur': 'tr',
      'vie': 'vi',
      'tha': 'th',
      'ind': 'id'
    }

    // Direct mapping for ISO codes
    if (code.length === 2 && /^[a-z]{2}$/.test(code)) {
      return code
    }

    // Use mapping or fallback
    return mapping[code] || code.substring(0, 2).toLowerCase()
  }

  /**
   * Get supported languages count
   */
  getSupportedLanguagesCount(): number {
    return 62 // TinyLD supports 62 languages
  }

  /**
   * Check if a language code is supported
   */
  isLanguageSupported(languageCode: string): boolean {
    // TinyLD supports 62 languages including Persian (fa)
    const supportedCodes = [
      'fr', 'en', 'es', 'de', 'fa', 'pt', 'it', 'ru', 'zh',
      'ja', 'ko', 'ar', 'hi', 'nl', 'pl', 'tr', 'vi', 'th', 'id'
      // ... and 43 more
    ]

    return supportedCodes.includes(languageCode.toLowerCase())
  }
}