# -*- coding: utf-8 -*-
#
# Copyright © 2007-2010 Dieter Verfaillie <dieterv@optionexplicit.be>
#
# This file is part of elib.intl.
#
# elib.intl is free software: you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# elib.intl is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with elib.intl. If not, see <http://www.gnu.org/licenses/>.
'''
The elib.intl module provides enhanced internationalization (I18N) services for
your Python modules and applications.
elib.intl wraps Python's :func:`gettext` functionality and adds the following on
Microsoft Windows systems:
- automatic detection of the current screen language (not necessarily the same
as the installation language) provided by MUI packs,
- makes sure internationalized C libraries which internally invoke gettext() or
dcgettext() can properly locate their message catalogs. This fixes a known
limitation in gettext's Windows support when using eg. Gtk.builder or Gtk.glade.
See http://www.gnu.org/software/gettext/FAQ.html#windows_setenv for more
information.
The elib.intl module defines the following functions:
'''
__all__ = ['install', 'install_module']
__version__ = '0.0.3'
__docformat__ = 'restructuredtext'
import os
import sys
import locale
import gettext
from logging import getLogger
logger = getLogger('elib.intl')
def _isofromlcid(lcid):
'''
:param lcid: Microsoft Windows LCID
:returns: the ISO 639-1 language code for a given lcid. If there is no
ISO 639-1 language code assigned to the language specified by lcid,
the ISO 639-2 language code is returned. If the language specified
by lcid is unknown in the ISO 639-x database, None is returned.
More information can be found on the following websites:
- List of ISO 639-1 and ISO 639-2 language codes: http://www.loc.gov/standards/iso639-2/
- List of known lcid's: http://www.microsoft.com/globaldev/reference/lcid-all.mspx
- List of known MUI packs: http://www.microsoft.com/globaldev/reference/win2k/setup/Langid.mspx
'''
mapping = {1078: 'af', #Afrikaans - South Africa
1052: 'sq', #Albanian - Albania
1118: 'am', #Amharic - Ethiopia
1025: 'ar', #Arabic - Saudi Arabia
5121: 'ar', #Arabic - Algeria
15361: 'ar', #Arabic - Bahrain
3073: 'ar', #Arabic - Egypt
2049: 'ar', #Arabic - Iraq
11265: 'ar', #Arabic - Jordan
13313: 'ar', #Arabic - Kuwait
12289: 'ar', #Arabic - Lebanon
4097: 'ar', #Arabic - Libya
6145: 'ar', #Arabic - Morocco
8193: 'ar', #Arabic - Oman
16385: 'ar', #Arabic - Qatar
10241: 'ar', #Arabic - Syria
7169: 'ar', #Arabic - Tunisia
14337: 'ar', #Arabic - U.A.E.
9217: 'ar', #Arabic - Yemen
1067: 'hy', #Armenian - Armenia
1101: 'as', #Assamese
2092: 'az', #Azeri (Cyrillic)
1068: 'az', #Azeri (Latin)
1069: 'eu', #Basque
1059: 'be', #Belarusian
1093: 'bn', #Bengali (India)
2117: 'bn', #Bengali (Bangladesh)
5146: 'bs', #Bosnian (Bosnia/Herzegovina)
1026: 'bg', #Bulgarian
1109: 'my', #Burmese
1027: 'ca', #Catalan
1116: 'chr', #Cherokee - United States
2052: 'zh', #Chinese - People's Republic of China
4100: 'zh', #Chinese - Singapore
1028: 'zh', #Chinese - Taiwan
3076: 'zh', #Chinese - Hong Kong SAR
5124: 'zh', #Chinese - Macao SAR
1050: 'hr', #Croatian
4122: 'hr', #Croatian (Bosnia/Herzegovina)
1029: 'cs', #Czech
1030: 'da', #Danish
1125: 'dv', #Divehi
1043: 'nl', #Dutch - Netherlands
2067: 'nl', #Dutch - Belgium
1126: 'bin', #Edo
1033: 'en', #English - United States
2057: 'en', #English - United Kingdom
3081: 'en', #English - Australia
10249: 'en', #English - Belize
4105: 'en', #English - Canada
9225: 'en', #English - Caribbean
15369: 'en', #English - Hong Kong SAR
16393: 'en', #English - India
14345: 'en', #English - Indonesia
6153: 'en', #English - Ireland
8201: 'en', #English - Jamaica
17417: 'en', #English - Malaysia
5129: 'en', #English - New Zealand
13321: 'en', #English - Philippines
18441: 'en', #English - Singapore
7177: 'en', #English - South Africa
11273: 'en', #English - Trinidad
12297: 'en', #English - Zimbabwe
1061: 'et', #Estonian
1080: 'fo', #Faroese
1065: None, #TODO: Farsi
1124: 'fil', #Filipino
1035: 'fi', #Finnish
1036: 'fr', #French - France
2060: 'fr', #French - Belgium
11276: 'fr', #French - Cameroon
3084: 'fr', #French - Canada
9228: 'fr', #French - Democratic Rep. of Congo
12300: 'fr', #French - Cote d'Ivoire
15372: 'fr', #French - Haiti
5132: 'fr', #French - Luxembourg
13324: 'fr', #French - Mali
6156: 'fr', #French - Monaco
14348: 'fr', #French - Morocco
58380: 'fr', #French - North Africa
8204: 'fr', #French - Reunion
10252: 'fr', #French - Senegal
4108: 'fr', #French - Switzerland
7180: 'fr', #French - West Indies
1122: 'fy', #Frisian - Netherlands
1127: None, #TODO: Fulfulde - Nigeria
1071: 'mk', #FYRO Macedonian
2108: 'ga', #Gaelic (Ireland)
1084: 'gd', #Gaelic (Scotland)
1110: 'gl', #Galician
1079: 'ka', #Georgian
1031: 'de', #German - Germany
3079: 'de', #German - Austria
5127: 'de', #German - Liechtenstein
4103: 'de', #German - Luxembourg
2055: 'de', #German - Switzerland
1032: 'el', #Greek
1140: 'gn', #Guarani - Paraguay
1095: 'gu', #Gujarati
1128: 'ha', #Hausa - Nigeria
1141: 'haw', #Hawaiian - United States
1037: 'he', #Hebrew
1081: 'hi', #Hindi
1038: 'hu', #Hungarian
1129: None, #TODO: Ibibio - Nigeria
1039: 'is', #Icelandic
1136: 'ig', #Igbo - Nigeria
1057: 'id', #Indonesian
1117: 'iu', #Inuktitut
1040: 'it', #Italian - Italy
2064: 'it', #Italian - Switzerland
1041: 'ja', #Japanese
1099: 'kn', #Kannada
1137: 'kr', #Kanuri - Nigeria
2144: 'ks', #Kashmiri
1120: 'ks', #Kashmiri (Arabic)
1087: 'kk', #Kazakh
1107: 'km', #Khmer
1111: 'kok', #Konkani
1042: 'ko', #Korean
1088: 'ky', #Kyrgyz (Cyrillic)
1108: 'lo', #Lao
1142: 'la', #Latin
1062: 'lv', #Latvian
1063: 'lt', #Lithuanian
1086: 'ms', #Malay - Malaysia
2110: 'ms', #Malay - Brunei Darussalam
1100: 'ml', #Malayalam
1082: 'mt', #Maltese
1112: 'mni', #Manipuri
1153: 'mi', #Maori - New Zealand
1102: 'mr', #Marathi
1104: 'mn', #Mongolian (Cyrillic)
2128: 'mn', #Mongolian (Mongolian)
1121: 'ne', #Nepali
2145: 'ne', #Nepali - India
1044: 'no', #Norwegian (Bokmￃﾥl)
2068: 'no', #Norwegian (Nynorsk)
1096: 'or', #Oriya
1138: 'om', #Oromo
1145: 'pap', #Papiamentu
1123: 'ps', #Pashto
1045: 'pl', #Polish
1046: 'pt', #Portuguese - Brazil
2070: 'pt', #Portuguese - Portugal
1094: 'pa', #Punjabi
2118: 'pa', #Punjabi (Pakistan)
1131: 'qu', #Quecha - Bolivia
2155: 'qu', #Quecha - Ecuador
3179: 'qu', #Quecha - Peru
1047: 'rm', #Rhaeto-Romanic
1048: 'ro', #Romanian
2072: 'ro', #Romanian - Moldava
1049: 'ru', #Russian
2073: 'ru', #Russian - Moldava
1083: 'se', #Sami (Lappish)
1103: 'sa', #Sanskrit
1132: 'nso', #Sepedi
3098: 'sr', #Serbian (Cyrillic)
2074: 'sr', #Serbian (Latin)
1113: 'sd', #Sindhi - India
2137: 'sd', #Sindhi - Pakistan
1115: 'si', #Sinhalese - Sri Lanka
1051: 'sk', #Slovak
1060: 'sl', #Slovenian
1143: 'so', #Somali
1070: 'wen', #Sorbian
3082: 'es', #Spanish - Spain (Modern Sort)
1034: 'es', #Spanish - Spain (Traditional Sort)
11274: 'es', #Spanish - Argentina
16394: 'es', #Spanish - Bolivia
13322: 'es', #Spanish - Chile
9226: 'es', #Spanish - Colombia
5130: 'es', #Spanish - Costa Rica
7178: 'es', #Spanish - Dominican Republic
12298: 'es', #Spanish - Ecuador
17418: 'es', #Spanish - El Salvador
4106: 'es', #Spanish - Guatemala
18442: 'es', #Spanish - Honduras
58378: 'es', #Spanish - Latin America
2058: 'es', #Spanish - Mexico
19466: 'es', #Spanish - Nicaragua
6154: 'es', #Spanish - Panama
15370: 'es', #Spanish - Paraguay
10250: 'es', #Spanish - Peru
20490: 'es', #Spanish - Puerto Rico
21514: 'es', #Spanish - United States
14346: 'es', #Spanish - Uruguay
8202: 'es', #Spanish - Venezuela
1072: None, #TODO: Sutu
1089: 'sw', #Swahili
1053: 'sv', #Swedish
2077: 'sv', #Swedish - Finland
1114: 'syr', #Syriac
1064: 'tg', #Tajik
1119: None, #TODO: Tamazight (Arabic)
2143: None, #TODO: Tamazight (Latin)
1097: 'ta', #Tamil
1092: 'tt', #Tatar
1098: 'te', #Telugu
1054: 'th', #Thai
2129: 'bo', #Tibetan - Bhutan
1105: 'bo', #Tibetan - People's Republic of China
2163: 'ti', #Tigrigna - Eritrea
1139: 'ti', #Tigrigna - Ethiopia
1073: 'ts', #Tsonga
1074: 'tn', #Tswana
1055: 'tr', #Turkish
1090: 'tk', #Turkmen
1152: 'ug', #Uighur - China
1058: 'uk', #Ukrainian
1056: 'ur', #Urdu
2080: 'ur', #Urdu - India
2115: 'uz', #Uzbek (Cyrillic)
1091: 'uz', #Uzbek (Latin)
1075: 've', #Venda
1066: 'vi', #Vietnamese
1106: 'cy', #Welsh
1076: 'xh', #Xhosa
1144: 'ii', #Yi
1085: 'yi', #Yiddish
1130: 'yo', #Yoruba
1077: 'zu'} #Zulu
return mapping[lcid]
def _getscreenlanguage():
'''
:returns: the ISO 639-x language code for this session.
If the LANGUAGE environment variable is set, it's value overrides the
screen language detection. Otherwise the screen language is determined by
the currently selected Microsoft Windows MUI language pack or the Microsoft
Windows installation language.
Works on Microsoft Windows 2000 and up.
'''
if sys.platform == 'win32' or sys.platform == 'nt':
# Start with nothing
lang = None
# Check the LANGUAGE environment variable
lang = os.getenv('LANGUAGE')
if lang is None:
# Start with nothing
lcid = None
try:
from ctypes import windll
lcid = windll.kernel32.GetUserDefaultUILanguage()
except:
logger.debug('Failed to get current screen language with \'GetUserDefaultUILanguage\'')
finally:
if lcid is None:
lang = 'C'
else:
lang = _isofromlcid(lcid)
logger.debug('Windows screen language is \'%s\' (lcid %s)' % (lang, lcid))
return lang
def _putenv(name, value):
'''
:param name: environment variable name
:param value: environment variable value
This function ensures that changes to an environment variable are applied
to each copy of the environment variables used by a process. Starting from
Python 2.4, os.environ changes only apply to the copy Python keeps (os.environ)
and are no longer automatically applied to the other copies for the process.
On Microsoft Windows, each process has multiple copies of the environment
variables, one managed by the OS and one managed by the C library. We also
need to take care of the fact that the C library used by Python is not
necessarily the same as the C library used by pygtk and friends. This because
the latest releases of pygtk and friends are built with mingw32 and are thus
linked against msvcrt.dll. The official gtk+ binaries have always been built
in this way.
'''
if sys.platform == 'win32' or sys.platform == 'nt':
from ctypes import windll
from ctypes import cdll
from ctypes.util import find_msvcrt
# Update Python's copy of the environment variables
os.environ[name] = value
# Update the copy maintained by Windows (so SysInternals Process Explorer sees it)
try:
result = windll.kernel32.SetEnvironmentVariableW(name, value)
if result == 0: raise Warning
except Exception:
logger.debug('Failed to set environment variable \'%s\' (\'kernel32.SetEnvironmentVariableW\')' % name)
else:
logger.debug('Set environment variable \'%s\' to \'%s\' (\'kernel32.SetEnvironmentVariableW\')' % (name, value))
# Update the copy maintained by msvcrt (used by gtk+ runtime)
try:
result = cdll.msvcrt._putenv('%s=%s' % (name, value))
if result !=0: raise Warning
except Exception:
logger.debug('Failed to set environment variable \'%s\' (\'msvcrt._putenv\')' % name)
else:
logger.debug('Set environment variable \'%s\' to \'%s\' (\'msvcrt._putenv\')' % (name, value))
# Update the copy maintained by whatever c runtime is used by Python
try:
msvcrt = find_msvcrt()
msvcrtname = str(msvcrt).split('.')[0] if '.' in msvcrt else str(msvcrt)
result = cdll.LoadLibrary(msvcrt)._putenv('%s=%s' % (name, value))
if result != 0: raise Warning
except Exception:
logger.debug('Failed to set environment variable \'%s\' (\'<unknown>._putenv\')' % (name))
else:
logger.debug('Set environment variable \'%s\' to \'%s\' (\'%s._putenv\')' % (name, value, msvcrtname))
def _dugettext(domain, message):
'''
:param domain: translation domain
:param message: message to translate
:returns: the translated message
Unicode version of :func:`gettext.dgettext`.
'''
try:
t = gettext.translation(domain, gettext._localedirs.get(domain, None),
codeset=gettext._localecodesets.get(domain))
except IOError:
return message
else:
return t.ugettext(message)
def _install(domain, localedir, asglobal=False, libintl='intl'):
'''
:param domain: translation domain
:param localedir: locale directory
:param asglobal: if True, installs the function _() in Python’s builtin namespace. Default is False
Private function doing all the work for the :func:`elib.intl.install` and
:func:`elib.intl.install_module` functions.
'''
# prep locale system
if asglobal:
locale.setlocale(locale.LC_ALL, '')
# on windows systems, set the LANGUAGE environment variable
if sys.platform == 'win32' or sys.platform == 'nt':
_putenv('LANGUAGE', _getscreenlanguage())
# The locale module on Max OS X lacks bindtextdomain so we specifically
# test on linux2 here. See commit 4ae8b26fd569382ab66a9e844daa0e01de409ceb
if sys.platform == 'linux2':
locale.bindtextdomain(domain, localedir)
locale.bind_textdomain_codeset(domain, 'UTF-8')
locale.textdomain(domain)
# initialize Python's gettext interface
gettext.bindtextdomain(domain, localedir)
gettext.bind_textdomain_codeset(domain, 'UTF-8')
if asglobal:
gettext.textdomain(domain)
# on windows systems, initialize libintl
if libintl and (sys.platform == 'win32' or sys.platform == 'nt'):
from ctypes import cdll
libintl = cdll.LoadLibrary(libintl)
libintl.bindtextdomain(domain, localedir)
libintl.bind_textdomain_codeset(domain, 'UTF-8')
if asglobal:
libintl.textdomain(domain)
del libintl
def install(domain, localedir, libintl='intl'):
'''
:param domain: translation domain
:param localedir: locale directory
Installs the function _() in Python’s builtin namespace, based on
domain and localedir. Codeset is always UTF-8.
As seen below, you usually mark the strings in your application that are
candidates for translation, by wrapping them in a call to the _() function,
like this:
.. sourcecode:: python
import elib.intl
elib.intl.install('myapplication', '/path/to/usr/share/locale')
print _('This string will be translated.')
Note that this is only one way, albeit the most convenient way,
to make the _() function available to your application. Because it affects
the entire application globally, and specifically Python’s built-in
namespace, localized modules should never install _(). Instead, you should
use :func:`elib.intl.install_module` to make _() available to your module.
'''
_install(domain, localedir, True, libintl=libintl)
gettext.install(domain, localedir)
def install_module(domain, localedir):
'''
:param domain: translation domain
:param localedir: locale directory
:returns: an anonymous function object, based on domain and localedir.
Codeset is always UTF-8.
You may find this function usefull when writing localized modules.
Use this code to make _() available to your module:
.. sourcecode:: python
import elib.intl
_ = elib.intl.install_module('mymodule', '/path/to/usr/share/locale')
print _('This string will be translated.')
When writing a package, you can usually do this in the package's __init__.py
file and import the _() function from the package namespace as needed.
'''
_install(domain, localedir, False)
return lambda message: _dugettext(domain, message)
