1/* 2 ********************************************************************** 3 * Copyright (C) 1996-2010, International Business Machines 4 * Corporation and others. All Rights Reserved. 5 ********************************************************************** 6 * 7 * Provides functionality for mapping between 8 * LCID and Posix IDs or ICU locale to codepage 9 * 10 * Note: All classes and code in this file are 11 * intended for internal use only. 12 * 13 * Methods of interest: 14 * unsigned long convertToLCID(const char*); 15 * const char* convertToPosix(unsigned long); 16 * 17 * Kathleen Wilson, 4/30/96 18 * 19 * Date Name Description 20 * 3/11/97 aliu Fixed off-by-one bug in assignment operator. Added 21 * setId() method and safety check against 22 * MAX_ID_LENGTH. 23 * 04/23/99 stephen Added C wrapper for convertToPosix. 24 * 09/18/00 george Removed the memory leaks. 25 * 08/23/01 george Convert to C 26 */ 27 28#include "locmap.h" 29#include "unicode/uloc.h" 30#include "cstring.h" 31#include "cmemory.h" 32 33#if defined(U_WINDOWS) && defined(_MSC_VER) && (_MSC_VER >= 1500) 34#define USE_WINDOWS_LOCALE_API 35#endif 36 37#ifdef USE_WINDOWS_LOCALE_API 38#include <windows.h> 39#include <winnls.h> 40#endif 41 42/* 43 * Note: 44 * The mapping from Win32 locale ID numbers to POSIX locale strings should 45 * be the faster one. 46 * 47 * Many LCID values come from winnt.h 48 * Some also come from http://www.microsoft.com/globaldev/reference/lcid-all.mspx 49 */ 50 51/* 52//////////////////////////////////////////////// 53// 54// Internal Classes for LCID <--> POSIX Mapping 55// 56///////////////////////////////////////////////// 57*/ 58 59typedef struct ILcidPosixElement 60{ 61 const uint32_t hostID; 62 const char * const posixID; 63} ILcidPosixElement; 64 65typedef struct ILcidPosixMap 66{ 67 const uint32_t numRegions; 68 const struct ILcidPosixElement* const regionMaps; 69} ILcidPosixMap; 70 71 72/* 73///////////////////////////////////////////////// 74// 75// Easy macros to make the LCID <--> POSIX Mapping 76// 77///////////////////////////////////////////////// 78*/ 79 80/** 81 * The standard one language/one country mapping for LCID. 82 * The first element must be the language, and the following 83 * elements are the language with the country. 84 * @param hostID LCID in host format such as 0x044d 85 * @param languageID posix ID of just the language such as 'de' 86 * @param posixID posix ID of the language_TERRITORY such as 'de_CH' 87 */ 88#define ILCID_POSIX_ELEMENT_ARRAY(hostID, languageID, posixID) \ 89static const ILcidPosixElement locmap_ ## languageID [] = { \ 90 {LANGUAGE_LCID(hostID), #languageID}, /* parent locale */ \ 91 {hostID, #posixID}, \ 92}; 93 94/** 95 * Define a subtable by ID 96 * @param id the POSIX ID, either a language or language_TERRITORY 97 */ 98#define ILCID_POSIX_SUBTABLE(id) \ 99static const ILcidPosixElement locmap_ ## id [] = 100 101 102/** 103 * Create the map for the posixID. This macro supposes that the language string 104 * name is the same as the global variable name, and that the first element 105 * in the ILcidPosixElement is just the language. 106 * @param _posixID the full POSIX ID for this entry. 107 */ 108#define ILCID_POSIX_MAP(_posixID) \ 109 {sizeof(locmap_ ## _posixID)/sizeof(ILcidPosixElement), locmap_ ## _posixID} 110 111/* 112//////////////////////////////////////////// 113// 114// Create the table of LCID to POSIX Mapping 115// None of it should be dynamically created. 116// 117// Keep static locale variables inside the function so that 118// it can be created properly during static init. 119// 120// Note: This table should be updated periodically. Check the National Lanaguage Support API Reference Website. 121// Microsoft is moving away from LCID in favor of locale name as of Vista. This table needs to be 122// maintained for support of older Windows version. 123// Update: Windows 7 (091130) 124//////////////////////////////////////////// 125*/ 126 127ILCID_POSIX_ELEMENT_ARRAY(0x0436, af, af_ZA) 128 129ILCID_POSIX_SUBTABLE(ar) { 130 {0x01, "ar"}, 131 {0x3801, "ar_AE"}, 132 {0x3c01, "ar_BH"}, 133 {0x1401, "ar_DZ"}, 134 {0x0c01, "ar_EG"}, 135 {0x0801, "ar_IQ"}, 136 {0x2c01, "ar_JO"}, 137 {0x3401, "ar_KW"}, 138 {0x3001, "ar_LB"}, 139 {0x1001, "ar_LY"}, 140 {0x1801, "ar_MA"}, 141 {0x2001, "ar_OM"}, 142 {0x4001, "ar_QA"}, 143 {0x0401, "ar_SA"}, 144 {0x2801, "ar_SY"}, 145 {0x1c01, "ar_TN"}, 146 {0x2401, "ar_YE"} 147}; 148 149ILCID_POSIX_ELEMENT_ARRAY(0x044d, as, as_IN) 150ILCID_POSIX_ELEMENT_ARRAY(0x045e, am, am_ET) 151ILCID_POSIX_ELEMENT_ARRAY(0x047a, arn,arn_CL) 152 153ILCID_POSIX_SUBTABLE(az) { 154 {0x2c, "az"}, 155 {0x082c, "az_Cyrl_AZ"}, /* Cyrillic based */ 156 {0x742c, "az_Cyrl"}, /* Cyrillic based */ 157 {0x042c, "az_Latn_AZ"}, /* Latin based */ 158 {0x782c, "az_Latn"}, /* Latin based */ 159 {0x042c, "az_AZ"} /* Latin based */ 160}; 161 162ILCID_POSIX_ELEMENT_ARRAY(0x046d, ba, ba_RU) 163ILCID_POSIX_ELEMENT_ARRAY(0x0423, be, be_BY) 164 165ILCID_POSIX_SUBTABLE(ber) { 166 {0x5f, "ber"}, 167 {0x045f, "ber_Arab_DZ"}, 168 {0x045f, "ber_Arab"}, 169 {0x085f, "ber_Latn_DZ"}, 170 {0x085f, "ber_Latn"} 171}; 172 173ILCID_POSIX_ELEMENT_ARRAY(0x0402, bg, bg_BG) 174 175ILCID_POSIX_SUBTABLE(bn) { 176 {0x45, "bn"}, 177 {0x0845, "bn_BD"}, 178 {0x0445, "bn_IN"} 179}; 180 181ILCID_POSIX_SUBTABLE(bo) { 182 {0x51, "bo"}, 183 {0x0851, "bo_BT"}, 184 {0x0451, "bo_CN"} 185}; 186 187ILCID_POSIX_ELEMENT_ARRAY(0x047e, br, br_FR) 188ILCID_POSIX_ELEMENT_ARRAY(0x0403, ca, ca_ES) 189ILCID_POSIX_ELEMENT_ARRAY(0x0483, co, co_FR) 190ILCID_POSIX_ELEMENT_ARRAY(0x045c, chr,chr_US) 191 192/* Declared as cs_CZ to get around compiler errors on z/OS, which defines cs as a function */ 193ILCID_POSIX_ELEMENT_ARRAY(0x0405, cs, cs_CZ) 194 195ILCID_POSIX_ELEMENT_ARRAY(0x0452, cy, cy_GB) 196ILCID_POSIX_ELEMENT_ARRAY(0x0406, da, da_DK) 197 198ILCID_POSIX_SUBTABLE(de) { 199 {0x07, "de"}, 200 {0x0c07, "de_AT"}, 201 {0x0807, "de_CH"}, 202 {0x0407, "de_DE"}, 203 {0x1407, "de_LI"}, 204 {0x1007, "de_LU"}, 205 {0x10407,"de_DE@collation=phonebook"}, /*This is really de_DE_PHONEBOOK on Windows*/ 206 {0x10407,"de@collation=phonebook"} /*This is really de_DE_PHONEBOOK on Windows*/ 207}; 208 209ILCID_POSIX_ELEMENT_ARRAY(0x0465, dv, dv_MV) 210ILCID_POSIX_ELEMENT_ARRAY(0x0408, el, el_GR) 211 212ILCID_POSIX_SUBTABLE(en) { 213 {0x09, "en"}, 214 {0x0c09, "en_AU"}, 215 {0x2809, "en_BZ"}, 216 {0x1009, "en_CA"}, 217 {0x0809, "en_GB"}, 218 {0x1809, "en_IE"}, 219 {0x4009, "en_IN"}, 220 {0x2009, "en_JM"}, 221 {0x4409, "en_MY"}, 222 {0x1409, "en_NZ"}, 223 {0x3409, "en_PH"}, 224 {0x4809, "en_SG"}, 225 {0x2C09, "en_TT"}, 226 {0x0409, "en_US"}, 227 {0x007f, "en_US_POSIX"}, /* duplicate for roundtripping */ 228 {0x2409, "en_VI"}, /* Virgin Islands AKA Caribbean Islands (en_CB). */ 229 {0x1c09, "en_ZA"}, 230 {0x3009, "en_ZW"}, 231 {0x2409, "en_029"}, 232 {0x0409, "en_AS"}, /* Alias for en_US. Leave last. */ 233 {0x0409, "en_GU"}, /* Alias for en_US. Leave last. */ 234 {0x0409, "en_MH"}, /* Alias for en_US. Leave last. */ 235 {0x0409, "en_MP"}, /* Alias for en_US. Leave last. */ 236 {0x0409, "en_UM"} /* Alias for en_US. Leave last. */ 237}; 238 239ILCID_POSIX_SUBTABLE(en_US_POSIX) { 240 {0x007f, "en_US_POSIX"} /* duplicate for roundtripping */ 241}; 242 243ILCID_POSIX_SUBTABLE(es) { 244 {0x0a, "es"}, 245 {0x2c0a, "es_AR"}, 246 {0x400a, "es_BO"}, 247 {0x340a, "es_CL"}, 248 {0x240a, "es_CO"}, 249 {0x140a, "es_CR"}, 250 {0x1c0a, "es_DO"}, 251 {0x300a, "es_EC"}, 252 {0x0c0a, "es_ES"}, /*Modern sort.*/ 253 {0x100a, "es_GT"}, 254 {0x480a, "es_HN"}, 255 {0x080a, "es_MX"}, 256 {0x4c0a, "es_NI"}, 257 {0x180a, "es_PA"}, 258 {0x280a, "es_PE"}, 259 {0x500a, "es_PR"}, 260 {0x3c0a, "es_PY"}, 261 {0x440a, "es_SV"}, 262 {0x540a, "es_US"}, 263 {0x380a, "es_UY"}, 264 {0x200a, "es_VE"}, 265 {0x040a, "es_ES@collation=traditional"}, 266 {0x040a, "es@collation=traditional"} 267}; 268 269ILCID_POSIX_ELEMENT_ARRAY(0x0425, et, et_EE) 270ILCID_POSIX_ELEMENT_ARRAY(0x042d, eu, eu_ES) 271 272/* ISO-639 doesn't distinguish between Persian and Dari.*/ 273ILCID_POSIX_SUBTABLE(fa) { 274 {0x29, "fa"}, 275 {0x0429, "fa_IR"}, /* Persian/Farsi (Iran) */ 276 {0x048c, "fa_AF"} /* Persian/Dari (Afghanistan) */ 277}; 278 279/* duplicate for roundtripping */ 280ILCID_POSIX_SUBTABLE(fa_AF) { 281 {0x8c, "fa_AF"}, /* Persian/Dari (Afghanistan) */ 282 {0x048c, "fa_AF"} /* Persian/Dari (Afghanistan) */ 283}; 284 285ILCID_POSIX_ELEMENT_ARRAY(0x040b, fi, fi_FI) 286ILCID_POSIX_ELEMENT_ARRAY(0x0464, fil,fil_PH) 287ILCID_POSIX_ELEMENT_ARRAY(0x0438, fo, fo_FO) 288 289ILCID_POSIX_SUBTABLE(fr) { 290 {0x0c, "fr"}, 291 {0x080c, "fr_BE"}, 292 {0x0c0c, "fr_CA"}, 293 {0x240c, "fr_CD"}, 294 {0x100c, "fr_CH"}, 295 {0x300c, "fr_CI"}, 296 {0x2c0c, "fr_CM"}, 297 {0x040c, "fr_FR"}, 298 {0x3c0c, "fr_HT"}, 299 {0x140c, "fr_LU"}, 300 {0x380c, "fr_MA"}, 301 {0x180c, "fr_MC"}, 302 {0x340c, "fr_ML"}, 303 {0x200c, "fr_RE"}, 304 {0x280c, "fr_SN"} 305}; 306 307ILCID_POSIX_ELEMENT_ARRAY(0x0462, fy, fy_NL) 308 309/* This LCID is really two different locales.*/ 310ILCID_POSIX_ELEMENT_ARRAY(0x083c, ga, ga_IE) /* Gaelic (Ireland) */ 311ILCID_POSIX_ELEMENT_ARRAY(0x0491, gd, gd_GB) /* Gaelic (Scotland) */ 312 313ILCID_POSIX_ELEMENT_ARRAY(0x0456, gl, gl_ES) 314ILCID_POSIX_ELEMENT_ARRAY(0x0447, gu, gu_IN) 315ILCID_POSIX_ELEMENT_ARRAY(0x0474, gn, gn_PY) 316ILCID_POSIX_ELEMENT_ARRAY(0x0484, gsw,gsw_FR) 317 318ILCID_POSIX_SUBTABLE(ha) { 319 {0x68, "ha"}, 320 {0x7c68, "ha_Latn"}, 321 {0x0468, "ha_Latn_NG"}, 322}; 323 324ILCID_POSIX_ELEMENT_ARRAY(0x0475, haw,haw_US) 325ILCID_POSIX_ELEMENT_ARRAY(0x040d, he, he_IL) 326ILCID_POSIX_ELEMENT_ARRAY(0x0439, hi, hi_IN) 327 328/* This LCID is really four different locales.*/ 329ILCID_POSIX_SUBTABLE(hr) { 330 {0x1a, "hr"}, 331 {0x141a, "bs_Latn_BA"}, /* Bosnian, Bosnia and Herzegovina */ 332 {0x681a, "bs_Latn"}, /* Bosnian, Bosnia and Herzegovina */ 333 {0x141a, "bs_BA"}, /* Bosnian, Bosnia and Herzegovina */ 334 {0x781a, "bs"}, /* Bosnian */ 335 {0x201a, "bs_Cyrl_BA"}, /* Bosnian, Bosnia and Herzegovina */ 336 {0x641a, "bs_Cyrl"}, /* Bosnian, Bosnia and Herzegovina */ 337 {0x101a, "hr_BA"}, /* Croatian in Bosnia */ 338 {0x041a, "hr_HR"}, /* Croatian*/ 339 {0x2c1a, "sr_Latn_ME"}, 340 {0x241a, "sr_Latn_RS"}, 341 {0x181a, "sr_Latn_BA"}, /* Serbo-Croatian in Bosnia */ 342 {0x081a, "sr_Latn_CS"}, /* Serbo-Croatian*/ 343 {0x701a, "sr_Latn"}, /* It's 0x1a or 0x081a, pick one to make the test program happy. */ 344 {0x1c1a, "sr_Cyrl_BA"}, /* Serbo-Croatian in Bosnia */ 345 {0x0c1a, "sr_Cyrl_CS"}, /* Serbian*/ 346 {0x301a, "sr_Cyrl_ME"}, 347 {0x281a, "sr_Cyrl_RS"}, 348 {0x6c1a, "sr_Cyrl"}, /* It's 0x1a or 0x0c1a, pick one to make the test program happy. */ 349 {0x7c1a, "sr"} /* In CLDR sr is sr_Cyrl. */ 350}; 351 352ILCID_POSIX_ELEMENT_ARRAY(0x040e, hu, hu_HU) 353ILCID_POSIX_ELEMENT_ARRAY(0x042b, hy, hy_AM) 354ILCID_POSIX_ELEMENT_ARRAY(0x0421, id, id_ID) 355ILCID_POSIX_ELEMENT_ARRAY(0x0470, ig, ig_NG) 356ILCID_POSIX_ELEMENT_ARRAY(0x0478, ii, ii_CN) 357ILCID_POSIX_ELEMENT_ARRAY(0x040f, is, is_IS) 358 359ILCID_POSIX_SUBTABLE(it) { 360 {0x10, "it"}, 361 {0x0810, "it_CH"}, 362 {0x0410, "it_IT"} 363}; 364 365ILCID_POSIX_SUBTABLE(iu) { 366 {0x5d, "iu"}, 367 {0x045d, "iu_Cans_CA"}, 368 {0x785d, "iu_Cans"}, 369 {0x085d, "iu_Latn_CA"}, 370 {0x7c5d, "iu_Latn"} 371}; 372 373ILCID_POSIX_ELEMENT_ARRAY(0x040d, iw, iw_IL) /*Left in for compatibility*/ 374ILCID_POSIX_ELEMENT_ARRAY(0x0411, ja, ja_JP) 375ILCID_POSIX_ELEMENT_ARRAY(0x0437, ka, ka_GE) 376ILCID_POSIX_ELEMENT_ARRAY(0x043f, kk, kk_KZ) 377ILCID_POSIX_ELEMENT_ARRAY(0x046f, kl, kl_GL) 378ILCID_POSIX_ELEMENT_ARRAY(0x0453, km, km_KH) 379ILCID_POSIX_ELEMENT_ARRAY(0x044b, kn, kn_IN) 380 381ILCID_POSIX_SUBTABLE(ko) { 382 {0x12, "ko"}, 383 {0x0812, "ko_KP"}, 384 {0x0412, "ko_KR"} 385}; 386 387ILCID_POSIX_ELEMENT_ARRAY(0x0457, kok, kok_IN) 388ILCID_POSIX_ELEMENT_ARRAY(0x0471, kr, kr_NG) 389 390ILCID_POSIX_SUBTABLE(ks) { /* We could add PK and CN too */ 391 {0x60, "ks"}, 392 {0x0860, "ks_IN"}, /* Documentation doesn't mention script */ 393 {0x0460, "ks_Arab_IN"} 394}; 395 396ILCID_POSIX_ELEMENT_ARRAY(0x0440, ky, ky_KG) /* Kyrgyz is spoken in Kyrgyzstan */ 397ILCID_POSIX_ELEMENT_ARRAY(0x0476, la, la_IT) /* TODO: Verify the country */ 398ILCID_POSIX_ELEMENT_ARRAY(0x046e, lb, lb_LU) 399ILCID_POSIX_ELEMENT_ARRAY(0x0454, lo, lo_LA) 400ILCID_POSIX_ELEMENT_ARRAY(0x0427, lt, lt_LT) 401ILCID_POSIX_ELEMENT_ARRAY(0x0426, lv, lv_LV) 402ILCID_POSIX_ELEMENT_ARRAY(0x0481, mi, mi_NZ) 403ILCID_POSIX_ELEMENT_ARRAY(0x042f, mk, mk_MK) 404ILCID_POSIX_ELEMENT_ARRAY(0x044c, ml, ml_IN) 405 406ILCID_POSIX_SUBTABLE(mn) { 407 {0x50, "mn"}, 408 {0x0450, "mn_MN"}, 409 {0x7c50, "mn_Mong"}, 410 {0x0850, "mn_Mong_CN"}, 411 {0x0850, "mn_CN"}, 412 {0x7850, "mn_Cyrl"} 413}; 414 415ILCID_POSIX_ELEMENT_ARRAY(0x0458, mni,mni_IN) 416ILCID_POSIX_ELEMENT_ARRAY(0x047c, moh,moh_CA) 417ILCID_POSIX_ELEMENT_ARRAY(0x044e, mr, mr_IN) 418 419ILCID_POSIX_SUBTABLE(ms) { 420 {0x3e, "ms"}, 421 {0x083e, "ms_BN"}, /* Brunei Darussalam*/ 422 {0x043e, "ms_MY"} /* Malaysia*/ 423}; 424 425ILCID_POSIX_ELEMENT_ARRAY(0x043a, mt, mt_MT) 426ILCID_POSIX_ELEMENT_ARRAY(0x0455, my, my_MM) 427 428ILCID_POSIX_SUBTABLE(ne) { 429 {0x61, "ne"}, 430 {0x0861, "ne_IN"}, /* India*/ 431 {0x0461, "ne_NP"} /* Nepal*/ 432}; 433 434ILCID_POSIX_SUBTABLE(nl) { 435 {0x13, "nl"}, 436 {0x0813, "nl_BE"}, 437 {0x0413, "nl_NL"} 438}; 439 440/* The "no" locale split into nb and nn. By default in ICU, "no" is nb.*/ 441ILCID_POSIX_SUBTABLE(no) { 442 {0x14, "no"}, /* really nb_NO */ 443 {0x7c14, "nb"}, /* really nb */ 444 {0x0414, "nb_NO"}, /* really nb_NO. Keep first in the 414 list. */ 445 {0x0414, "no_NO"}, /* really nb_NO */ 446 {0x0814, "nn_NO"}, /* really nn_NO. Keep first in the 814 list. */ 447 {0x7814, "nn"}, /* It's 0x14 or 0x814, pick one to make the test program happy. */ 448 {0x0814, "no_NO_NY"}/* really nn_NO */ 449}; 450 451ILCID_POSIX_ELEMENT_ARRAY(0x046c, nso,nso_ZA) /* TODO: Verify the ISO-639 code */ 452ILCID_POSIX_ELEMENT_ARRAY(0x0482, oc, oc_FR) 453ILCID_POSIX_ELEMENT_ARRAY(0x0472, om, om_ET) /* TODO: Verify the country */ 454 455/* Declared as or_IN to get around compiler errors*/ 456ILCID_POSIX_SUBTABLE(or_IN) { 457 {0x48, "or"}, 458 {0x0448, "or_IN"}, 459}; 460 461 462ILCID_POSIX_SUBTABLE(pa) { 463 {0x46, "pa"}, 464 {0x0446, "pa_IN"}, 465 {0x0846, "pa_PK"} 466}; 467 468ILCID_POSIX_ELEMENT_ARRAY(0x0415, pl, pl_PL) 469ILCID_POSIX_ELEMENT_ARRAY(0x0463, ps, ps_AF) 470 471ILCID_POSIX_SUBTABLE(pt) { 472 {0x16, "pt"}, 473 {0x0416, "pt_BR"}, 474 {0x0816, "pt_PT"} 475}; 476 477ILCID_POSIX_SUBTABLE(qu) { 478 {0x6b, "qu"}, 479 {0x046b, "qu_BO"}, 480 {0x086b, "qu_EC"}, 481 {0x0C6b, "qu_PE"} 482}; 483 484ILCID_POSIX_ELEMENT_ARRAY(0x0486, qut, qut_GT) /* qut is an ISO-639-3 code */ 485ILCID_POSIX_ELEMENT_ARRAY(0x0417, rm, rm_CH) 486ILCID_POSIX_ELEMENT_ARRAY(0x0418, ro, ro_RO) 487 488ILCID_POSIX_SUBTABLE(root) { 489 {0x00, "root"} 490}; 491 492ILCID_POSIX_ELEMENT_ARRAY(0x0419, ru, ru_RU) 493ILCID_POSIX_ELEMENT_ARRAY(0x0487, rw, rw_RW) 494ILCID_POSIX_ELEMENT_ARRAY(0x044f, sa, sa_IN) 495ILCID_POSIX_ELEMENT_ARRAY(0x0485, sah,sah_RU) 496 497ILCID_POSIX_SUBTABLE(sd) { 498 {0x59, "sd"}, 499 {0x0459, "sd_IN"}, 500 {0x0859, "sd_PK"} 501}; 502 503ILCID_POSIX_SUBTABLE(se) { 504 {0x3b, "se"}, 505 {0x0c3b, "se_FI"}, 506 {0x043b, "se_NO"}, 507 {0x083b, "se_SE"}, 508 {0x783b, "sma"}, 509 {0x183b, "sma_NO"}, 510 {0x1c3b, "sma_SE"}, 511 {0x7c3b, "smj"}, 512 {0x703b, "smn"}, 513 {0x743b, "sms"}, 514 {0x103b, "smj_NO"}, 515 {0x143b, "smj_SE"}, 516 {0x243b, "smn_FI"}, 517 {0x203b, "sms_FI"}, 518}; 519 520ILCID_POSIX_ELEMENT_ARRAY(0x045b, si, si_LK) 521ILCID_POSIX_ELEMENT_ARRAY(0x041b, sk, sk_SK) 522ILCID_POSIX_ELEMENT_ARRAY(0x0424, sl, sl_SI) 523ILCID_POSIX_ELEMENT_ARRAY(0x0477, so, so_ET) /* TODO: Verify the country */ 524ILCID_POSIX_ELEMENT_ARRAY(0x041c, sq, sq_AL) 525 526ILCID_POSIX_SUBTABLE(sv) { 527 {0x1d, "sv"}, 528 {0x081d, "sv_FI"}, 529 {0x041d, "sv_SE"} 530}; 531 532ILCID_POSIX_ELEMENT_ARRAY(0x0441, sw, sw_KE) 533ILCID_POSIX_ELEMENT_ARRAY(0x045A, syr, syr_SY) 534ILCID_POSIX_ELEMENT_ARRAY(0x0449, ta, ta_IN) 535ILCID_POSIX_ELEMENT_ARRAY(0x044a, te, te_IN) 536 537/* Cyrillic based by default */ 538ILCID_POSIX_SUBTABLE(tg) { 539 {0x28, "tg"}, 540 {0x7c28, "tg_Cyrl"}, 541 {0x0428, "tg_Cyrl_TJ"} 542}; 543 544ILCID_POSIX_ELEMENT_ARRAY(0x041e, th, th_TH) 545 546ILCID_POSIX_SUBTABLE(ti) { 547 {0x73, "ti"}, 548 {0x0873, "ti_ER"}, 549 {0x0473, "ti_ET"} 550}; 551 552ILCID_POSIX_ELEMENT_ARRAY(0x0442, tk, tk_TM) 553ILCID_POSIX_ELEMENT_ARRAY(0x0432, tn, tn_BW) 554ILCID_POSIX_ELEMENT_ARRAY(0x041f, tr, tr_TR) 555ILCID_POSIX_ELEMENT_ARRAY(0x0444, tt, tt_RU) 556 557ILCID_POSIX_SUBTABLE(tzm) { 558 {0x5f, "tzm"}, 559 {0x7c5f, "tzm_Latn"}, 560 {0x085f, "tzm_Latn_DZ"} 561}; 562 563ILCID_POSIX_ELEMENT_ARRAY(0x0480, ug, ug_CN) 564ILCID_POSIX_ELEMENT_ARRAY(0x0422, uk, uk_UA) 565 566ILCID_POSIX_SUBTABLE(ur) { 567 {0x20, "ur"}, 568 {0x0820, "ur_IN"}, 569 {0x0420, "ur_PK"} 570}; 571 572ILCID_POSIX_SUBTABLE(uz) { 573 {0x43, "uz"}, 574 {0x0843, "uz_Cyrl_UZ"}, /* Cyrillic based */ 575 {0x7843, "uz_Cyrl"}, /* Cyrillic based */ 576 {0x0843, "uz_UZ"}, /* Cyrillic based */ 577 {0x0443, "uz_Latn_UZ"}, /* Latin based */ 578 {0x7c43, "uz_Latn"} /* Latin based */ 579}; 580 581ILCID_POSIX_ELEMENT_ARRAY(0x0433, ve, ve_ZA) /* TODO: Verify the country */ 582ILCID_POSIX_ELEMENT_ARRAY(0x042a, vi, vi_VN) 583 584ILCID_POSIX_SUBTABLE(wen) { 585 {0x2E, "wen"}, 586 {0x042E, "wen_DE"}, 587 {0x042E, "hsb_DE"}, 588 {0x082E, "dsb_DE"}, 589 {0x7C2E, "dsb"}, 590 {0x2E, "hsb"} 591}; 592 593ILCID_POSIX_ELEMENT_ARRAY(0x0488, wo, wo_SN) 594ILCID_POSIX_ELEMENT_ARRAY(0x0434, xh, xh_ZA) 595ILCID_POSIX_ELEMENT_ARRAY(0x046a, yo, yo_NG) 596 597ILCID_POSIX_SUBTABLE(zh) { 598 {0x0004, "zh_Hans"}, 599 {0x7804, "zh"}, 600 {0x0804, "zh_CN"}, 601 {0x0804, "zh_Hans_CN"}, 602 {0x0c04, "zh_Hant_HK"}, 603 {0x0c04, "zh_HK"}, 604 {0x1404, "zh_Hant_MO"}, 605 {0x1404, "zh_MO"}, 606 {0x1004, "zh_Hans_SG"}, 607 {0x1004, "zh_SG"}, 608 {0x0404, "zh_Hant_TW"}, 609 {0x7c04, "zh_Hant"}, 610 {0x0404, "zh_TW"}, 611 {0x30404,"zh_Hant_TW"}, /* Bopomofo order */ 612 {0x30404,"zh_TW"}, /* Bopomofo order */ 613 {0x20004,"zh@collation=stroke"}, 614 {0x20404,"zh_Hant@collation=stroke"}, 615 {0x20404,"zh_Hant_TW@collation=stroke"}, 616 {0x20404,"zh_TW@collation=stroke"}, 617 {0x20804,"zh_Hans@collation=stroke"}, 618 {0x20804,"zh_Hans_CN@collation=stroke"}, 619 {0x20804,"zh_CN@collation=stroke"} 620}; 621 622ILCID_POSIX_ELEMENT_ARRAY(0x0435, zu, zu_ZA) 623 624/* This must be static and grouped by LCID. */ 625 626/* non-existent ISO-639-2 codes */ 627/* 6280x466 Edo 6290x467 Fulfulde - Nigeria 6300x486 K'iche - Guatemala 6310x430 Sutu 632*/ 633static const ILcidPosixMap gPosixIDmap[] = { 634 ILCID_POSIX_MAP(af), /* af Afrikaans 0x36 */ 635 ILCID_POSIX_MAP(am), /* am Amharic 0x5e */ 636 ILCID_POSIX_MAP(ar), /* ar Arabic 0x01 */ 637 ILCID_POSIX_MAP(arn), /* arn Araucanian/Mapudungun 0x7a */ 638 ILCID_POSIX_MAP(as), /* as Assamese 0x4d */ 639 ILCID_POSIX_MAP(az), /* az Azerbaijani 0x2c */ 640 ILCID_POSIX_MAP(ba), /* ba Bashkir 0x6d */ 641 ILCID_POSIX_MAP(be), /* be Belarusian 0x23 */ 642/* ILCID_POSIX_MAP(ber), ber Berber/Tamazight 0x5f */ 643 ILCID_POSIX_MAP(bg), /* bg Bulgarian 0x02 */ 644 ILCID_POSIX_MAP(bn), /* bn Bengali; Bangla 0x45 */ 645 ILCID_POSIX_MAP(bo), /* bo Tibetan 0x51 */ 646 ILCID_POSIX_MAP(br), /* br Breton 0x7e */ 647 ILCID_POSIX_MAP(ca), /* ca Catalan 0x03 */ 648 ILCID_POSIX_MAP(chr), /* chr Cherokee 0x5c */ 649 ILCID_POSIX_MAP(co), /* co Corsican 0x83 */ 650 ILCID_POSIX_MAP(cs), /* cs Czech 0x05 */ 651 ILCID_POSIX_MAP(cy), /* cy Welsh 0x52 */ 652 ILCID_POSIX_MAP(da), /* da Danish 0x06 */ 653 ILCID_POSIX_MAP(de), /* de German 0x07 */ 654 ILCID_POSIX_MAP(dv), /* dv Divehi 0x65 */ 655 ILCID_POSIX_MAP(el), /* el Greek 0x08 */ 656 ILCID_POSIX_MAP(en), /* en English 0x09 */ 657 ILCID_POSIX_MAP(en_US_POSIX), /* invariant 0x7f */ 658 ILCID_POSIX_MAP(es), /* es Spanish 0x0a */ 659 ILCID_POSIX_MAP(et), /* et Estonian 0x25 */ 660 ILCID_POSIX_MAP(eu), /* eu Basque 0x2d */ 661 ILCID_POSIX_MAP(fa), /* fa Persian/Farsi 0x29 */ 662 ILCID_POSIX_MAP(fa_AF), /* fa Persian/Dari 0x8c */ 663 ILCID_POSIX_MAP(fi), /* fi Finnish 0x0b */ 664 ILCID_POSIX_MAP(fil), /* fil Filipino 0x64 */ 665 ILCID_POSIX_MAP(fo), /* fo Faroese 0x38 */ 666 ILCID_POSIX_MAP(fr), /* fr French 0x0c */ 667 ILCID_POSIX_MAP(fy), /* fy Frisian 0x62 */ 668 ILCID_POSIX_MAP(ga), /* * Gaelic (Ireland,Scotland) 0x3c */ 669 ILCID_POSIX_MAP(gd), /* gd Gaelic (United Kingdom) 0x91 */ 670 ILCID_POSIX_MAP(gl), /* gl Galician 0x56 */ 671 ILCID_POSIX_MAP(gn), /* gn Guarani 0x74 */ 672 ILCID_POSIX_MAP(gsw), /* gsw Alemanic/Alsatian/Swiss German 0x84 */ 673 ILCID_POSIX_MAP(gu), /* gu Gujarati 0x47 */ 674 ILCID_POSIX_MAP(ha), /* ha Hausa 0x68 */ 675 ILCID_POSIX_MAP(haw), /* haw Hawaiian 0x75 */ 676 ILCID_POSIX_MAP(he), /* he Hebrew (formerly iw) 0x0d */ 677 ILCID_POSIX_MAP(hi), /* hi Hindi 0x39 */ 678 ILCID_POSIX_MAP(hr), /* * Croatian and others 0x1a */ 679 ILCID_POSIX_MAP(hu), /* hu Hungarian 0x0e */ 680 ILCID_POSIX_MAP(hy), /* hy Armenian 0x2b */ 681 ILCID_POSIX_MAP(id), /* id Indonesian (formerly in) 0x21 */ 682 ILCID_POSIX_MAP(ig), /* ig Igbo 0x70 */ 683 ILCID_POSIX_MAP(ii), /* ii Sichuan Yi 0x78 */ 684 ILCID_POSIX_MAP(is), /* is Icelandic 0x0f */ 685 ILCID_POSIX_MAP(it), /* it Italian 0x10 */ 686 ILCID_POSIX_MAP(iu), /* iu Inuktitut 0x5d */ 687 ILCID_POSIX_MAP(iw), /* iw Hebrew 0x0d */ 688 ILCID_POSIX_MAP(ja), /* ja Japanese 0x11 */ 689 ILCID_POSIX_MAP(ka), /* ka Georgian 0x37 */ 690 ILCID_POSIX_MAP(kk), /* kk Kazakh 0x3f */ 691 ILCID_POSIX_MAP(kl), /* kl Kalaallisut 0x6f */ 692 ILCID_POSIX_MAP(km), /* km Khmer 0x53 */ 693 ILCID_POSIX_MAP(kn), /* kn Kannada 0x4b */ 694 ILCID_POSIX_MAP(ko), /* ko Korean 0x12 */ 695 ILCID_POSIX_MAP(kok), /* kok Konkani 0x57 */ 696 ILCID_POSIX_MAP(kr), /* kr Kanuri 0x71 */ 697 ILCID_POSIX_MAP(ks), /* ks Kashmiri 0x60 */ 698 ILCID_POSIX_MAP(ky), /* ky Kyrgyz 0x40 */ 699 ILCID_POSIX_MAP(lb), /* lb Luxembourgish 0x6e */ 700 ILCID_POSIX_MAP(la), /* la Latin 0x76 */ 701 ILCID_POSIX_MAP(lo), /* lo Lao 0x54 */ 702 ILCID_POSIX_MAP(lt), /* lt Lithuanian 0x27 */ 703 ILCID_POSIX_MAP(lv), /* lv Latvian, Lettish 0x26 */ 704 ILCID_POSIX_MAP(mi), /* mi Maori 0x81 */ 705 ILCID_POSIX_MAP(mk), /* mk Macedonian 0x2f */ 706 ILCID_POSIX_MAP(ml), /* ml Malayalam 0x4c */ 707 ILCID_POSIX_MAP(mn), /* mn Mongolian 0x50 */ 708 ILCID_POSIX_MAP(mni), /* mni Manipuri 0x58 */ 709 ILCID_POSIX_MAP(moh), /* moh Mohawk 0x7c */ 710 ILCID_POSIX_MAP(mr), /* mr Marathi 0x4e */ 711 ILCID_POSIX_MAP(ms), /* ms Malay 0x3e */ 712 ILCID_POSIX_MAP(mt), /* mt Maltese 0x3a */ 713 ILCID_POSIX_MAP(my), /* my Burmese 0x55 */ 714/* ILCID_POSIX_MAP(nb), // no Norwegian 0x14 */ 715 ILCID_POSIX_MAP(ne), /* ne Nepali 0x61 */ 716 ILCID_POSIX_MAP(nl), /* nl Dutch 0x13 */ 717/* ILCID_POSIX_MAP(nn), // no Norwegian 0x14 */ 718 ILCID_POSIX_MAP(no), /* * Norwegian 0x14 */ 719 ILCID_POSIX_MAP(nso), /* nso Sotho, Northern (Sepedi dialect) 0x6c */ 720 ILCID_POSIX_MAP(oc), /* oc Occitan 0x82 */ 721 ILCID_POSIX_MAP(om), /* om Oromo 0x72 */ 722 ILCID_POSIX_MAP(or_IN), /* or Oriya 0x48 */ 723 ILCID_POSIX_MAP(pa), /* pa Punjabi 0x46 */ 724 ILCID_POSIX_MAP(pl), /* pl Polish 0x15 */ 725 ILCID_POSIX_MAP(ps), /* ps Pashto 0x63 */ 726 ILCID_POSIX_MAP(pt), /* pt Portuguese 0x16 */ 727 ILCID_POSIX_MAP(qu), /* qu Quechua 0x6B */ 728 ILCID_POSIX_MAP(qut), /* qut K'iche 0x86 */ 729 ILCID_POSIX_MAP(rm), /* rm Raeto-Romance/Romansh 0x17 */ 730 ILCID_POSIX_MAP(ro), /* ro Romanian 0x18 */ 731 ILCID_POSIX_MAP(root), /* root 0x00 */ 732 ILCID_POSIX_MAP(ru), /* ru Russian 0x19 */ 733 ILCID_POSIX_MAP(rw), /* rw Kinyarwanda 0x87 */ 734 ILCID_POSIX_MAP(sa), /* sa Sanskrit 0x4f */ 735 ILCID_POSIX_MAP(sah), /* sah Yakut 0x85 */ 736 ILCID_POSIX_MAP(sd), /* sd Sindhi 0x59 */ 737 ILCID_POSIX_MAP(se), /* se Sami 0x3b */ 738/* ILCID_POSIX_MAP(sh), // sh Serbo-Croatian 0x1a */ 739 ILCID_POSIX_MAP(si), /* si Sinhalese 0x5b */ 740 ILCID_POSIX_MAP(sk), /* sk Slovak 0x1b */ 741 ILCID_POSIX_MAP(sl), /* sl Slovenian 0x24 */ 742 ILCID_POSIX_MAP(so), /* so Somali 0x77 */ 743 ILCID_POSIX_MAP(sq), /* sq Albanian 0x1c */ 744/* ILCID_POSIX_MAP(sr), // sr Serbian 0x1a */ 745 ILCID_POSIX_MAP(sv), /* sv Swedish 0x1d */ 746 ILCID_POSIX_MAP(sw), /* sw Swahili 0x41 */ 747 ILCID_POSIX_MAP(syr), /* syr Syriac 0x5A */ 748 ILCID_POSIX_MAP(ta), /* ta Tamil 0x49 */ 749 ILCID_POSIX_MAP(te), /* te Telugu 0x4a */ 750 ILCID_POSIX_MAP(tg), /* tg Tajik 0x28 */ 751 ILCID_POSIX_MAP(th), /* th Thai 0x1e */ 752 ILCID_POSIX_MAP(ti), /* ti Tigrigna 0x73 */ 753 ILCID_POSIX_MAP(tk), /* tk Turkmen 0x42 */ 754 ILCID_POSIX_MAP(tn), /* tn Tswana 0x32 */ 755 ILCID_POSIX_MAP(tr), /* tr Turkish 0x1f */ 756 ILCID_POSIX_MAP(tt), /* tt Tatar 0x44 */ 757 ILCID_POSIX_MAP(tzm), /* tzm 0x5f */ 758 ILCID_POSIX_MAP(ug), /* ug Uighur 0x80 */ 759 ILCID_POSIX_MAP(uk), /* uk Ukrainian 0x22 */ 760 ILCID_POSIX_MAP(ur), /* ur Urdu 0x20 */ 761 ILCID_POSIX_MAP(uz), /* uz Uzbek 0x43 */ 762 ILCID_POSIX_MAP(ve), /* ve Venda 0x33 */ 763 ILCID_POSIX_MAP(vi), /* vi Vietnamese 0x2a */ 764 ILCID_POSIX_MAP(wen), /* wen Sorbian 0x2e */ 765 ILCID_POSIX_MAP(wo), /* wo Wolof 0x88 */ 766 ILCID_POSIX_MAP(xh), /* xh Xhosa 0x34 */ 767 ILCID_POSIX_MAP(yo), /* yo Yoruba 0x6a */ 768 ILCID_POSIX_MAP(zh), /* zh Chinese 0x04 */ 769 ILCID_POSIX_MAP(zu), /* zu Zulu 0x35 */ 770}; 771 772static const uint32_t gLocaleCount = sizeof(gPosixIDmap)/sizeof(ILcidPosixMap); 773 774/** 775 * Do not call this function. It is called by hostID. 776 * The function is not private because this struct must stay as a C struct, 777 * and this is an internal class. 778 */ 779static int32_t 780idCmp(const char* id1, const char* id2) 781{ 782 int32_t diffIdx = 0; 783 while (*id1 == *id2 && *id1 != 0) { 784 diffIdx++; 785 id1++; 786 id2++; 787 } 788 return diffIdx; 789} 790 791/** 792 * Searches for a Windows LCID 793 * 794 * @param posixid the Posix style locale id. 795 * @param status gets set to U_ILLEGAL_ARGUMENT_ERROR when the Posix ID has 796 * no equivalent Windows LCID. 797 * @return the LCID 798 */ 799static uint32_t 800getHostID(const ILcidPosixMap *this_0, const char* posixID, UErrorCode* status) 801{ 802 int32_t bestIdx = 0; 803 int32_t bestIdxDiff = 0; 804 int32_t posixIDlen = (int32_t)uprv_strlen(posixID); 805 uint32_t idx; 806 807 for (idx = 0; idx < this_0->numRegions; idx++ ) { 808 int32_t sameChars = idCmp(posixID, this_0->regionMaps[idx].posixID); 809 if (sameChars > bestIdxDiff && this_0->regionMaps[idx].posixID[sameChars] == 0) { 810 if (posixIDlen == sameChars) { 811 /* Exact match */ 812 return this_0->regionMaps[idx].hostID; 813 } 814 bestIdxDiff = sameChars; 815 bestIdx = idx; 816 } 817 } 818 /* We asked for something unusual, like en_ZZ, and we try to return the number for the same language. */ 819 /* We also have to make sure that sid and si and similar string subsets don't match. */ 820 if ((posixID[bestIdxDiff] == '_' || posixID[bestIdxDiff] == '@') 821 && this_0->regionMaps[bestIdx].posixID[bestIdxDiff] == 0) 822 { 823 *status = U_USING_FALLBACK_WARNING; 824 return this_0->regionMaps[bestIdx].hostID; 825 } 826 827 /*no match found */ 828 *status = U_ILLEGAL_ARGUMENT_ERROR; 829 return this_0->regionMaps->hostID; 830} 831 832static const char* 833getPosixID(const ILcidPosixMap *this_0, uint32_t hostID) 834{ 835 uint32_t i; 836 for (i = 0; i <= this_0->numRegions; i++) 837 { 838 if (this_0->regionMaps[i].hostID == hostID) 839 { 840 return this_0->regionMaps[i].posixID; 841 } 842 } 843 844 /* If you get here, then no matching region was found, 845 so return the language id with the wild card region. */ 846 return this_0->regionMaps[0].posixID; 847} 848 849/* 850////////////////////////////////////// 851// 852// LCID --> POSIX 853// 854///////////////////////////////////// 855*/ 856#ifdef USE_WINDOWS_LOCALE_API 857/* 858 * Change the tag separator from '-' to '_' 859 */ 860#define FIX_LOCALE_ID_TAG_SEPARATOR(buffer, len, i) \ 861 for(i = 0; i < len; i++) \ 862 if (buffer[i] == '-') buffer[i] = '_'; 863 864/* 865 * Various language tags needs to be changed: 866 * quz -> qu 867 * prs -> fa 868 */ 869#define FIX_LANGUAGE_ID_TAG(buffer, len) \ 870 if (len >= 3) { \ 871 if (buffer[0] == 'q' && buffer[1] == 'u' && buffer[2] == 'z') {\ 872 buffer[2] = 0; \ 873 uprv_strcat(buffer, buffer+3); \ 874 } else if (buffer[0] == 'p' && buffer[1] == 'r' && buffer[2] == 's') {\ 875 buffer[0] = 'f'; buffer[1] = 'a'; buffer[2] = 0; \ 876 uprv_strcat(buffer, buffer+3); \ 877 } \ 878 } 879 880static char gPosixFromLCID[ULOC_FULLNAME_CAPACITY]; 881#endif 882U_CAPI const char * 883uprv_convertToPosix(uint32_t hostid, UErrorCode* status) 884{ 885 uint16_t langID; 886 uint32_t localeIndex; 887#ifdef USE_WINDOWS_LOCALE_API 888 int32_t ret = 0; 889 890 uprv_memset(gPosixFromLCID, 0, sizeof(gPosixFromLCID)); 891 892 ret = GetLocaleInfoA(hostid, LOCALE_SNAME, (LPSTR)gPosixFromLCID, sizeof(gPosixFromLCID)); 893 if (ret > 1) { 894 FIX_LOCALE_ID_TAG_SEPARATOR(gPosixFromLCID, ret, localeIndex) 895 FIX_LANGUAGE_ID_TAG(gPosixFromLCID, ret) 896 897 return gPosixFromLCID; 898 } 899#endif 900 langID = LANGUAGE_LCID(hostid); 901 902 for (localeIndex = 0; localeIndex < gLocaleCount; localeIndex++) 903 { 904 if (langID == gPosixIDmap[localeIndex].regionMaps->hostID) 905 { 906 return getPosixID(&gPosixIDmap[localeIndex], hostid); 907 } 908 } 909 910 /* no match found */ 911 *status = U_ILLEGAL_ARGUMENT_ERROR; 912 return NULL; 913} 914 915/* 916////////////////////////////////////// 917// 918// POSIX --> LCID 919// This should only be called from uloc_getLCID. 920// The locale ID must be in canonical form. 921// langID is separate so that this file doesn't depend on the uloc_* API. 922// 923///////////////////////////////////// 924*/ 925 926U_CAPI uint32_t 927uprv_convertToLCID(const char *langID, const char* posixID, UErrorCode* status) 928{ 929 930 uint32_t low = 0; 931 uint32_t high = gLocaleCount; 932 uint32_t mid = high; 933 uint32_t oldmid = 0; 934 int32_t compVal; 935 936 uint32_t value = 0; 937 uint32_t fallbackValue = (uint32_t)-1; 938 UErrorCode myStatus; 939 uint32_t idx; 940 941 /* Check for incomplete id. */ 942 if (!langID || !posixID || uprv_strlen(langID) < 2 || uprv_strlen(posixID) < 2) { 943 return 0; 944 } 945 946 /*Binary search for the map entry for normal cases */ 947 948 while (high > low) /*binary search*/{ 949 950 mid = (high+low) >> 1; /*Finds median*/ 951 952 if (mid == oldmid) 953 break; 954 955 compVal = uprv_strcmp(langID, gPosixIDmap[mid].regionMaps->posixID); 956 if (compVal < 0){ 957 high = mid; 958 } 959 else if (compVal > 0){ 960 low = mid; 961 } 962 else /*we found it*/{ 963 return getHostID(&gPosixIDmap[mid], posixID, status); 964 } 965 oldmid = mid; 966 } 967 968 /* 969 * Sometimes we can't do a binary search on posixID because some LCIDs 970 * go to different locales. We hit one of those special cases. 971 */ 972 for (idx = 0; idx < gLocaleCount; idx++ ) { 973 myStatus = U_ZERO_ERROR; 974 value = getHostID(&gPosixIDmap[idx], posixID, &myStatus); 975 if (myStatus == U_ZERO_ERROR) { 976 return value; 977 } 978 else if (myStatus == U_USING_FALLBACK_WARNING) { 979 fallbackValue = value; 980 } 981 } 982 983 if (fallbackValue != (uint32_t)-1) { 984 *status = U_USING_FALLBACK_WARNING; 985 return fallbackValue; 986 } 987 988 /* no match found */ 989 *status = U_ILLEGAL_ARGUMENT_ERROR; 990 return 0; /* return international (root) */ 991} 992 993