Module core

Returns the system default locale for a given category, based on environment variables.

>>> for name in ['LANGUAGE', 'LC_ALL', 'LC_CTYPE']:
...     os.environ[name] = ''
>>> os.environ['LANG'] = 'fr_FR.UTF-8'
>>> default_locale('LC_MESSAGES')
'fr_FR'

The "C" or "POSIX" pseudo-locales are treated as aliases for the "en_US_POSIX" locale:

>>> os.environ['LC_MESSAGES'] = 'POSIX'
>>> default_locale('LC_MESSAGES')
'en_US_POSIX'

Parameters:

• category - one of the LC_XXX environment variable names
• aliases - a dictionary of aliases for locale identifiers

Returns: str

the value of the variable, or any of the fallbacks (LANGUAGE, LC_ALL, LC_CTYPE, and LANG)

Find the best match between available and requested locale strings.

>>> negotiate_locale(['de_DE', 'en_US'], ['de_DE', 'de_AT'])
'de_DE'
>>> negotiate_locale(['de_DE', 'en_US'], ['en', 'de'])
'de'

Case is ignored by the algorithm, the result uses the case of the preferred locale identifier:

>>> negotiate_locale(['de_DE', 'en_US'], ['de_de', 'de_at'])
'de_DE'

>>> negotiate_locale(['de_DE', 'en_US'], ['de_de', 'de_at'])
'de_DE'

By default, some web browsers unfortunately do not include the territory in the locale identifier for many locales, and some don't even allow the user to easily add the territory. So while you may prefer using qualified locale identifiers in your web-application, they would not normally match the language-only locale sent by such browsers. To workaround that, this function uses a default mapping of commonly used langauge-only locale identifiers to identifiers including the territory:

>>> negotiate_locale(['ja', 'en_US'], ['ja_JP', 'en_US'])
'ja_JP'

Some browsers even use an incorrect or outdated language code, such as "no" for Norwegian, where the correct locale identifier would actually be "nb_NO" (Bokmål) or "nn_NO" (Nynorsk). The aliases are intended to take care of such cases, too:

>>> negotiate_locale(['no', 'sv'], ['nb_NO', 'sv_SE'])
'nb_NO'

You can override this default mapping by passing a different aliases dictionary to this function, or you can bypass the behavior althogher by setting the aliases parameter to None.

Parameters:

• preferred - the list of locale strings preferred by the user
• available - the list of locale strings available
• sep - character that separates the different parts of the locale strings
• aliases - a dictionary of aliases for locale identifiers

Returns: str

the locale identifier for the best match, or None if no match was found

Parse a locale identifier into a tuple of the form:

``(language, territory, script, variant)``

>>> parse_locale('zh_CN')
('zh', 'CN', None, None)
>>> parse_locale('zh_Hans_CN')
('zh', 'CN', 'Hans', None)

The default component separator is "_", but a different separator can be specified using the sep parameter:

>>> parse_locale('zh-CN', sep='-')
('zh', 'CN', None, None)

If the identifier cannot be parsed into a locale, a ValueError exception is raised:

>>> parse_locale('not_a_LOCALE_String')
Traceback (most recent call last):
  ...
ValueError: 'not_a_LOCALE_String' is not a valid locale identifier

Encoding information and locale modifiers are removed from the identifier:

>>> parse_locale('it_IT@euro')
('it', 'IT', None, None)
>>> parse_locale('en_US.UTF-8')
('en', 'US', None, None)
>>> parse_locale('de_DE.iso885915@euro')
('de', 'DE', None, None)

Parameters:

• identifier - the locale identifier string
• sep - character that separates the different components of the locale identifier

Returns: tuple

the (language, territory, script, variant) tuple

Raises:

• ValueError - if the string does not appear to be a valid locale identifier