Since it is a language processing library, it probably makes more sense to use language codes ISO_639-1, ISO_639-2, ISO_639-3 than country codes ISO 3166-1.

• Belarusian language code is be not by (ISO 3166-1)

• Montenegrin language code is cnr (ISO_639-2) not me (ISO 3166-1)

• Serbian language code is sr not rs (ISO 3166-1)

• Tajik language code is tg not tj (ISO 3166-1)

• Ukrainian language code is uk not ua (ISO 3166-1)

• Use Case : If I want to display programmatically the list of language supported, I need to get supported codes from cyrtranslit.supported() and then do the conversion from the country ISO 3166 to the language ISO_639 to have the name of the language and not the name of the country.

• Use Case : I am using langdetect library to detect the language of the text to transcript and use this language code (ISO_639) as a parameter of cyrtranslit. It will not work for the languages listed above because cyrtranslit is using country codes.

• Finally, I will recommend to use only ISO_639-2 (not ISO_639-1) since Montenegrin do not have ISO_639-1
  In such case the list of supported languages will be:

  • Belarusian : bel
  • Bulgarian : bul
  • Greek : ell
  • Montenegrin : cnr
  • Macedonian : mkd
  • Mongolian : mon
  • Russian : rus
  • Serbian : srp
  • Tajik : tgk
  • Ukrainian : ukr