!7

lib 1400180 // github.com/optimaize/language-detector
lib 1400181 // jsonic
lib 1011966 // slf4j-api-1.7.25.jar
lib 1400182 // guava

static LS _stickyLibs_langDetect = ll(#1400180, #1400181, #1011966, #1400182);

import com.optimaize.langdetect.*;
import com.optimaize.langdetect.i18n.*;
import com.optimaize.langdetect.ngram.*;
import com.optimaize.langdetect.profiles.*;
import com.optimaize.langdetect.text.*;

cmodule LanguageDetectorModule {
  transient LanguageDetector languageDetector;
  
  void init() {
    lock lock;
    if (languageDetector != null) ret;
    final new LanguageProfileReader profileReader;
    L<LanguageProfile> languageProfiles = map(ll("de", "en"), func(S lang) -> LanguageProfile { profileReader.readBuiltIn(LdLocale.fromString(lang)) });
    languageDetector = LanguageDetectorBuilder.create(NgramExtractors.standard())
      .withProfiles(languageProfiles)
      .build();
  }

  // API
  
  // returns "en" or "de" or null
  S detectLanguage(S text) {
    S lang = dm_findAndCallModule("#1021121/WordToLanguageCRUD", 'languageForText, text);
    if (lang != null) ret languageToTwoLetters(lang);
    init();
    L<DetectedLanguage> languages = languageDetector.getProbabilities(text);
    ret empty(languages) ? null : first(languages).getLocale().getLanguage();
  }
}