cmlenz@263: # -*- coding: utf-8 -*-
cmlenz@263: #
cmlenz@263: # Copyright (C) 2007 Edgewall Software
cmlenz@263: # All rights reserved.
cmlenz@263: #
cmlenz@263: # This software is licensed as described in the file COPYING, which
cmlenz@263: # you should have received as part of this distribution. The terms
cmlenz@263: # are also available at http://babel.edgewall.org/wiki/License.
cmlenz@263: #
cmlenz@263: # This software consists of voluntary contributions made by many
cmlenz@263: # individuals. For the exact contribution history, see the revision
cmlenz@263: # history and logs, available at http://babel.edgewall.org/log/.
cmlenz@263: 
cmlenz@263: """Plural form definitions."""
cmlenz@263: 
cmlenz@381: 
cmlenz@381: from operator import itemgetter
cmlenz@381: from babel.core import default_locale, Locale
cmlenz@381: 
cmlenz@381: 
cmlenz@381: LC_CTYPE = default_locale('LC_CTYPE')
cmlenz@381: 
cmlenz@381: 
cmlenz@263: PLURALS = {
cmlenz@320:     # Afar
cmlenz@320:     # 'aa': (),
cmlenz@320:     # Abkhazian
cmlenz@320:     # 'ab': (),
cmlenz@320:     # Avestan
cmlenz@320:     # 'ae': (),
cmlenz@263:     # Afrikaans - From Pootle's PO's
cmlenz@263:     'af': (2, '(n != 1)'),
cmlenz@320:     # Akan
cmlenz@320:     # 'ak': (),
cmlenz@320:     # Amharic
cmlenz@320:     # 'am': (),
cmlenz@320:     # Aragonese
cmlenz@320:     # 'an': (),
cmlenz@263:     # Arabic - From Pootle's PO's
cmlenz@263:     'ar': (6, '(n==0 ? 0 : n==1 ? 1 : n==2 ? 2 : n>=3 && n<=10 ? 3 : n>=11 && n<=99 ? 4 : 5)'),
cmlenz@320:     # Assamese
cmlenz@320:     # 'as': (),
cmlenz@320:     # Avaric
cmlenz@320:     # 'av': (),
cmlenz@320:     # Aymara
cmlenz@320:     # 'ay': (),
cmlenz@320:     # Azerbaijani
cmlenz@320:     # 'az': (),
cmlenz@320:     # Bashkir
cmlenz@320:     # 'ba': (),
cmlenz@320:     # Belarusian
cmlenz@320:     # 'be': (),
cmlenz@263:     # Bulgarian - From Pootle's PO's
cmlenz@263:     'bg': (2, '(n != 1)'),
cmlenz@320:     # Bihari
cmlenz@320:     # 'bh': (),
cmlenz@320:     # Bislama
cmlenz@320:     # 'bi': (),
cmlenz@320:     # Bambara
cmlenz@320:     # 'bm': (),
cmlenz@263:     # Bengali - From Pootle's PO's
cmlenz@263:     'bn': (2, '(n != 1)'),
cmlenz@302:     # Tibetan - as discussed in private with Andrew West
cmlenz@302:     'bo': (1, '0'),
cmlenz@320:     # Breton
cmlenz@320:     # 'br': (),
cmlenz@320:     # Bosnian
cmlenz@320:     # 'bs': (),
cmlenz@263:     # Catalan - From Pootle's PO's
cmlenz@263:     'ca': (2, '(n != 1)'),
cmlenz@320:     # Chechen
cmlenz@320:     # 'ce': (),
cmlenz@320:     # Chamorro
cmlenz@320:     # 'ch': (),
cmlenz@320:     # Corsican
cmlenz@320:     # 'co': (),
cmlenz@320:     # Cree
cmlenz@320:     # 'cr': (),
cmlenz@263:     # Czech
cmlenz@263:     'cs': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@320:     # Church Slavic
cmlenz@320:     # 'cu': (),
cmlenz@320:     # Chuvash
cmlenz@320:     'cv': (1, '0'),
cmlenz@320:     # Welsh
cmlenz@348:     'cy': (5, '(n==1 ? 1 : n==2 ? 2 : n==3 ? 3 : n==6 ? 4 : 0)'),
cmlenz@263:     # Danish
cmlenz@263:     'da': (2, '(n != 1)'),
cmlenz@263:     # German
cmlenz@263:     'de': (2, '(n != 1)'),
cmlenz@320:     # Divehi
cmlenz@320:     # 'dv': (),
cmlenz@320:     # Dzongkha
cmlenz@320:     'dz': (1, '0'),
cmlenz@263:     # Greek
cmlenz@263:     'el': (2, '(n != 1)'),
cmlenz@263:     # English
cmlenz@263:     'en': (2, '(n != 1)'),
cmlenz@263:     # Esperanto
cmlenz@263:     'eo': (2, '(n != 1)'),
cmlenz@263:     # Spanish
cmlenz@263:     'es': (2, '(n != 1)'),
cmlenz@263:     # Estonian
cmlenz@263:     'et': (2, '(n != 1)'),
cmlenz@263:     # Basque - From Pootle's PO's
cmlenz@263:     'eu': (2, '(n != 1)'),
cmlenz@263:     # Persian - From Pootle's PO's
cmlenz@263:     'fa': (1, '0'),
cmlenz@263:     # Finnish
cmlenz@263:     'fi': (2, '(n != 1)'),
cmlenz@263:     # French
cmlenz@263:     'fr': (2, '(n > 1)'),
cmlenz@302:     # Friulian - From Pootle's PO's
cmlenz@263:     'fur': (2, '(n > 1)'),
cmlenz@263:     # Irish
cmlenz@348:     'ga': (3, '(n==1 ? 0 : n==2 ? 1 : 2)'),
cmlenz@302:     # Galician - From Pootle's PO's
cmlenz@263:     'gl': (2, '(n != 1)'),
cmlenz@263:     # Hausa - From Pootle's PO's
cmlenz@348:     'ha': (2, '(n != 1)'),
cmlenz@263:     # Hebrew
cmlenz@263:     'he': (2, '(n != 1)'),
cmlenz@263:     # Hindi - From Pootle's PO's
cmlenz@348:     'hi': (2, '(n != 1)'),
cmlenz@263:     # Croatian
cmlenz@263:     'hr': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Hungarian
cmlenz@263:     'hu': (1, '0'),
cmlenz@263:     # Armenian - From Pootle's PO's
cmlenz@263:     'hy': (1, '0'),
cmlenz@263:     # Icelandic - From Pootle's PO's
cmlenz@263:     'is': (2, '(n != 1)'),
cmlenz@263:     # Italian
cmlenz@263:     'it': (2, '(n != 1)'),
cmlenz@263:     # Japanese
cmlenz@263:     'ja': (1, '0'),
cmlenz@263:     # Georgian - From Pootle's PO's
cmlenz@263:     'ka': (1, '0'),
cmlenz@263:     # Kongo - From Pootle's PO's
cmlenz@263:     'kg': (2, '(n != 1)'),
cmlenz@263:     # Khmer - From Pootle's PO's
cmlenz@263:     'km': (1, '0'),
cmlenz@263:     # Korean
cmlenz@263:     'ko': (1, '0'),
cmlenz@302:     # Kurdish - From Pootle's PO's
cmlenz@263:     'ku': (2, '(n != 1)'),
cmlenz@320:     # Lao - Another member of the Tai language family, like Thai.
cmlenz@320:     'lo': (1, '0'),
cmlenz@263:     # Lithuanian
cmlenz@263:     'lt': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Latvian
cmlenz@263:     'lv': (3, '(n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2)'),
cmlenz@263:     # Maltese - From Pootle's PO's
cmlenz@263:     'mt': (4, '(n==1 ? 0 : n==0 || ( n%100>1 && n%100<11) ? 1 : (n%100>10 && n%100<20 ) ? 2 : 3)'),
cmlenz@302:     # Norwegian Bokmål
cmlenz@263:     'nb': (2, '(n != 1)'),
cmlenz@263:     # Dutch
cmlenz@263:     'nl': (2, '(n != 1)'),
cmlenz@263:     # Norwegian Nynorsk
cmlenz@263:     'nn': (2, '(n != 1)'),
cmlenz@263:     # Norwegian
cmlenz@263:     'no': (2, '(n != 1)'),
cmlenz@263:     # Punjabi - From Pootle's PO's
cmlenz@263:     'pa': (2, '(n != 1)'),
cmlenz@263:     # Polish
cmlenz@263:     'pl': (3, '(n==1 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Portuguese
cmlenz@263:     'pt': (2, '(n != 1)'),
cmlenz@263:     # Brazilian
cmlenz@263:     'pt_BR': (2, '(n > 1)'),
cmlenz@263:     # Romanian - From Pootle's PO's
cmlenz@263:     'ro': (3, '(n==1 ? 0 : (n==0 || (n%100 > 0 && n%100 < 20)) ? 1 : 2)'),
cmlenz@263:     # Russian
cmlenz@263:     'ru': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Slovak
cmlenz@263:     'sk': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Slovenian
cmlenz@263:     'sl': (4, '(n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3)'),
cmlenz@263:     # Serbian - From Pootle's PO's
cmlenz@263:     'sr': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10< =4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@302:     # Southern Sotho - From Pootle's PO's
cmlenz@263:     'st': (2, '(n != 1)'),
cmlenz@263:     # Swedish
cmlenz@263:     'sv': (2, '(n != 1)'),
cmlenz@302:     # Thai
cmlenz@302:     'th': (1, '0'),
cmlenz@263:     # Turkish
cmlenz@263:     'tr': (1, '0'),
cmlenz@263:     # Ukrainian
cmlenz@263:     'uk': (3, '(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)'),
cmlenz@263:     # Venda - From Pootle's PO's
cmlenz@263:     've': (2, '(n != 1)'),
cmlenz@263:     # Vietnamese - From Pootle's PO's
cmlenz@263:     'vi': (1, '0'),
cmlenz@263:     # Xhosa - From Pootle's PO's
cmlenz@263:     'xh': (2, '(n != 1)'),
cmlenz@263:     # Chinese - From Pootle's PO's
cmlenz@263:     'zh_CN': (1, '0'),
cmlenz@263:     'zh_HK': (1, '0'),
cmlenz@263:     'zh_TW': (1, '0'),
cmlenz@263: }
cmlenz@381: 
cmlenz@381: 
cmlenz@381: DEFAULT_PLURAL = (2, '(n != 1)')
cmlenz@381: 
cmlenz@381: 
cmlenz@381: class _PluralTuple(tuple):
cmlenz@381:     """A tuple with plural information."""
cmlenz@381: 
cmlenz@381:     __slots__ = ()
cmlenz@381:     num_plurals = property(itemgetter(0), doc="""
cmlenz@381:     The number of plurals used by the locale.""")
cmlenz@381:     plural_expr = property(itemgetter(1), doc="""
cmlenz@381:     The plural expression used by the locale.""")
cmlenz@381:     plural_forms = property(lambda x: 'npurals=%s; plural=%s' % x, doc="""
cmlenz@381:     The plural expression used by the catalog or locale.""")
cmlenz@381: 
cmlenz@381:     def __str__(self):
cmlenz@381:         return self.plural_forms
cmlenz@381: 
cmlenz@381: 
cmlenz@381: def get_plural(locale=LC_CTYPE):
cmlenz@381:     """A tuple with the information catalogs need to perform proper
cmlenz@381:     pluralization.  The first item of the tuple is the number of plural
cmlenz@381:     forms, the second the plural expression.
cmlenz@381: 
cmlenz@381:     >>> get_plural(locale='en')
cmlenz@381:     (2, '(n != 1)')
cmlenz@381:     >>> get_plural(locale='ga')
cmlenz@381:     (3, '(n==1 ? 0 : n==2 ? 1 : 2)')
cmlenz@381: 
cmlenz@381:     The object returned is a special tuple with additional members:
cmlenz@381: 
cmlenz@381:     >>> tup = get_plural("ja")
cmlenz@381:     >>> tup.num_plurals
cmlenz@381:     1
cmlenz@381:     >>> tup.plural_expr
cmlenz@381:     '0'
cmlenz@381:     >>> tup.plural_forms
cmlenz@381:     'npurals=1; plural=0'
cmlenz@381: 
cmlenz@381:     Converting the tuple into a string prints the plural forms for a
cmlenz@381:     gettext catalog:
cmlenz@381: 
cmlenz@381:     >>> str(tup)
cmlenz@381:     'npurals=1; plural=0'
cmlenz@381:     """
cmlenz@381:     locale = Locale.parse(locale)
cmlenz@381:     try:
cmlenz@381:         tup = PLURALS[str(locale)]
cmlenz@381:     except KeyError:
cmlenz@381:         try:
cmlenz@381:             tup = PLURALS[locale.language]
cmlenz@381:         except KeyError:
cmlenz@381:             tup = DEFAULT_PLURAL
cmlenz@381:     return _PluralTuple(tup)