bcp47_language_tag.py

def make_bcp47_language_tag_re():
    """
    Reference: http://www.rfc-editor.org/rfc/bcp/bcp47.txt#
    Validator: http://schneegans.de/lv/
    """
    regular = (r"(art-lojban|cel-gaulish|no-bok|no-nyn|zh-guoyu|zh-hakka"
               "|zh-min|zh-min-nan|zh-xiang)")
    irregular = (r"(en-GB-oed|i-ami|i-bnn|i-default|i-enochian|i-hak|i-klingon"
                 "|i-lux|i-mingo|i-navajo|i-pwn|i-tao|i-tay|i-tsu|sgn-BE-FR"
                 "|sgn-BE-NL|sgn-CH-DE)")
    grandfathered = r"(?P<grandfathered>" + irregular + "|" + regular + ")"
    privateUse = r"(x(-[A-Za-z0-9]{1,8})+)"
    singleton = r"[0-9A-WY-Za-wy-z]"
    extension = r"(?P<extension>" + singleton + "(-[A-Za-z0-9]{2,8})+)"
    variant = r"(?P<variant>[A-Za-z0-9]{5,8}|[0-9][A-Za-z0-9]{3})"
    region = r"(?P<region>[A-Za-z]{2}|[0-9]{3})"
    script = r"(?P<script>[A-Za-z]{4})"
    extlang = r"(?P<extlang>[A-Za-z]{3}(-[A-Za-z]{3}){0,2})"
    language = (r"(?P<language>([A-Za-z]{2,3}(-" + extlang + ")?)|[A-Za-z]{4}"
                "|[A-Za-z]{5,8})")
    langtag = (r"(" + language + "(-" + script + ")?" + "(-" + region + ")?" +
               "(-" + variant + ")*" + "(-" + extension + ")*" + "(-" +
               privateUse + ")?" + ")")
    language_tag = (r"^(" + langtag + "|" + privateUse + "|" + grandfathered +
                   ")$")
    return re.compile(language_tag)


RE_BCP47_LANGUAGE_TAG = make_bcp47_language_tag_re()


def parse_language_tag(tag):
    match = RE_BCP47_LANGUAGE_TAG.match(tag)
    return None if match is None else match.groupdict()


def validate_language_tag(tag):
    validate = Schema(Match(RE_BCP47_LANGUAGE_TAG, msg='Invalid language tag'))
    return validate(tag)