!7

lib 1400180 // github.com/optimaize/language-detector

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

p-exp {
  new LanguageProfileReader profileReader;
  //L<LanguageProfile> languageProfiles = profileReader.readAllBuiltIn();
  L<LanguageProfile> languageProfiles = ll(profileReader.readBuiltIn(LdLocale.fromString("de")), profileReader.readBuiltIn(LdLocale.fromString("en")));
  
  LanguageDetector languageDetector = LanguageDetectorBuilder.create(NgramExtractors.standard())
    .withProfiles(languageProfiles)
    .build();

  // create a text object factory
  TextObjectFactory textObjectFactory = CommonTextObjectFactories.forDetectingOnLargeText();

  // query
  for (S text : ll("hello world", "hallo welt")) {
    TextObject textObject = textObjectFactory.forText(text);
    LdLocale lang = languageDetector.detect(textObject).orElse(null);
    print(text + " => " + lang);
  }
}