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 value != 'C':
  73             raise Error, '_locale emulation only supports "C" locale'
  74         return 'C'
  75
  76     def strcoll(a,b):
  77         """ strcoll(string,string) -> int.
  78             Compares two strings according to the locale.
  79         """
  80         return cmp(a,b)
  81
  82     def strxfrm(s):
  83         """ strxfrm(string) -> string.
  84             Returns a string that behaves for cmp locale-aware.
  85         """
  86         return s
  87
  88 ### Number formatting APIs
  89
  90 # Author: Martin von Loewis
  91
  92 #perform the grouping from right to left
  93 def _group(s):
  94     conv=localeconv()
  95     grouping=conv['grouping']
  96     if not grouping:return (s, 0)
  97     result=""
  98     seps = 0
  99     spaces = ""
 100     if s[-1] == ' ':
 101         sp = s.find(' ')
 102         spaces = s[sp:]
 103         s = s[:sp]
 104     while s and grouping:
 105         # if grouping is -1, we are done
 106         if grouping[0]==CHAR_MAX:
 107             break
 108         # 0: re-use last group ad infinitum
 109         elif grouping[0]!=0:
 110             #process last group
 111             group=grouping[0]
 112             grouping=grouping[1:]
 113         if result:
 114             result=s[-group:]+conv['thousands_sep']+result
 115             seps += 1
 116         else:
 117             result=s[-group:]
 118         s=s[:-group]
 119         if s and s[-1] not in "0123456789":
 120             # the leading string is only spaces and signs
 121             return s+result+spaces,seps
 122     if not result:
 123         return s+spaces,seps
 124     if s:
 125         result=s+conv['thousands_sep']+result
 126         seps += 1
 127     return result+spaces,seps
 128
 129 def format(f,val,grouping=0):
 130     """Formats a value in the same way that the % formatting would use,
 131     but takes the current locale into account.
 132     Grouping is applied if the third parameter is true."""
 133     result = f % val
 134     fields = result.split(".")
 135     seps = 0
 136     if grouping:
 137         fields[0],seps=_group(fields[0])
 138     if len(fields)==2:
 139         result = fields[0]+localeconv()['decimal_point']+fields[1]
 140     elif len(fields)==1:
 141         result = fields[0]
 142     else:
 143         raise Error, "Too many decimal points in result string"
 144
 145     while seps:
 146         # If the number was formatted for a specific width, then it
 147         # might have been filled with spaces to the left or right. If
 148         # so, kill as much spaces as there where separators.
 149         # Leading zeroes as fillers are not yet dealt with, as it is
 150         # not clear how they should interact with grouping.
 151         sp = result.find(" ")
 152         if sp==-1:break
 153         result = result[:sp]+result[sp+1:]
 154         seps -= 1
 155
 156     return result
 157
 158 def str(val):
 159     """Convert float to integer, taking the locale into account."""
 160     return format("%.12g",val)
 161
 162 def atof(str,func=float):
 163     "Parses a string as a float according to the locale settings."
 164     #First, get rid of the grouping
 165     ts = localeconv()['thousands_sep']
 166     if ts:
 167         s=str.split(ts)
 168         str="".join(s)
 169     #next, replace the decimal point with a dot
 170     dd = localeconv()['decimal_point']
 171     if dd:
 172         s=str.split(dd)
 173         str='.'.join(s)
 174     #finally, parse the string
 175     return func(str)
 176
 177 def atoi(str):
 178     "Converts a string to an integer according to the locale settings."
 179     return atof(str, int)
 180
 181 def _test():
 182     setlocale(LC_ALL, "")
 183     #do grouping
 184     s1=format("%d", 123456789,1)
 185     print s1, "is", atoi(s1)
 186     #standard formatting
 187     s1=str(3.14)
 188     print s1, "is", atof(s1)
 189
 190 ### Locale name aliasing engine
 191
 192 # Author: Marc-Andre Lemburg, mal@lemburg.com
 193 # Various tweaks by Fredrik Lundh <effbot@telia.com>
 194
 195 # store away the low-level version of setlocale (it's
 196 # overridden below)
 197 _setlocale = setlocale
 198
 199 def normalize(localename):
 200
 201     """ Returns a normalized locale code for the given locale
 202         name.
 203
 204         The returned locale code is formatted for use with
 205         setlocale().
 206
 207         If normalization fails, the original name is returned
 208         unchanged.
 209
 210         If the given encoding is not known, the function defaults to
 211         the default encoding for the locale code just like setlocale()
 212         does.
 213
 214     """
 215     # Normalize the locale name and extract the encoding
 216     fullname = localename.lower()
 217     if ':' in fullname:
 218         # ':' is sometimes used as encoding delimiter.
 219         fullname = fullname.replace(':', '.')
 220     if '.' in fullname:
 221         langname, encoding = fullname.split('.')[:2]
 222         fullname = langname + '.' + encoding
 223     else:
 224         langname = fullname
 225         encoding = ''
 226
 227     # First lookup: fullname (possibly with encoding)
 228     code = locale_alias.get(fullname, None)
 229     if code is not None:
 230         return code
 231
 232     # Second try: langname (without encoding)
 233     code = locale_alias.get(langname, None)
 234     if code is not None:
 235         if '.' in code:
 236             langname, defenc = code.split('.')
 237         else:
 238             langname = code
 239             defenc = ''
 240         if encoding:
 241             encoding = encoding_alias.get(encoding, encoding)
 242         else:
 243             encoding = defenc
 244         if encoding:
 245             return langname + '.' + encoding
 246         else:
 247             return langname
 248
 249     else:
 250         return localename
 251
 252 def _parse_localename(localename):
 253
 254     """ Parses the locale code for localename and returns the
 255         result as tuple (language code, encoding).
 256
 257         The localename is normalized and passed through the locale
 258         alias engine. A ValueError is raised in case the locale name
 259         cannot be parsed.
 260
 261         The language code corresponds to RFC 1766.  code and encoding
 262         can be None in case the values cannot be determined or are
 263         unknown to this implementation.
 264
 265     """
 266     code = normalize(localename)
 267     if '.' in code:
 268         return code.split('.')[:2]
 269     elif code == 'C':
 270         return None, None
 271     raise ValueError, 'unknown locale: %s' % localename
 272
 273 def _build_localename(localetuple):
 274
 275     """ Builds a locale code from the given tuple (language code,
 276         encoding).
 277
 278         No aliasing or normalizing takes place.
 279
 280     """
 281     language, encoding = localetuple
 282     if language is None:
 283         language = 'C'
 284     if encoding is None:
 285         return language
 286     else:
 287         return language + '.' + encoding
 288
 289 def getdefaultlocale(envvars=('LANGUAGE', 'LC_ALL', 'LC_CTYPE', 'LANG')):
 290
 291     """ Tries to determine the default locale settings and returns
 292         them as tuple (language code, encoding).
 293
 294         According to POSIX, a program which has not called
 295         setlocale(LC_ALL, "") runs using the portable 'C' locale.
 296         Calling setlocale(LC_ALL, "") lets it use the default locale as
 297         defined by the LANG variable. Since we don't want to interfere
 298         with the current locale setting we thus emulate the behavior
 299         in the way described above.
 300
 301         To maintain compatibility with other platforms, not only the
 302         LANG variable is tested, but a list of variables given as
 303         envvars parameter. The first found to be defined will be
 304         used. envvars defaults to the search path used in GNU gettext;
 305         it must always contain the variable name 'LANG'.
 306
 307         Except for the code 'C', the language code corresponds to RFC
 308         1766.  code and encoding can be None in case the values cannot
 309         be determined.
 310
 311     """
 312
 313     try:
 314         # check if it's supported by the _locale module
 315         import _locale
 316         code, encoding = _locale._getdefaultlocale()
 317     except (ImportError, AttributeError):
 318         pass
 319     else:
 320         # make sure the code/encoding values are valid
 321         if sys.platform == "win32" and code and code[:2] == "0x":
 322             # map windows language identifier to language name
 323             code = windows_locale.get(int(code, 0))
 324         # ...add other platform-specific processing here, if
 325         # necessary...
 326         return code, encoding
 327
 328     # fall back on POSIX behaviour
 329     import os
 330     lookup = os.environ.get
 331     for variable in envvars:
 332         localename = lookup(variable,None)
 333         if localename is not None:
 334             break
 335     else:
 336         localename = 'C'
 337     return _parse_localename(localename)
 338
 339
 340 def getlocale(category=LC_CTYPE):
 341
 342     """ Returns the current setting for the given locale category as
 343         tuple (language code, encoding).
 344
 345         category may be one of the LC_* value except LC_ALL. It
 346         defaults to LC_CTYPE.
 347
 348         Except for the code 'C', the language code corresponds to RFC
 349         1766.  code and encoding can be None in case the values cannot
 350         be determined.
 351
 352     """
 353     localename = _setlocale(category)
 354     if category == LC_ALL and ';' in localename:
 355         raise TypeError, 'category LC_ALL is not supported'
 356     return _parse_localename(localename)
 357
 358 def setlocale(category, locale=None):
 359
 360     """ Set the locale for the given category.  The locale can be
 361         a string, a locale tuple (language code, encoding), or None.
 362
 363         Locale tuples are converted to strings the locale aliasing
 364         engine.  Locale strings are passed directly to the C lib.
 365
 366         category may be given as one of the LC_* values.
 367
 368     """
 369     if locale and type(locale) is not type(""):
 370         # convert to string
 371         locale = normalize(_build_localename(locale))
 372     return _setlocale(category, locale)
 373
 374 def resetlocale(category=LC_ALL):
 375
 376     """ Sets the locale for category to the default setting.
 377
 378         The default setting is determined by calling
 379         getdefaultlocale(). category defaults to LC_ALL.
 380
 381     """
 382     _setlocale(category, _build_localename(getdefaultlocale()))
 383
 384 ### Database
 385 #
 386 # The following data was extracted from the locale.alias file which
 387 # comes with X11 and then hand edited removing the explicit encoding
 388 # definitions and adding some more aliases. The file is usually
 389 # available as /usr/lib/X11/locale/locale.alias.
 390 #
 391
 392 #
 393 # The encoding_alias table maps lowercase encoding alias names to C
 394 # locale encoding names (case-sensitive).
 395 #
 396 encoding_alias = {
 397         '437':                          'C',
 398         'c':                            'C',
 399         'iso8859':                      'ISO8859-1',
 400         '8859':                         'ISO8859-1',
 401         '88591':                        'ISO8859-1',
 402         'ascii':                        'ISO8859-1',
 403         'en':                           'ISO8859-1',
 404         'iso88591':                     'ISO8859-1',
 405         'iso_8859-1':                   'ISO8859-1',
 406         '885915':                       'ISO8859-15',
 407         'iso885915':                    'ISO8859-15',
 408         'iso_8859-15':                  'ISO8859-15',
 409         'iso8859-2':                    'ISO8859-2',
 410         'iso88592':                     'ISO8859-2',
 411         'iso_8859-2':                   'ISO8859-2',
 412         'iso88595':                     'ISO8859-5',
 413         'iso88596':                     'ISO8859-6',
 414         'iso88597':                     'ISO8859-7',
 415         'iso88598':                     'ISO8859-8',
 416         'iso88599':                     'ISO8859-9',
 417         'iso-2022-jp':                  'JIS7',
 418         'jis':                          'JIS7',
 419         'jis7':                         'JIS7',
 420         'sjis':                         'SJIS',
 421         'tis620':                       'TACTIS',
 422         'ajec':                         'eucJP',
 423         'eucjp':                        'eucJP',
 424         'ujis':                         'eucJP',
 425         'utf-8':                        'utf',
 426         'utf8':                         'utf',
 427         'utf8@ucs4':                    'utf',
 428 }
 429
 430 #
 431 # The locale_alias table maps lowercase alias names to C locale names
 432 # (case-sensitive). Encodings are always separated from the locale
 433 # name using a dot ('.'); they should only be given in case the
 434 # language name is needed to interpret the given encoding alias
 435 # correctly (CJK codes often have this need).
 436 #
 437 locale_alias = {
 438         'american':                      'en_US.ISO8859-1',
 439         'ar':                            'ar_AA.ISO8859-6',
 440         'ar_aa':                         'ar_AA.ISO8859-6',
 441         'ar_sa':                         'ar_SA.ISO8859-6',
 442         'arabic':                        'ar_AA.ISO8859-6',
 443         'bg':                            'bg_BG.ISO8859-5',
 444         'bg_bg':                         'bg_BG.ISO8859-5',
 445         'bulgarian':                     'bg_BG.ISO8859-5',
 446         'c-french':                      'fr_CA.ISO8859-1',
 447         'c':                             'C',
 448         'c_c':                           'C',
 449         'cextend':                       'en_US.ISO8859-1',
 450         'chinese-s':                     'zh_CN.eucCN',
 451         'chinese-t':                     'zh_TW.eucTW',
 452         'croatian':                      'hr_HR.ISO8859-2',
 453         'cs':                            'cs_CZ.ISO8859-2',
 454         'cs_cs':                         'cs_CZ.ISO8859-2',
 455         'cs_cz':                         'cs_CZ.ISO8859-2',
 456         'cz':                            'cz_CZ.ISO8859-2',
 457         'cz_cz':                         'cz_CZ.ISO8859-2',
 458         'czech':                         'cs_CS.ISO8859-2',
 459         'da':                            'da_DK.ISO8859-1',
 460         'da_dk':                         'da_DK.ISO8859-1',
 461         'danish':                        'da_DK.ISO8859-1',
 462         'de':                            'de_DE.ISO8859-1',
 463         'de_at':                         'de_AT.ISO8859-1',
 464         'de_ch':                         'de_CH.ISO8859-1',
 465         'de_de':                         'de_DE.ISO8859-1',
 466         'dutch':                         'nl_BE.ISO8859-1',
 467         'ee':                            'ee_EE.ISO8859-4',
 468         'el':                            'el_GR.ISO8859-7',
 469         'el_gr':                         'el_GR.ISO8859-7',
 470         'en':                            'en_US.ISO8859-1',
 471         'en_au':                         'en_AU.ISO8859-1',
 472         'en_ca':                         'en_CA.ISO8859-1',
 473         'en_gb':                         'en_GB.ISO8859-1',
 474         'en_ie':                         'en_IE.ISO8859-1',
 475         'en_nz':                         'en_NZ.ISO8859-1',
 476         'en_uk':                         'en_GB.ISO8859-1',
 477         'en_us':                         'en_US.ISO8859-1',
 478         'eng_gb':                        'en_GB.ISO8859-1',
 479         'english':                       'en_EN.ISO8859-1',
 480         'english_uk':                    'en_GB.ISO8859-1',
 481         'english_united-states':         'en_US.ISO8859-1',
 482         'english_us':                    'en_US.ISO8859-1',
 483         'es':                            'es_ES.ISO8859-1',
 484         'es_ar':                         'es_AR.ISO8859-1',
 485         'es_bo':                         'es_BO.ISO8859-1',
 486         'es_cl':                         'es_CL.ISO8859-1',
 487         'es_co':                         'es_CO.ISO8859-1',
 488         'es_cr':                         'es_CR.ISO8859-1',
 489         'es_ec':                         'es_EC.ISO8859-1',
 490         'es_es':                         'es_ES.ISO8859-1',
 491         'es_gt':                         'es_GT.ISO8859-1',
 492         'es_mx':                         'es_MX.ISO8859-1',
 493         'es_ni':                         'es_NI.ISO8859-1',
 494         'es_pa':                         'es_PA.ISO8859-1',
 495         'es_pe':                         'es_PE.ISO8859-1',
 496         'es_py':                         'es_PY.ISO8859-1',
 497         'es_sv':                         'es_SV.ISO8859-1',
 498         'es_uy':                         'es_UY.ISO8859-1',
 499         'es_ve':                         'es_VE.ISO8859-1',
 500         'et':                            'et_EE.ISO8859-4',
 501         'et_ee':                         'et_EE.ISO8859-4',
 502         'fi':                            'fi_FI.ISO8859-1',
 503         'fi_fi':                         'fi_FI.ISO8859-1',
 504         'finnish':                       'fi_FI.ISO8859-1',
 505         'fr':                            'fr_FR.ISO8859-1',
 506         'fr_be':                         'fr_BE.ISO8859-1',
 507         'fr_ca':                         'fr_CA.ISO8859-1',
 508         'fr_ch':                         'fr_CH.ISO8859-1',
 509         'fr_fr':                         'fr_FR.ISO8859-1',
 510         'fre_fr':                        'fr_FR.ISO8859-1',
 511         'french':                        'fr_FR.ISO8859-1',
 512         'french_france':                 'fr_FR.ISO8859-1',
 513         'ger_de':                        'de_DE.ISO8859-1',
 514         'german':                        'de_DE.ISO8859-1',
 515         'german_germany':                'de_DE.ISO8859-1',
 516         'greek':                         'el_GR.ISO8859-7',
 517         'hebrew':                        'iw_IL.ISO8859-8',
 518         'hr':                            'hr_HR.ISO8859-2',
 519         'hr_hr':                         'hr_HR.ISO8859-2',
 520         'hu':                            'hu_HU.ISO8859-2',
 521         'hu_hu':                         'hu_HU.ISO8859-2',
 522         'hungarian':                     'hu_HU.ISO8859-2',
 523         'icelandic':                     'is_IS.ISO8859-1',
 524         'id':                            'id_ID.ISO8859-1',
 525         'id_id':                         'id_ID.ISO8859-1',
 526         'is':                            'is_IS.ISO8859-1',
 527         'is_is':                         'is_IS.ISO8859-1',
 528         'iso-8859-1':                    'en_US.ISO8859-1',
 529         'iso-8859-15':                   'en_US.ISO8859-15',
 530         'iso8859-1':                     'en_US.ISO8859-1',
 531         'iso8859-15':                    'en_US.ISO8859-15',
 532         'iso_8859_1':                    'en_US.ISO8859-1',
 533         'iso_8859_15':                   'en_US.ISO8859-15',
 534         'it':                            'it_IT.ISO8859-1',
 535         'it_ch':                         'it_CH.ISO8859-1',
 536         'it_it':                         'it_IT.ISO8859-1',
 537         'italian':                       'it_IT.ISO8859-1',
 538         'iw':                            'iw_IL.ISO8859-8',
 539         'iw_il':                         'iw_IL.ISO8859-8',
 540         'ja':                            'ja_JP.eucJP',
 541         'ja.jis':                        'ja_JP.JIS7',
 542         'ja.sjis':                       'ja_JP.SJIS',
 543         'ja_jp':                         'ja_JP.eucJP',
 544         'ja_jp.ajec':                    'ja_JP.eucJP',
 545         'ja_jp.euc':                     'ja_JP.eucJP',
 546         'ja_jp.eucjp':                   'ja_JP.eucJP',
 547         'ja_jp.iso-2022-jp':             'ja_JP.JIS7',
 548         'ja_jp.jis':                     'ja_JP.JIS7',
 549         'ja_jp.jis7':                    'ja_JP.JIS7',
 550         'ja_jp.mscode':                  'ja_JP.SJIS',
 551         'ja_jp.sjis':                    'ja_JP.SJIS',
 552         'ja_jp.ujis':                    'ja_JP.eucJP',
 553         'japan':                         'ja_JP.eucJP',
 554         'japanese':                      'ja_JP.SJIS',
 555         'japanese-euc':                  'ja_JP.eucJP',
 556         'japanese.euc':                  'ja_JP.eucJP',
 557         'jp_jp':                         'ja_JP.eucJP',
 558         'ko':                            'ko_KR.eucKR',
 559         'ko_kr':                         'ko_KR.eucKR',
 560         'ko_kr.euc':                     'ko_KR.eucKR',
 561         'korean':                        'ko_KR.eucKR',
 562         'lt':                            'lt_LT.ISO8859-4',
 563         'lv':                            'lv_LV.ISO8859-4',
 564         'mk':                            'mk_MK.ISO8859-5',
 565         'mk_mk':                         'mk_MK.ISO8859-5',
 566         'nl':                            'nl_NL.ISO8859-1',
 567         'nl_be':                         'nl_BE.ISO8859-1',
 568         'nl_nl':                         'nl_NL.ISO8859-1',
 569         'no':                            'no_NO.ISO8859-1',
 570         'no_no':                         'no_NO.ISO8859-1',
 571         'norwegian':                     'no_NO.ISO8859-1',
 572         'pl':                            'pl_PL.ISO8859-2',
 573         'pl_pl':                         'pl_PL.ISO8859-2',
 574         'polish':                        'pl_PL.ISO8859-2',
 575         'portuguese':                    'pt_PT.ISO8859-1',
 576         'portuguese_brazil':             'pt_BR.ISO8859-1',
 577         'posix':                         'C',
 578         'posix-utf2':                    'C',
 579         'pt':                            'pt_PT.ISO8859-1',
 580         'pt_br':                         'pt_BR.ISO8859-1',
 581         'pt_pt':                         'pt_PT.ISO8859-1',
 582         'ro':                            'ro_RO.ISO8859-2',
 583         'ro_ro':                         'ro_RO.ISO8859-2',
 584         'ru':                            'ru_RU.ISO8859-5',
 585         'ru_ru':                         'ru_RU.ISO8859-5',
 586         'rumanian':                      'ro_RO.ISO8859-2',
 587         'russian':                       'ru_RU.ISO8859-5',
 588         'serbocroatian':                 'sh_YU.ISO8859-2',
 589         'sh':                            'sh_YU.ISO8859-2',
 590         'sh_hr':                         'sh_HR.ISO8859-2',
 591         'sh_sp':                         'sh_YU.ISO8859-2',
 592         'sh_yu':                         'sh_YU.ISO8859-2',
 593         'sk':                            'sk_SK.ISO8859-2',
 594         'sk_sk':                         'sk_SK.ISO8859-2',
 595         'sl':                            'sl_CS.ISO8859-2',
 596         'sl_cs':                         'sl_CS.ISO8859-2',
 597         'sl_si':                         'sl_SI.ISO8859-2',
 598         'slovak':                        'sk_SK.ISO8859-2',
 599         'slovene':                       'sl_CS.ISO8859-2',
 600         'sp':                            'sp_YU.ISO8859-5',
 601         'sp_yu':                         'sp_YU.ISO8859-5',
 602         'spanish':                       'es_ES.ISO8859-1',
 603         'spanish_spain':                 'es_ES.ISO8859-1',
 604         'sr_sp':                         'sr_SP.ISO8859-2',
 605         'sv':                            'sv_SE.ISO8859-1',
 606         'sv_se':                         'sv_SE.ISO8859-1',
 607         'swedish':                       'sv_SE.ISO8859-1',
 608         'th_th':                         'th_TH.TACTIS',
 609         'tr':                            'tr_TR.ISO8859-9',
 610         'tr_tr':                         'tr_TR.ISO8859-9',
 611         'turkish':                       'tr_TR.ISO8859-9',
 612         'univ':                          'en_US.utf',
 613         'universal':                     'en_US.utf',
 614         'zh':                            'zh_CN.eucCN',
 615         'zh_cn':                         'zh_CN.eucCN',
 616         'zh_cn.big5':                    'zh_TW.eucTW',
 617         'zh_cn.euc':                     'zh_CN.eucCN',
 618         'zh_tw':                         'zh_TW.eucTW',
 619         'zh_tw.euc':                     'zh_TW.eucTW',
 620 }
 621
 622 #
 623 # this maps windows language identifiers (as used on Windows 95 and
 624 # earlier) to locale strings.
 625 #
 626 # NOTE: this mapping is incomplete.  If your language is missing, send
 627 # a note with the missing language identifier and the suggested locale
 628 # code to Fredrik Lundh <effbot@telia.com>.  Thanks /F
 629
 630 windows_locale = {
 631     0x0404: "zh_TW", # Chinese (Taiwan)
 632     0x0804: "zh_CN", # Chinese (PRC)
 633     0x0406: "da_DK", # Danish
 634     0x0413: "nl_NL", # Dutch (Netherlands)
 635     0x0409: "en_US", # English (United States)
 636     0x0809: "en_UK", # English (United Kingdom)
 637     0x0c09: "en_AU", # English (Australian)
 638     0x1009: "en_CA", # English (Canadian)
 639     0x1409: "en_NZ", # English (New Zealand)
 640     0x1809: "en_IE", # English (Ireland)
 641     0x1c09: "en_ZA", # English (South Africa)
 642     0x040b: "fi_FI", # Finnish
 643     0x040c: "fr_FR", # French (Standard)
 644     0x080c: "fr_BE", # French (Belgian)
 645     0x0c0c: "fr_CA", # French (Canadian)
 646     0x100c: "fr_CH", # French (Switzerland)
 647     0x0407: "de_DE", # German (Standard)
 648     0x0408: "el_GR", # Greek
 649     0x040d: "iw_IL", # Hebrew
 650     0x040f: "is_IS", # Icelandic
 651     0x0410: "it_IT", # Italian (Standard)
 652     0x0411: "ja_JA", # Japanese
 653     0x0414: "no_NO", # Norwegian (Bokmal)
 654     0x0816: "pt_PT", # Portuguese (Standard)
 655     0x0c0a: "es_ES", # Spanish (Modern Sort)
 656     0x0441: "sw_KE", # Swahili (Kenya)
 657     0x041d: "sv_SE", # Swedish
 658     0x081d: "sv_FI", # Swedish (Finland)
 659     0x041f: "tr_TR", # Turkish
 660 }
 661
 662 def _print_locale():
 663
 664     """ Test function.
 665     """
 666     categories = {}
 667     def _init_categories(categories=categories):
 668         for k,v in globals().items():
 669             if k[:3] == 'LC_':
 670                 categories[k] = v
 671     _init_categories()
 672     del categories['LC_ALL']
 673
 674     print 'Locale defaults as determined by getdefaultlocale():'
 675     print '-'*72
 676     lang, enc = getdefaultlocale()
 677     print 'Language: ', lang or '(undefined)'
 678     print 'Encoding: ', enc or '(undefined)'
 679     print
 680
 681     print 'Locale settings on startup:'
 682     print '-'*72
 683     for name,category in categories.items():
 684         print name, '...'
 685         lang, enc = getlocale(category)
 686         print '   Language: ', lang or '(undefined)'
 687         print '   Encoding: ', enc or '(undefined)'
 688         print
 689
 690     print
 691     print 'Locale settings after calling resetlocale():'
 692     print '-'*72
 693     resetlocale()
 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     try:
 702         setlocale(LC_ALL, "")
 703     except:
 704         print 'NOTE:'
 705         print 'setlocale(LC_ALL, "") does not support the default locale'
 706         print 'given in the OS environment variables.'
 707     else:
 708         print
 709         print 'Locale settings after calling setlocale(LC_ALL, ""):'
 710         print '-'*72
 711         for name,category in categories.items():
 712             print name, '...'
 713             lang, enc = getlocale(category)
 714             print '   Language: ', lang or '(undefined)'
 715             print '   Encoding: ', enc or '(undefined)'
 716             print
 717
 718 ###
 719
 720 try:
 721     LC_MESSAGES
 722 except:
 723     pass
 724 else:
 725     __all__.append("LC_MESSAGES")
 726
 727 if __name__=='__main__':
 728     print 'Locale aliasing:'
 729     print
 730     _print_locale()
 731     print
 732     print 'Number formatting:'
 733     print
 734     _test()