Lib/locale.py

   1 """ Locale support.
   2
   3     The module provides low-level access to the C lib's locale APIs
   4     and adds high level number formatting APIs as well as a locale
   5     aliasing engine to complement these.
   6
   7     The aliasing engine includes support for many commonly used locale
   8     names and maps them to values suitable for passing to the C lib's
   9     setlocale() function. It also includes default encodings for all
  10     supported locale names.
  11
  12 """
  13
  14 import sys
  15
  16 # Try importing the _locale module.
  17 #
  18 # If this fails, fall back on a basic 'C' locale emulation.
  19
  20 # Yuck:  LC_MESSAGES is non-standard:  can't tell whether it exists before
  21 # trying the import.  So __all__ is also fiddled at the end of the file.
  22 __all__ = ["setlocale","Error","localeconv","strcoll","strxfrm",
  23            "format","str","atof","atoi","LC_CTYPE","LC_COLLATE",
  24            "LC_TIME","LC_MONETARY","LC_NUMERIC", "LC_ALL","CHAR_MAX"]
  25
  26 try:
  27
  28     from _locale import *
  29
  30 except ImportError:
  31
  32     # Locale emulation
  33
  34     CHAR_MAX = 127
  35     LC_ALL = 6
  36     LC_COLLATE = 3
  37     LC_CTYPE = 0
  38     LC_MESSAGES = 5
  39     LC_MONETARY = 4
  40     LC_NUMERIC = 1
  41     LC_TIME = 2
  42     Error = ValueError
  43
  44     def localeconv():
  45         """ localeconv() -> dict.
  46             Returns numeric and monetary locale-specific parameters.
  47         """
  48         # 'C' locale default values
  49         return {'grouping': [127],
  50                 'currency_symbol': '',
  51                 'n_sign_posn': 127,
  52                 'p_cs_precedes': 127,
  53                 'n_cs_precedes': 127,
  54                 'mon_grouping': [],
  55                 'n_sep_by_space': 127,
  56                 'decimal_point': '.',
  57                 'negative_sign': '',
  58                 'positive_sign': '',
  59                 'p_sep_by_space': 127,
  60                 'int_curr_symbol': '',
  61                 'p_sign_posn': 127,
  62                 'thousands_sep': '',
  63                 'mon_thousands_sep': '',
  64                 'frac_digits': 127,
  65                 'mon_decimal_point': '',
  66                 'int_frac_digits': 127}
  67
  68     def setlocale(category, value=None):
  69         """ setlocale(integer,string=None) -> string.
  70             Activates/queries locale processing.
  71         """
  72         if value is not None and \
  73            value is not 'C':
  74             raise Error, '_locale emulation only supports "C" locale'
  75         return 'C'
  76
  77     def strcoll(a,b):
  78         """ strcoll(string,string) -> int.
  79             Compares two strings according to the locale.
  80         """
  81         return cmp(a,b)
  82
  83     def strxfrm(s):
  84         """ strxfrm(string) -> string.
  85             Returns a string that behaves for cmp locale-aware.
  86         """
  87         return s
  88
  89 ### Number formatting APIs
  90
  91 # Author: Martin von Loewis
  92
  93 #perform the grouping from right to left
  94 def _group(s):
  95     conv=localeconv()
  96     grouping=conv['grouping']
  97     if not grouping:return s
  98     result=""
  99     while s and grouping:
 100         # if grouping is -1, we are done
 101         if grouping[0]==CHAR_MAX:
 102             break
 103         # 0: re-use last group ad infinitum
 104         elif grouping[0]!=0:
 105             #process last group
 106             group=grouping[0]
 107             grouping=grouping[1:]
 108         if result:
 109             result=s[-group:]+conv['thousands_sep']+result
 110         else:
 111             result=s[-group:]
 112         s=s[:-group]
 113     if not result:
 114         return s
 115     if s:
 116         result=s+conv['thousands_sep']+result
 117     return result
 118
 119 def format(f,val,grouping=0):
 120     """Formats a value in the same way that the % formatting would use,
 121     but takes the current locale into account.
 122     Grouping is applied if the third parameter is true."""
 123     result = f % abs(val)
 124     fields = result.split(".")
 125     if grouping:
 126         fields[0]=_group(fields[0])
 127     if len(fields)==2:
 128         res = fields[0]+localeconv()['decimal_point']+fields[1]
 129     elif len(fields)==1:
 130         res = fields[0]
 131     else:
 132         raise Error, "Too many decimal points in result string"
 133
 134     if val < 0:
 135         return '-'+res
 136     else:
 137         return res
 138
 139 def str(val):
 140     """Convert float to integer, taking the locale into account."""
 141     return format("%.12g",val)
 142
 143 def atof(str,func=float):
 144     "Parses a string as a float according to the locale settings."
 145     #First, get rid of the grouping
 146     ts = localeconv()['thousands_sep']
 147     if ts:
 148         s=str.split(ts)
 149         str="".join(s)
 150     #next, replace the decimal point with a dot
 151     dd = localeconv()['decimal_point']
 152     if dd:
 153         s=str.split(dd)
 154         str='.'.join(s)
 155     #finally, parse the string
 156     return func(str)
 157
 158 def atoi(str):
 159     "Converts a string to an integer according to the locale settings."
 160     return atof(str, int)
 161
 162 def _test():
 163     setlocale(LC_ALL, "")
 164     #do grouping
 165     s1=format("%d", 123456789,1)
 166     print s1, "is", atoi(s1)
 167     #standard formatting
 168     s1=str(3.14)
 169     print s1, "is", atof(s1)
 170
 171 ### Locale name aliasing engine
 172
 173 # Author: Marc-Andre Lemburg, mal@lemburg.com
 174 # Various tweaks by Fredrik Lundh <effbot@telia.com>
 175
 176 # store away the low-level version of setlocale (it's
 177 # overridden below)
 178 _setlocale = setlocale
 179
 180 def normalize(localename):
 181
 182     """ Returns a normalized locale code for the given locale
 183         name.
 184
 185         The returned locale code is formatted for use with
 186         setlocale().
 187
 188         If normalization fails, the original name is returned
 189         unchanged.
 190
 191         If the given encoding is not known, the function defaults to
 192         the default encoding for the locale code just like setlocale()
 193         does.
 194
 195     """
 196     # Normalize the locale name and extract the encoding
 197     fullname = localename.lower()
 198     if ':' in fullname:
 199         # ':' is sometimes used as encoding delimiter.
 200         fullname = fullname.replace(':', '.')
 201     if '.' in fullname:
 202         langname, encoding = fullname.split('.')[:2]
 203         fullname = langname + '.' + encoding
 204     else:
 205         langname = fullname
 206         encoding = ''
 207
 208     # First lookup: fullname (possibly with encoding)
 209     code = locale_alias.get(fullname, None)
 210     if code is not None:
 211         return code
 212
 213     # Second try: langname (without encoding)
 214     code = locale_alias.get(langname, None)
 215     if code is not None:
 216         if '.' in code:
 217             langname, defenc = code.split('.')
 218         else:
 219             langname = code
 220             defenc = ''
 221         if encoding:
 222             encoding = encoding_alias.get(encoding, encoding)
 223         else:
 224             encoding = defenc
 225         if encoding:
 226             return langname + '.' + encoding
 227         else:
 228             return langname
 229
 230     else:
 231         return localename
 232
 233 def _parse_localename(localename):
 234
 235     """ Parses the locale code for localename and returns the
 236         result as tuple (language code, encoding).
 237
 238         The localename is normalized and passed through the locale
 239         alias engine. A ValueError is raised in case the locale name
 240         cannot be parsed.
 241
 242         The language code corresponds to RFC 1766.  code and encoding
 243         can be None in case the values cannot be determined or are
 244         unknown to this implementation.
 245
 246     """
 247     code = normalize(localename)
 248     if '.' in code:
 249         return code.split('.')[:2]
 250     elif code == 'C':
 251         return None, None
 252     else:
 253         raise ValueError, 'unknown locale: %s' % localename
 254     return l
 255
 256 def _build_localename(localetuple):
 257
 258     """ Builds a locale code from the given tuple (language code,
 259         encoding).
 260
 261         No aliasing or normalizing takes place.
 262
 263     """
 264     language, encoding = localetuple
 265     if language is None:
 266         language = 'C'
 267     if encoding is None:
 268         return language
 269     else:
 270         return language + '.' + encoding
 271
 272 def getdefaultlocale(envvars=('LANGUAGE', 'LC_ALL', 'LC_CTYPE', 'LANG')):
 273
 274     """ Tries to determine the default locale settings and returns
 275         them as tuple (language code, encoding).
 276
 277         According to POSIX, a program which has not called
 278         setlocale(LC_ALL, "") runs using the portable 'C' locale.
 279         Calling setlocale(LC_ALL, "") lets it use the default locale as
 280         defined by the LANG variable. Since we don't want to interfere
 281         with the current locale setting we thus emulate the behavior
 282         in the way described above.
 283
 284         To maintain compatibility with other platforms, not only the
 285         LANG variable is tested, but a list of variables given as
 286         envvars parameter. The first found to be defined will be
 287         used. envvars defaults to the search path used in GNU gettext;
 288         it must always contain the variable name 'LANG'.
 289
 290         Except for the code 'C', the language code corresponds to RFC
 291         1766.  code and encoding can be None in case the values cannot
 292         be determined.
 293
 294     """
 295
 296     try:
 297         # check if it's supported by the _locale module
 298         import _locale
 299         code, encoding = _locale._getdefaultlocale()
 300     except (ImportError, AttributeError):
 301         pass
 302     else:
 303         # make sure the code/encoding values are valid
 304         if sys.platform == "win32" and code and code[:2] == "0x":
 305             # map windows language identifier to language name
 306             code = windows_locale.get(int(code, 0))
 307         # ...add other platform-specific processing here, if
 308         # necessary...
 309         return code, encoding
 310
 311     # fall back on POSIX behaviour
 312     import os
 313     lookup = os.environ.get
 314     for variable in envvars:
 315         localename = lookup(variable,None)
 316         if localename is not None:
 317             break
 318     else:
 319         localename = 'C'
 320     return _parse_localename(localename)
 321
 322
 323 def getlocale(category=LC_CTYPE):
 324
 325     """ Returns the current setting for the given locale category as
 326         tuple (language code, encoding).
 327
 328         category may be one of the LC_* value except LC_ALL. It
 329         defaults to LC_CTYPE.
 330
 331         Except for the code 'C', the language code corresponds to RFC
 332         1766.  code and encoding can be None in case the values cannot
 333         be determined.
 334
 335     """
 336     localename = _setlocale(category)
 337     if category == LC_ALL and ';' in localename:
 338         raise TypeError, 'category LC_ALL is not supported'
 339     return _parse_localename(localename)
 340
 341 def setlocale(category, locale=None):
 342
 343     """ Set the locale for the given category.  The locale can be
 344         a string, a locale tuple (language code, encoding), or None.
 345
 346         Locale tuples are converted to strings the locale aliasing
 347         engine.  Locale strings are passed directly to the C lib.
 348
 349         category may be given as one of the LC_* values.
 350
 351     """
 352     if locale and type(locale) is not type(""):
 353         # convert to string
 354         locale = normalize(_build_localename(locale))
 355     return _setlocale(category, locale)
 356
 357 def resetlocale(category=LC_ALL):
 358
 359     """ Sets the locale for category to the default setting.
 360
 361         The default setting is determined by calling
 362         getdefaultlocale(). category defaults to LC_ALL.
 363
 364     """
 365     _setlocale(category, _build_localename(getdefaultlocale()))
 366
 367 ### Database
 368 #
 369 # The following data was extracted from the locale.alias file which
 370 # comes with X11 and then hand edited removing the explicit encoding
 371 # definitions and adding some more aliases. The file is usually
 372 # available as /usr/lib/X11/locale/locale.alias.
 373 #
 374
 375 #
 376 # The encoding_alias table maps lowercase encoding alias names to C
 377 # locale encoding names (case-sensitive).
 378 #
 379 encoding_alias = {
 380         '437':                          'C',
 381         'c':                            'C',
 382         'iso8859':                      'ISO8859-1',
 383         '8859':                         'ISO8859-1',
 384         '88591':                        'ISO8859-1',
 385         'ascii':                        'ISO8859-1',
 386         'en':                           'ISO8859-1',
 387         'iso88591':                     'ISO8859-1',
 388         'iso_8859-1':                   'ISO8859-1',
 389         '885915':                       'ISO8859-15',
 390         'iso885915':                    'ISO8859-15',
 391         'iso_8859-15':                  'ISO8859-15',
 392         'iso8859-2':                    'ISO8859-2',
 393         'iso88592':                     'ISO8859-2',
 394         'iso_8859-2':                   'ISO8859-2',
 395         'iso88595':                     'ISO8859-5',
 396         'iso88596':                     'ISO8859-6',
 397         'iso88597':                     'ISO8859-7',
 398         'iso88598':                     'ISO8859-8',
 399         'iso88599':                     'ISO8859-9',
 400         'iso-2022-jp':                  'JIS7',
 401         'jis':                          'JIS7',
 402         'jis7':                         'JIS7',
 403         'sjis':                         'SJIS',
 404         'tis620':                       'TACTIS',
 405         'ajec':                         'eucJP',
 406         'eucjp':                        'eucJP',
 407         'ujis':                         'eucJP',
 408         'utf-8':                        'utf',
 409         'utf8':                         'utf',
 410         'utf8@ucs4':                    'utf',
 411 }
 412
 413 #
 414 # The locale_alias table maps lowercase alias names to C locale names
 415 # (case-sensitive). Encodings are always separated from the locale
 416 # name using a dot ('.'); they should only be given in case the
 417 # language name is needed to interpret the given encoding alias
 418 # correctly (CJK codes often have this need).
 419 #
 420 locale_alias = {
 421         'american':                      'en_US.ISO8859-1',
 422         'ar':                            'ar_AA.ISO8859-6',
 423         'ar_aa':                         'ar_AA.ISO8859-6',
 424         'ar_sa':                         'ar_SA.ISO8859-6',
 425         'arabic':                        'ar_AA.ISO8859-6',
 426         'bg':                            'bg_BG.ISO8859-5',
 427         'bg_bg':                         'bg_BG.ISO8859-5',
 428         'bulgarian':                     'bg_BG.ISO8859-5',
 429         'c-french':                      'fr_CA.ISO8859-1',
 430         'c':                             'C',
 431         'c_c':                           'C',
 432         'cextend':                       'en_US.ISO8859-1',
 433         'chinese-s':                     'zh_CN.eucCN',
 434         'chinese-t':                     'zh_TW.eucTW',
 435         'croatian':                      'hr_HR.ISO8859-2',
 436         'cs':                            'cs_CZ.ISO8859-2',
 437         'cs_cs':                         'cs_CZ.ISO8859-2',
 438         'cs_cz':                         'cs_CZ.ISO8859-2',
 439         'cz':                            'cz_CZ.ISO8859-2',
 440         'cz_cz':                         'cz_CZ.ISO8859-2',
 441         'czech':                         'cs_CS.ISO8859-2',
 442         'da':                            'da_DK.ISO8859-1',
 443         'da_dk':                         'da_DK.ISO8859-1',
 444         'danish':                        'da_DK.ISO8859-1',
 445         'de':                            'de_DE.ISO8859-1',
 446         'de_at':                         'de_AT.ISO8859-1',
 447         'de_ch':                         'de_CH.ISO8859-1',
 448         'de_de':                         'de_DE.ISO8859-1',
 449         'dutch':                         'nl_BE.ISO8859-1',
 450         'ee':                            'ee_EE.ISO8859-4',
 451         'el':                            'el_GR.ISO8859-7',
 452         'el_gr':                         'el_GR.ISO8859-7',
 453         'en':                            'en_US.ISO8859-1',
 454         'en_au':                         'en_AU.ISO8859-1',
 455         'en_ca':                         'en_CA.ISO8859-1',
 456         'en_gb':                         'en_GB.ISO8859-1',
 457         'en_ie':                         'en_IE.ISO8859-1',
 458         'en_nz':                         'en_NZ.ISO8859-1',
 459         'en_uk':                         'en_GB.ISO8859-1',
 460         'en_us':                         'en_US.ISO8859-1',
 461         'eng_gb':                        'en_GB.ISO8859-1',
 462         'english':                       'en_EN.ISO8859-1',
 463         'english_uk':                    'en_GB.ISO8859-1',
 464         'english_united-states':         'en_US.ISO8859-1',
 465         'english_us':                    'en_US.ISO8859-1',
 466         'es':                            'es_ES.ISO8859-1',
 467         'es_ar':                         'es_AR.ISO8859-1',
 468         'es_bo':                         'es_BO.ISO8859-1',
 469         'es_cl':                         'es_CL.ISO8859-1',
 470         'es_co':                         'es_CO.ISO8859-1',
 471         'es_cr':                         'es_CR.ISO8859-1',
 472         'es_ec':                         'es_EC.ISO8859-1',
 473         'es_es':                         'es_ES.ISO8859-1',
 474         'es_gt':                         'es_GT.ISO8859-1',
 475         'es_mx':                         'es_MX.ISO8859-1',
 476         'es_ni':                         'es_NI.ISO8859-1',
 477         'es_pa':                         'es_PA.ISO8859-1',
 478         'es_pe':                         'es_PE.ISO8859-1',
 479         'es_py':                         'es_PY.ISO8859-1',
 480         'es_sv':                         'es_SV.ISO8859-1',
 481         'es_uy':                         'es_UY.ISO8859-1',
 482         'es_ve':                         'es_VE.ISO8859-1',
 483         'et':                            'et_EE.ISO8859-4',
 484         'et_ee':                         'et_EE.ISO8859-4',
 485         'fi':                            'fi_FI.ISO8859-1',
 486         'fi_fi':                         'fi_FI.ISO8859-1',
 487         'finnish':                       'fi_FI.ISO8859-1',
 488         'fr':                            'fr_FR.ISO8859-1',
 489         'fr_be':                         'fr_BE.ISO8859-1',
 490         'fr_ca':                         'fr_CA.ISO8859-1',
 491         'fr_ch':                         'fr_CH.ISO8859-1',
 492         'fr_fr':                         'fr_FR.ISO8859-1',
 493         'fre_fr':                        'fr_FR.ISO8859-1',
 494         'french':                        'fr_FR.ISO8859-1',
 495         'french_france':                 'fr_FR.ISO8859-1',
 496         'ger_de':                        'de_DE.ISO8859-1',
 497         'german':                        'de_DE.ISO8859-1',
 498         'german_germany':                'de_DE.ISO8859-1',
 499         'greek':                         'el_GR.ISO8859-7',
 500         'hebrew':                        'iw_IL.ISO8859-8',
 501         'hr':                            'hr_HR.ISO8859-2',
 502         'hr_hr':                         'hr_HR.ISO8859-2',
 503         'hu':                            'hu_HU.ISO8859-2',
 504         'hu_hu':                         'hu_HU.ISO8859-2',
 505         'hungarian':                     'hu_HU.ISO8859-2',
 506         'icelandic':                     'is_IS.ISO8859-1',
 507         'id':                            'id_ID.ISO8859-1',
 508         'id_id':                         'id_ID.ISO8859-1',
 509         'is':                            'is_IS.ISO8859-1',
 510         'is_is':                         'is_IS.ISO8859-1',
 511         'iso-8859-1':                    'en_US.ISO8859-1',
 512         'iso-8859-15':                   'en_US.ISO8859-15',
 513         'iso8859-1':                     'en_US.ISO8859-1',
 514         'iso8859-15':                    'en_US.ISO8859-15',
 515         'iso_8859_1':                    'en_US.ISO8859-1',
 516         'iso_8859_15':                   'en_US.ISO8859-15',
 517         'it':                            'it_IT.ISO8859-1',
 518         'it_ch':                         'it_CH.ISO8859-1',
 519         'it_it':                         'it_IT.ISO8859-1',
 520         'italian':                       'it_IT.ISO8859-1',
 521         'iw':                            'iw_IL.ISO8859-8',
 522         'iw_il':                         'iw_IL.ISO8859-8',
 523         'ja':                            'ja_JP.eucJP',
 524         'ja.jis':                        'ja_JP.JIS7',
 525         'ja.sjis':                       'ja_JP.SJIS',
 526         'ja_jp':                         'ja_JP.eucJP',
 527         'ja_jp.ajec':                    'ja_JP.eucJP',
 528         'ja_jp.euc':                     'ja_JP.eucJP',
 529         'ja_jp.eucjp':                   'ja_JP.eucJP',
 530         'ja_jp.iso-2022-jp':             'ja_JP.JIS7',
 531         'ja_jp.jis':                     'ja_JP.JIS7',
 532         'ja_jp.jis7':                    'ja_JP.JIS7',
 533         'ja_jp.mscode':                  'ja_JP.SJIS',
 534         'ja_jp.sjis':                    'ja_JP.SJIS',
 535         'ja_jp.ujis':                    'ja_JP.eucJP',
 536         'japan':                         'ja_JP.eucJP',
 537         'japanese':                      'ja_JP.SJIS',
 538         'japanese-euc':                  'ja_JP.eucJP',
 539         'japanese.euc':                  'ja_JP.eucJP',
 540         'jp_jp':                         'ja_JP.eucJP',
 541         'ko':                            'ko_KR.eucKR',
 542         'ko_kr':                         'ko_KR.eucKR',
 543         'ko_kr.euc':                     'ko_KR.eucKR',
 544         'korean':                        'ko_KR.eucKR',
 545         'lt':                            'lt_LT.ISO8859-4',
 546         'lv':                            'lv_LV.ISO8859-4',
 547         'mk':                            'mk_MK.ISO8859-5',
 548         'mk_mk':                         'mk_MK.ISO8859-5',
 549         'nl':                            'nl_NL.ISO8859-1',
 550         'nl_be':                         'nl_BE.ISO8859-1',
 551         'nl_nl':                         'nl_NL.ISO8859-1',
 552         'no':                            'no_NO.ISO8859-1',
 553         'no_no':                         'no_NO.ISO8859-1',
 554         'norwegian':                     'no_NO.ISO8859-1',
 555         'pl':                            'pl_PL.ISO8859-2',
 556         'pl_pl':                         'pl_PL.ISO8859-2',
 557         'polish':                        'pl_PL.ISO8859-2',
 558         'portuguese':                    'pt_PT.ISO8859-1',
 559         'portuguese_brazil':             'pt_BR.ISO8859-1',
 560         'posix':                         'C',
 561         'posix-utf2':                    'C',
 562         'pt':                            'pt_PT.ISO8859-1',
 563         'pt_br':                         'pt_BR.ISO8859-1',
 564         'pt_pt':                         'pt_PT.ISO8859-1',
 565         'ro':                            'ro_RO.ISO8859-2',
 566         'ro_ro':                         'ro_RO.ISO8859-2',
 567         'ru':                            'ru_RU.ISO8859-5',
 568         'ru_ru':                         'ru_RU.ISO8859-5',
 569         'rumanian':                      'ro_RO.ISO8859-2',
 570         'russian':                       'ru_RU.ISO8859-5',
 571         'serbocroatian':                 'sh_YU.ISO8859-2',
 572         'sh':                            'sh_YU.ISO8859-2',
 573         'sh_hr':                         'sh_HR.ISO8859-2',
 574         'sh_sp':                         'sh_YU.ISO8859-2',
 575         'sh_yu':                         'sh_YU.ISO8859-2',
 576         'sk':                            'sk_SK.ISO8859-2',
 577         'sk_sk':                         'sk_SK.ISO8859-2',
 578         'sl':                            'sl_CS.ISO8859-2',
 579         'sl_cs':                         'sl_CS.ISO8859-2',
 580         'sl_si':                         'sl_SI.ISO8859-2',
 581         'slovak':                        'sk_SK.ISO8859-2',
 582         'slovene':                       'sl_CS.ISO8859-2',
 583         'sp':                            'sp_YU.ISO8859-5',
 584         'sp_yu':                         'sp_YU.ISO8859-5',
 585         'spanish':                       'es_ES.ISO8859-1',
 586         'spanish_spain':                 'es_ES.ISO8859-1',
 587         'sr_sp':                         'sr_SP.ISO8859-2',
 588         'sv':                            'sv_SE.ISO8859-1',
 589         'sv_se':                         'sv_SE.ISO8859-1',
 590         'swedish':                       'sv_SE.ISO8859-1',
 591         'th_th':                         'th_TH.TACTIS',
 592         'tr':                            'tr_TR.ISO8859-9',
 593         'tr_tr':                         'tr_TR.ISO8859-9',
 594         'turkish':                       'tr_TR.ISO8859-9',
 595         'univ':                          'en_US.utf',
 596         'universal':                     'en_US.utf',
 597         'zh':                            'zh_CN.eucCN',
 598         'zh_cn':                         'zh_CN.eucCN',
 599         'zh_cn.big5':                    'zh_TW.eucTW',
 600         'zh_cn.euc':                     'zh_CN.eucCN',
 601         'zh_tw':                         'zh_TW.eucTW',
 602         'zh_tw.euc':                     'zh_TW.eucTW',
 603 }
 604
 605 #
 606 # this maps windows language identifiers (as used on Windows 95 and
 607 # earlier) to locale strings.
 608 #
 609 # NOTE: this mapping is incomplete.  If your language is missing, send
 610 # a note with the missing language identifier and the suggested locale
 611 # code to Fredrik Lundh <effbot@telia.com>.  Thanks /F
 612
 613 windows_locale = {
 614     0x0404: "zh_TW", # Chinese (Taiwan)
 615     0x0804: "zh_CN", # Chinese (PRC)
 616     0x0406: "da_DK", # Danish
 617     0x0413: "nl_NL", # Dutch (Netherlands)
 618     0x0409: "en_US", # English (United States)
 619     0x0809: "en_UK", # English (United Kingdom)
 620     0x0c09: "en_AU", # English (Australian)
 621     0x1009: "en_CA", # English (Canadian)
 622     0x1409: "en_NZ", # English (New Zealand)
 623     0x1809: "en_IE", # English (Ireland)
 624     0x1c09: "en_ZA", # English (South Africa)
 625     0x040b: "fi_FI", # Finnish
 626     0x040c: "fr_FR", # French (Standard)
 627     0x080c: "fr_BE", # French (Belgian)
 628     0x0c0c: "fr_CA", # French (Canadian)
 629     0x100c: "fr_CH", # French (Switzerland)
 630     0x0407: "de_DE", # German (Standard)
 631     0x0408: "el_GR", # Greek
 632     0x040d: "iw_IL", # Hebrew
 633     0x040f: "is_IS", # Icelandic
 634     0x0410: "it_IT", # Italian (Standard)
 635     0x0411: "ja_JA", # Japanese
 636     0x0414: "no_NO", # Norwegian (Bokmal)
 637     0x0816: "pt_PT", # Portuguese (Standard)
 638     0x0c0a: "es_ES", # Spanish (Modern Sort)
 639     0x0441: "sw_KE", # Swahili (Kenya)
 640     0x041d: "sv_SE", # Swedish
 641     0x081d: "sv_FI", # Swedish (Finland)
 642     0x041f: "tr_TR", # Turkish
 643 }
 644
 645 def _print_locale():
 646
 647     """ Test function.
 648     """
 649     categories = {}
 650     def _init_categories(categories=categories):
 651         for k,v in globals().items():
 652             if k[:3] == 'LC_':
 653                 categories[k] = v
 654     _init_categories()
 655     del categories['LC_ALL']
 656
 657     print 'Locale defaults as determined by getdefaultlocale():'
 658     print '-'*72
 659     lang, enc = getdefaultlocale()
 660     print 'Language: ', lang or '(undefined)'
 661     print 'Encoding: ', enc or '(undefined)'
 662     print
 663
 664     print 'Locale settings on startup:'
 665     print '-'*72
 666     for name,category in categories.items():
 667         print name, '...'
 668         lang, enc = getlocale(category)
 669         print '   Language: ', lang or '(undefined)'
 670         print '   Encoding: ', enc or '(undefined)'
 671         print
 672
 673     print
 674     print 'Locale settings after calling resetlocale():'
 675     print '-'*72
 676     resetlocale()
 677     for name,category in categories.items():
 678         print name, '...'
 679         lang, enc = getlocale(category)
 680         print '   Language: ', lang or '(undefined)'
 681         print '   Encoding: ', enc or '(undefined)'
 682         print
 683
 684     try:
 685         setlocale(LC_ALL, "")
 686     except:
 687         print 'NOTE:'
 688         print 'setlocale(LC_ALL, "") does not support the default locale'
 689         print 'given in the OS environment variables.'
 690     else:
 691         print
 692         print 'Locale settings after calling setlocale(LC_ALL, ""):'
 693         print '-'*72
 694         for name,category in categories.items():
 695             print name, '...'
 696             lang, enc = getlocale(category)
 697             print '   Language: ', lang or '(undefined)'
 698             print '   Encoding: ', enc or '(undefined)'
 699             print
 700
 701 ###
 702
 703 try:
 704     LC_MESSAGES
 705 except:
 706     pass
 707 else:
 708     __all__.append("LC_MESSAGES")
 709
 710 if __name__=='__main__':
 711     print 'Locale aliasing:'
 712     print
 713     _print_locale()
 714     print
 715     print 'Number formatting:'
 716     print
 717     _test()