1/*
2 **********************************************************************
3 *   Copyright (C) 1996-2012, 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 U_PLATFORM == U_PF_WINDOWS && defined(_MSC_VER) && (_MSC_VER >= 1500)
34/*
35 * TODO: It seems like we should widen this to
36 * either U_PLATFORM_USES_ONLY_WIN32_API (includes MinGW)
37 * or U_PLATFORM_HAS_WIN32_API (includes MinGW and Cygwin)
38 * but those use gcc and won't have defined(_MSC_VER).
39 * We might need to #include some Windows header and test for some version macro from there.
40 * Or call some Windows function and see what it returns.
41 */
42#define USE_WINDOWS_LOCALE_API
43#endif
44
45#ifdef USE_WINDOWS_LOCALE_API
46#include <windows.h>
47#include <winnls.h>
48#endif
49
50/*
51 * Note:
52 * The mapping from Win32 locale ID numbers to POSIX locale strings should
53 * be the faster one.
54 *
55 * Many LCID values come from winnt.h
56 * Some also come from http://www.microsoft.com/globaldev/reference/lcid-all.mspx
57 */
58
59/*
60////////////////////////////////////////////////
61//
62// Internal Classes for LCID <--> POSIX Mapping
63//
64/////////////////////////////////////////////////
65*/
66
67typedef struct ILcidPosixElement
68{
69    const uint32_t hostID;
70    const char * const posixID;
71} ILcidPosixElement;
72
73typedef struct ILcidPosixMap
74{
75    const uint32_t numRegions;
76    const struct ILcidPosixElement* const regionMaps;
77} ILcidPosixMap;
78
79
80/*
81/////////////////////////////////////////////////
82//
83// Easy macros to make the LCID <--> POSIX Mapping
84//
85/////////////////////////////////////////////////
86*/
87
88/**
89 * The standard one language/one country mapping for LCID.
90 * The first element must be the language, and the following
91 * elements are the language with the country.
92 * @param hostID LCID in host format such as 0x044d
93 * @param languageID posix ID of just the language such as 'de'
94 * @param posixID posix ID of the language_TERRITORY such as 'de_CH'
95 */
96#define ILCID_POSIX_ELEMENT_ARRAY(hostID, languageID, posixID) \
97static const ILcidPosixElement locmap_ ## languageID [] = { \
98    {LANGUAGE_LCID(hostID), #languageID},     /* parent locale */ \
99    {hostID, #posixID}, \
100};
101
102/**
103 * Define a subtable by ID
104 * @param id the POSIX ID, either a language or language_TERRITORY
105 */
106#define ILCID_POSIX_SUBTABLE(id) \
107static const ILcidPosixElement locmap_ ## id [] =
108
109
110/**
111 * Create the map for the posixID. This macro supposes that the language string
112 * name is the same as the global variable name, and that the first element
113 * in the ILcidPosixElement is just the language.
114 * @param _posixID the full POSIX ID for this entry.
115 */
116#define ILCID_POSIX_MAP(_posixID) \
117    {sizeof(locmap_ ## _posixID)/sizeof(ILcidPosixElement), locmap_ ## _posixID}
118
119/*
120////////////////////////////////////////////
121//
122// Create the table of LCID to POSIX Mapping
123// None of it should be dynamically created.
124//
125// Keep static locale variables inside the function so that
126// it can be created properly during static init.
127//
128// Note: This table should be updated periodically. Check the National Lanaguage Support API Reference Website.
129//       Microsoft is moving away from LCID in favor of locale name as of Vista.  This table needs to be
130//       maintained for support of older Windows version.
131//       Update: Windows 7 (091130)
132////////////////////////////////////////////
133*/
134
135ILCID_POSIX_ELEMENT_ARRAY(0x0436, af, af_ZA)
136
137ILCID_POSIX_SUBTABLE(ar) {
138    {0x01,   "ar"},
139    {0x3801, "ar_AE"},
140    {0x3c01, "ar_BH"},
141    {0x1401, "ar_DZ"},
142    {0x0c01, "ar_EG"},
143    {0x0801, "ar_IQ"},
144    {0x2c01, "ar_JO"},
145    {0x3401, "ar_KW"},
146    {0x3001, "ar_LB"},
147    {0x1001, "ar_LY"},
148    {0x1801, "ar_MA"},
149    {0x1801, "ar_MO"},
150    {0x2001, "ar_OM"},
151    {0x4001, "ar_QA"},
152    {0x0401, "ar_SA"},
153    {0x2801, "ar_SY"},
154    {0x1c01, "ar_TN"},
155    {0x2401, "ar_YE"}
156};
157
158ILCID_POSIX_ELEMENT_ARRAY(0x044d, as, as_IN)
159ILCID_POSIX_ELEMENT_ARRAY(0x045e, am, am_ET)
160ILCID_POSIX_ELEMENT_ARRAY(0x047a, arn,arn_CL)
161
162ILCID_POSIX_SUBTABLE(az) {
163    {0x2c,   "az"},
164    {0x082c, "az_Cyrl_AZ"},  /* Cyrillic based */
165    {0x742c, "az_Cyrl"},  /* Cyrillic based */
166    {0x042c, "az_Latn_AZ"}, /* Latin based */
167    {0x782c, "az_Latn"}, /* Latin based */
168    {0x042c, "az_AZ"} /* Latin based */
169};
170
171ILCID_POSIX_ELEMENT_ARRAY(0x046d, ba, ba_RU)
172ILCID_POSIX_ELEMENT_ARRAY(0x0423, be, be_BY)
173
174/*ILCID_POSIX_SUBTABLE(ber) {
175    {0x5f,   "ber"},
176    {0x045f, "ber_Arab_DZ"},
177    {0x045f, "ber_Arab"},
178    {0x085f, "ber_Latn_DZ"},
179    {0x085f, "ber_Latn"}
180};*/
181
182ILCID_POSIX_ELEMENT_ARRAY(0x0402, bg, bg_BG)
183
184ILCID_POSIX_ELEMENT_ARRAY(0x0466, bin, bin_NG)
185
186ILCID_POSIX_SUBTABLE(bn) {
187    {0x45,   "bn"},
188    {0x0845, "bn_BD"},
189    {0x0445, "bn_IN"}
190};
191
192ILCID_POSIX_SUBTABLE(bo) {
193    {0x51,   "bo"},
194    {0x0851, "bo_BT"},
195    {0x0451, "bo_CN"}
196};
197
198ILCID_POSIX_ELEMENT_ARRAY(0x047e, br, br_FR)
199ILCID_POSIX_ELEMENT_ARRAY(0x0403, ca, ca_ES)
200ILCID_POSIX_ELEMENT_ARRAY(0x0483, co, co_FR)
201ILCID_POSIX_ELEMENT_ARRAY(0x045c, chr,chr_US)
202
203/* Declared as cs_CZ to get around compiler errors on z/OS, which defines cs as a function */
204ILCID_POSIX_ELEMENT_ARRAY(0x0405, cs, cs_CZ)
205
206ILCID_POSIX_ELEMENT_ARRAY(0x0452, cy, cy_GB)
207ILCID_POSIX_ELEMENT_ARRAY(0x0406, da, da_DK)
208
209ILCID_POSIX_SUBTABLE(de) {
210    {0x07,   "de"},
211    {0x0c07, "de_AT"},
212    {0x0807, "de_CH"},
213    {0x0407, "de_DE"},
214    {0x1407, "de_LI"},
215    {0x1007, "de_LU"},
216    {0x10407,"de_DE@collation=phonebook"},  /*This is really de_DE_PHONEBOOK on Windows*/
217    {0x10407,"de@collation=phonebook"}  /*This is really de_DE_PHONEBOOK on Windows*/
218};
219
220ILCID_POSIX_ELEMENT_ARRAY(0x0465, dv, dv_MV)
221ILCID_POSIX_ELEMENT_ARRAY(0x0408, el, el_GR)
222
223ILCID_POSIX_SUBTABLE(en) {
224    {0x09,   "en"},
225    {0x0c09, "en_AU"},
226    {0x2809, "en_BZ"},
227    {0x1009, "en_CA"},
228    {0x0809, "en_GB"},
229    {0x3c09, "en_HK"},
230    {0x3809, "en_ID"},
231    {0x1809, "en_IE"},
232    {0x4009, "en_IN"},
233    {0x2009, "en_JM"},
234    {0x4409, "en_MY"},
235    {0x1409, "en_NZ"},
236    {0x3409, "en_PH"},
237    {0x4809, "en_SG"},
238    {0x2C09, "en_TT"},
239    {0x0409, "en_US"},
240    {0x007f, "en_US_POSIX"}, /* duplicate for roundtripping */
241    {0x2409, "en_VI"},  /* Virgin Islands AKA Caribbean Islands (en_CB). */
242    {0x1c09, "en_ZA"},
243    {0x3009, "en_ZW"},
244    {0x2409, "en_029"},
245    {0x0409, "en_AS"},  /* Alias for en_US. Leave last. */
246    {0x0409, "en_GU"},  /* Alias for en_US. Leave last. */
247    {0x0409, "en_MH"},  /* Alias for en_US. Leave last. */
248    {0x0409, "en_MP"},  /* Alias for en_US. Leave last. */
249    {0x0409, "en_UM"}   /* Alias for en_US. Leave last. */
250};
251
252ILCID_POSIX_SUBTABLE(en_US_POSIX) {
253    {0x007f, "en_US_POSIX"} /* duplicate for roundtripping */
254};
255
256ILCID_POSIX_SUBTABLE(es) {
257    {0x0a,   "es"},
258    {0x2c0a, "es_AR"},
259    {0x400a, "es_BO"},
260    {0x340a, "es_CL"},
261    {0x240a, "es_CO"},
262    {0x140a, "es_CR"},
263    {0x1c0a, "es_DO"},
264    {0x300a, "es_EC"},
265    {0x0c0a, "es_ES"},      /*Modern sort.*/
266    {0x100a, "es_GT"},
267    {0x480a, "es_HN"},
268    {0x080a, "es_MX"},
269    {0x4c0a, "es_NI"},
270    {0x180a, "es_PA"},
271    {0x280a, "es_PE"},
272    {0x500a, "es_PR"},
273    {0x3c0a, "es_PY"},
274    {0x440a, "es_SV"},
275    {0x540a, "es_US"},
276    {0x380a, "es_UY"},
277    {0x200a, "es_VE"},
278    {0xe40a, "es_419"},
279    {0x040a, "es_ES@collation=traditional"},
280    {0x040a, "es@collation=traditional"}
281};
282
283ILCID_POSIX_ELEMENT_ARRAY(0x0425, et, et_EE)
284ILCID_POSIX_ELEMENT_ARRAY(0x042d, eu, eu_ES)
285
286/* ISO-639 doesn't distinguish between Persian and Dari.*/
287ILCID_POSIX_SUBTABLE(fa) {
288    {0x29,   "fa"},
289    {0x0429, "fa_IR"},  /* Persian/Farsi (Iran) */
290    {0x048c, "fa_AF"}   /* Persian/Dari (Afghanistan) */
291};
292
293/* duplicate for roundtripping */
294ILCID_POSIX_SUBTABLE(fa_AF) {
295    {0x8c,   "fa_AF"},  /* Persian/Dari (Afghanistan) */
296    {0x048c, "fa_AF"}   /* Persian/Dari (Afghanistan) */
297};
298
299ILCID_POSIX_ELEMENT_ARRAY(0x040b, fi, fi_FI)
300ILCID_POSIX_ELEMENT_ARRAY(0x0464, fil,fil_PH)
301ILCID_POSIX_ELEMENT_ARRAY(0x0438, fo, fo_FO)
302
303ILCID_POSIX_SUBTABLE(fr) {
304    {0x0c,   "fr"},
305    {0x080c, "fr_BE"},
306    {0x0c0c, "fr_CA"},
307    {0x240c, "fr_CD"},
308    {0x240c, "fr_CG"},
309    {0x100c, "fr_CH"},
310    {0x300c, "fr_CI"},
311    {0x2c0c, "fr_CM"},
312    {0x040c, "fr_FR"},
313    {0x3c0c, "fr_HT"},
314    {0x140c, "fr_LU"},
315    {0x380c, "fr_MA"},
316    {0x180c, "fr_MC"},
317    {0x340c, "fr_ML"},
318    {0x200c, "fr_RE"},
319    {0x280c, "fr_SN"},
320    {0xe40c, "fr_015"},
321    {0x1c0c, "fr_029"}
322};
323
324ILCID_POSIX_ELEMENT_ARRAY(0x0467, fuv, fuv_NG)
325
326ILCID_POSIX_ELEMENT_ARRAY(0x0462, fy, fy_NL)
327
328ILCID_POSIX_SUBTABLE(ga) { /* Gaelic (Ireland) */
329    {0x3c,   "ga"},
330    {0x083c, "ga_IE"},
331    {0x043c, "gd_GB"}
332};
333
334ILCID_POSIX_SUBTABLE(gd) { /* Gaelic (Scotland) */
335    {0x91,   "gd"},
336    {0x0491, "gd_GB"}
337};
338
339ILCID_POSIX_ELEMENT_ARRAY(0x0456, gl, gl_ES)
340ILCID_POSIX_ELEMENT_ARRAY(0x0447, gu, gu_IN)
341ILCID_POSIX_ELEMENT_ARRAY(0x0474, gn, gn_PY)
342ILCID_POSIX_ELEMENT_ARRAY(0x0484, gsw,gsw_FR)
343
344ILCID_POSIX_SUBTABLE(ha) {
345    {0x68,   "ha"},
346    {0x7c68, "ha_Latn"},
347    {0x0468, "ha_Latn_NG"},
348};
349
350ILCID_POSIX_ELEMENT_ARRAY(0x0475, haw,haw_US)
351ILCID_POSIX_ELEMENT_ARRAY(0x040d, he, he_IL)
352ILCID_POSIX_ELEMENT_ARRAY(0x0439, hi, hi_IN)
353
354/* This LCID is really four different locales.*/
355ILCID_POSIX_SUBTABLE(hr) {
356    {0x1a,   "hr"},
357    {0x141a, "bs_Latn_BA"},  /* Bosnian, Bosnia and Herzegovina */
358    {0x681a, "bs_Latn"},  /* Bosnian, Bosnia and Herzegovina */
359    {0x141a, "bs_BA"},  /* Bosnian, Bosnia and Herzegovina */
360    {0x781a, "bs"},     /* Bosnian */
361    {0x201a, "bs_Cyrl_BA"},  /* Bosnian, Bosnia and Herzegovina */
362    {0x641a, "bs_Cyrl"},  /* Bosnian, Bosnia and Herzegovina */
363    {0x101a, "hr_BA"},  /* Croatian in Bosnia */
364    {0x041a, "hr_HR"},  /* Croatian*/
365    {0x2c1a, "sr_Latn_ME"},
366    {0x241a, "sr_Latn_RS"},
367    {0x181a, "sr_Latn_BA"}, /* Serbo-Croatian in Bosnia */
368    {0x081a, "sr_Latn_CS"}, /* Serbo-Croatian*/
369    {0x701a, "sr_Latn"},    /* It's 0x1a or 0x081a, pick one to make the test program happy. */
370    {0x1c1a, "sr_Cyrl_BA"}, /* Serbo-Croatian in Bosnia */
371    {0x0c1a, "sr_Cyrl_CS"}, /* Serbian*/
372    {0x301a, "sr_Cyrl_ME"},
373    {0x281a, "sr_Cyrl_RS"},
374    {0x6c1a, "sr_Cyrl"},    /* It's 0x1a or 0x0c1a, pick one to make the test program happy. */
375    {0x7c1a, "sr"}          /* In CLDR sr is sr_Cyrl. */
376};
377
378ILCID_POSIX_ELEMENT_ARRAY(0x040e, hu, hu_HU)
379ILCID_POSIX_ELEMENT_ARRAY(0x042b, hy, hy_AM)
380ILCID_POSIX_ELEMENT_ARRAY(0x0469, ibb, ibb_NG)
381ILCID_POSIX_ELEMENT_ARRAY(0x0421, id, id_ID)
382ILCID_POSIX_ELEMENT_ARRAY(0x0470, ig, ig_NG)
383ILCID_POSIX_ELEMENT_ARRAY(0x0478, ii, ii_CN)
384ILCID_POSIX_ELEMENT_ARRAY(0x040f, is, is_IS)
385
386ILCID_POSIX_SUBTABLE(it) {
387    {0x10,   "it"},
388    {0x0810, "it_CH"},
389    {0x0410, "it_IT"}
390};
391
392ILCID_POSIX_SUBTABLE(iu) {
393    {0x5d,   "iu"},
394    {0x045d, "iu_Cans_CA"},
395    {0x785d, "iu_Cans"},
396    {0x085d, "iu_Latn_CA"},
397    {0x7c5d, "iu_Latn"}
398};
399
400ILCID_POSIX_ELEMENT_ARRAY(0x040d, iw, iw_IL)    /*Left in for compatibility*/
401ILCID_POSIX_ELEMENT_ARRAY(0x0411, ja, ja_JP)
402ILCID_POSIX_ELEMENT_ARRAY(0x0437, ka, ka_GE)
403ILCID_POSIX_ELEMENT_ARRAY(0x043f, kk, kk_KZ)
404ILCID_POSIX_ELEMENT_ARRAY(0x046f, kl, kl_GL)
405ILCID_POSIX_ELEMENT_ARRAY(0x0453, km, km_KH)
406ILCID_POSIX_ELEMENT_ARRAY(0x044b, kn, kn_IN)
407
408ILCID_POSIX_SUBTABLE(ko) {
409    {0x12,   "ko"},
410    {0x0812, "ko_KP"},
411    {0x0412, "ko_KR"}
412};
413
414ILCID_POSIX_ELEMENT_ARRAY(0x0457, kok, kok_IN)
415ILCID_POSIX_ELEMENT_ARRAY(0x0471, kr,  kr_NG)
416
417ILCID_POSIX_SUBTABLE(ks) {         /* We could add PK and CN too */
418    {0x60,   "ks"},
419    {0x0860, "ks_IN"},              /* Documentation doesn't mention script */
420    {0x0460, "ks_Arab_IN"},
421    {0x0860, "ks_Deva_IN"}
422};
423
424ILCID_POSIX_ELEMENT_ARRAY(0x0440, ky, ky_KG)   /* Kyrgyz is spoken in Kyrgyzstan */
425ILCID_POSIX_ELEMENT_ARRAY(0x0476, la, la_IT)   /* TODO: Verify the country */
426ILCID_POSIX_ELEMENT_ARRAY(0x046e, lb, lb_LU)
427ILCID_POSIX_ELEMENT_ARRAY(0x0454, lo, lo_LA)
428ILCID_POSIX_ELEMENT_ARRAY(0x0427, lt, lt_LT)
429ILCID_POSIX_ELEMENT_ARRAY(0x0426, lv, lv_LV)
430ILCID_POSIX_ELEMENT_ARRAY(0x0481, mi, mi_NZ)
431ILCID_POSIX_ELEMENT_ARRAY(0x042f, mk, mk_MK)
432ILCID_POSIX_ELEMENT_ARRAY(0x044c, ml, ml_IN)
433
434ILCID_POSIX_SUBTABLE(mn) {
435    {0x50,   "mn"},
436    {0x0450, "mn_MN"},
437    {0x7c50, "mn_Mong"},
438    {0x0850, "mn_Mong_CN"},
439    {0x0850, "mn_CN"},
440    {0x7850, "mn_Cyrl"}
441};
442
443ILCID_POSIX_ELEMENT_ARRAY(0x0458, mni,mni_IN)
444ILCID_POSIX_ELEMENT_ARRAY(0x047c, moh,moh_CA)
445ILCID_POSIX_ELEMENT_ARRAY(0x044e, mr, mr_IN)
446
447ILCID_POSIX_SUBTABLE(ms) {
448    {0x3e,   "ms"},
449    {0x083e, "ms_BN"},   /* Brunei Darussalam*/
450    {0x043e, "ms_MY"}    /* Malaysia*/
451};
452
453ILCID_POSIX_ELEMENT_ARRAY(0x043a, mt, mt_MT)
454ILCID_POSIX_ELEMENT_ARRAY(0x0455, my, my_MM)
455
456ILCID_POSIX_SUBTABLE(ne) {
457    {0x61,   "ne"},
458    {0x0861, "ne_IN"},   /* India*/
459    {0x0461, "ne_NP"}    /* Nepal*/
460};
461
462ILCID_POSIX_SUBTABLE(nl) {
463    {0x13,   "nl"},
464    {0x0813, "nl_BE"},
465    {0x0413, "nl_NL"}
466};
467
468/* The "no" locale split into nb and nn.  By default in ICU, "no" is nb.*/
469ILCID_POSIX_SUBTABLE(no) {
470    {0x14,   "no"},     /* really nb_NO */
471    {0x7c14, "nb"},     /* really nb */
472    {0x0414, "nb_NO"},  /* really nb_NO. Keep first in the 414 list. */
473    {0x0414, "no_NO"},  /* really nb_NO */
474    {0x0814, "nn_NO"},  /* really nn_NO. Keep first in the 814 list.  */
475    {0x7814, "nn"},     /* It's 0x14 or 0x814, pick one to make the test program happy. */
476    {0x0814, "no_NO_NY"}/* really nn_NO */
477};
478
479ILCID_POSIX_ELEMENT_ARRAY(0x046c, nso,nso_ZA)   /* TODO: Verify the ISO-639 code */
480ILCID_POSIX_ELEMENT_ARRAY(0x0482, oc, oc_FR)
481
482ILCID_POSIX_SUBTABLE(om) { /* TODO: Verify the country */
483    {0x72,   "om"},
484    {0x0472, "om_ET"},
485    {0x0472, "gaz_ET"}
486};
487
488/* Declared as or_IN to get around compiler errors*/
489ILCID_POSIX_SUBTABLE(or_IN) {
490    {0x48,   "or"},
491    {0x0448, "or_IN"},
492};
493
494
495ILCID_POSIX_SUBTABLE(pa) {
496    {0x46,   "pa"},
497    {0x0446, "pa_IN"},
498    {0x0846, "pa_PK"}
499};
500
501ILCID_POSIX_ELEMENT_ARRAY(0x0479, pap, pap_AN)
502ILCID_POSIX_ELEMENT_ARRAY(0x0415, pl, pl_PL)
503ILCID_POSIX_ELEMENT_ARRAY(0x0463, ps, ps_AF)
504
505ILCID_POSIX_SUBTABLE(pt) {
506    {0x16,   "pt"},
507    {0x0416, "pt_BR"},
508    {0x0816, "pt_PT"}
509};
510
511ILCID_POSIX_SUBTABLE(qu) {
512    {0x6b,   "qu"},
513    {0x046b, "qu_BO"},
514    {0x086b, "qu_EC"},
515    {0x0C6b, "qu_PE"},
516    {0x046b, "quz_BO"},
517    {0x086b, "quz_EC"},
518    {0x0C6b, "quz_PE"}
519};
520
521ILCID_POSIX_ELEMENT_ARRAY(0x0486, qut, qut_GT) /* qut is an ISO-639-3 code */
522ILCID_POSIX_ELEMENT_ARRAY(0x0417, rm, rm_CH)
523
524ILCID_POSIX_SUBTABLE(ro) {
525    {0x18,   "ro"},
526    {0x0418, "ro_RO"},
527    {0x0818, "ro_MD"}
528};
529
530ILCID_POSIX_SUBTABLE(root) {
531    {0x00,   "root"}
532};
533
534ILCID_POSIX_SUBTABLE(ru) {
535    {0x19,   "ru"},
536    {0x0419, "ru_RU"},
537    {0x0819, "ru_MD"}
538};
539
540ILCID_POSIX_ELEMENT_ARRAY(0x0487, rw, rw_RW)
541ILCID_POSIX_ELEMENT_ARRAY(0x044f, sa, sa_IN)
542ILCID_POSIX_ELEMENT_ARRAY(0x0485, sah,sah_RU)
543
544ILCID_POSIX_SUBTABLE(sd) {
545    {0x59,   "sd"},
546    {0x0459, "sd_IN"},
547    {0x0859, "sd_PK"}
548};
549
550ILCID_POSIX_SUBTABLE(se) {
551    {0x3b,   "se"},
552    {0x0c3b, "se_FI"},
553    {0x043b, "se_NO"},
554    {0x083b, "se_SE"},
555    {0x783b, "sma"},
556    {0x183b, "sma_NO"},
557    {0x1c3b, "sma_SE"},
558    {0x7c3b, "smj"},
559    {0x703b, "smn"},
560    {0x743b, "sms"},
561    {0x103b, "smj_NO"},
562    {0x143b, "smj_SE"},
563    {0x243b, "smn_FI"},
564    {0x203b, "sms_FI"},
565};
566
567ILCID_POSIX_ELEMENT_ARRAY(0x045b, si, si_LK)
568ILCID_POSIX_ELEMENT_ARRAY(0x041b, sk, sk_SK)
569ILCID_POSIX_ELEMENT_ARRAY(0x0424, sl, sl_SI)
570
571ILCID_POSIX_SUBTABLE(so) { /* TODO: Verify the country */
572    {0x77,   "so"},
573    {0x0477, "so_ET"},
574    {0x0477, "so_SO"}
575};
576
577ILCID_POSIX_ELEMENT_ARRAY(0x041c, sq, sq_AL)
578ILCID_POSIX_ELEMENT_ARRAY(0x0430, st, st_ZA)
579
580ILCID_POSIX_SUBTABLE(sv) {
581    {0x1d,   "sv"},
582    {0x081d, "sv_FI"},
583    {0x041d, "sv_SE"}
584};
585
586ILCID_POSIX_ELEMENT_ARRAY(0x0441, sw, sw_KE)
587ILCID_POSIX_ELEMENT_ARRAY(0x045A, syr, syr_SY)
588ILCID_POSIX_ELEMENT_ARRAY(0x0449, ta, ta_IN)
589ILCID_POSIX_ELEMENT_ARRAY(0x044a, te, te_IN)
590
591/* Cyrillic based by default */
592ILCID_POSIX_SUBTABLE(tg) {
593    {0x28,   "tg"},
594    {0x7c28, "tg_Cyrl"},
595    {0x0428, "tg_Cyrl_TJ"}
596};
597
598ILCID_POSIX_ELEMENT_ARRAY(0x041e, th, th_TH)
599
600ILCID_POSIX_SUBTABLE(ti) {
601    {0x73,   "ti"},
602    {0x0473, "ti_ER"},
603    {0x0873, "ti_ER"},
604    {0x0873, "ti_ET"},
605    {0x0473, "ti_ET"}
606};
607
608ILCID_POSIX_ELEMENT_ARRAY(0x0442, tk, tk_TM)
609
610ILCID_POSIX_SUBTABLE(tn) {
611    {0x32,   "tn"},
612    {0x0432, "tn_BW"},
613    {0x0432, "tn_ZA"}
614};
615
616ILCID_POSIX_ELEMENT_ARRAY(0x041f, tr, tr_TR)
617ILCID_POSIX_ELEMENT_ARRAY(0x0431, ts, ts_ZA)
618ILCID_POSIX_ELEMENT_ARRAY(0x0444, tt, tt_RU)
619
620ILCID_POSIX_SUBTABLE(tzm) {
621    {0x5f,   "tzm"},
622    {0x7c5f, "tzm_Latn"},
623    {0x085f, "tzm_Latn_DZ"},
624    {0x045f, "tmz"}
625};
626
627ILCID_POSIX_SUBTABLE(ug) {
628    {0x80,   "ug"},
629    {0x0480, "ug_CN"},
630    {0x0480, "ug_Arab_CN"}
631};
632
633ILCID_POSIX_ELEMENT_ARRAY(0x0422, uk, uk_UA)
634
635ILCID_POSIX_SUBTABLE(ur) {
636    {0x20,   "ur"},
637    {0x0820, "ur_IN"},
638    {0x0420, "ur_PK"}
639};
640
641ILCID_POSIX_SUBTABLE(uz) {
642    {0x43,   "uz"},
643    {0x0843, "uz_Cyrl_UZ"},  /* Cyrillic based */
644    {0x7843, "uz_Cyrl"},  /* Cyrillic based */
645    {0x0843, "uz_UZ"},  /* Cyrillic based */
646    {0x0443, "uz_Latn_UZ"}, /* Latin based */
647    {0x7c43, "uz_Latn"} /* Latin based */
648};
649
650ILCID_POSIX_SUBTABLE(ve) { /* TODO: Verify the country */
651    {0x33,   "ve"},
652    {0x0433, "ve_ZA"},
653    {0x0433, "ven_ZA"}
654};
655
656ILCID_POSIX_ELEMENT_ARRAY(0x042a, vi, vi_VN)
657
658ILCID_POSIX_SUBTABLE(wen) {
659    {0x2E,   "wen"},
660    {0x042E, "wen_DE"},
661    {0x042E, "hsb_DE"},
662    {0x082E, "dsb_DE"},
663    {0x7C2E, "dsb"},
664    {0x2E,   "hsb"}
665};
666
667ILCID_POSIX_ELEMENT_ARRAY(0x0488, wo, wo_SN)
668ILCID_POSIX_ELEMENT_ARRAY(0x0434, xh, xh_ZA)
669ILCID_POSIX_ELEMENT_ARRAY(0x043d, yi, yi)
670ILCID_POSIX_ELEMENT_ARRAY(0x046a, yo, yo_NG)
671
672ILCID_POSIX_SUBTABLE(zh) {
673    {0x0004, "zh_Hans"},
674    {0x7804, "zh"},
675    {0x0804, "zh_CN"},
676    {0x0804, "zh_Hans_CN"},
677    {0x0c04, "zh_Hant_HK"},
678    {0x0c04, "zh_HK"},
679    {0x1404, "zh_Hant_MO"},
680    {0x1404, "zh_MO"},
681    {0x1004, "zh_Hans_SG"},
682    {0x1004, "zh_SG"},
683    {0x0404, "zh_Hant_TW"},
684    {0x7c04, "zh_Hant"},
685    {0x0404, "zh_TW"},
686    {0x30404,"zh_Hant_TW"},     /* Bopomofo order */
687    {0x30404,"zh_TW"},          /* Bopomofo order */
688    {0x20004,"zh@collation=stroke"},
689    {0x20404,"zh_Hant@collation=stroke"},
690    {0x20404,"zh_Hant_TW@collation=stroke"},
691    {0x20404,"zh_TW@collation=stroke"},
692    {0x20804,"zh_Hans@collation=stroke"},
693    {0x20804,"zh_Hans_CN@collation=stroke"},
694    {0x20804,"zh_CN@collation=stroke"}
695};
696
697ILCID_POSIX_ELEMENT_ARRAY(0x0435, zu, zu_ZA)
698
699/* This must be static and grouped by LCID. */
700static const ILcidPosixMap gPosixIDmap[] = {
701    ILCID_POSIX_MAP(af),    /*  af  Afrikaans                 0x36 */
702    ILCID_POSIX_MAP(am),    /*  am  Amharic                   0x5e */
703    ILCID_POSIX_MAP(ar),    /*  ar  Arabic                    0x01 */
704    ILCID_POSIX_MAP(arn),   /*  arn Araucanian/Mapudungun     0x7a */
705    ILCID_POSIX_MAP(as),    /*  as  Assamese                  0x4d */
706    ILCID_POSIX_MAP(az),    /*  az  Azerbaijani               0x2c */
707    ILCID_POSIX_MAP(ba),    /*  ba  Bashkir                   0x6d */
708    ILCID_POSIX_MAP(be),    /*  be  Belarusian                0x23 */
709/*    ILCID_POSIX_MAP(ber),     ber Berber/Tamazight          0x5f */
710    ILCID_POSIX_MAP(bg),    /*  bg  Bulgarian                 0x02 */
711    ILCID_POSIX_MAP(bin),   /*  bin Edo                       0x66 */
712    ILCID_POSIX_MAP(bn),    /*  bn  Bengali; Bangla           0x45 */
713    ILCID_POSIX_MAP(bo),    /*  bo  Tibetan                   0x51 */
714    ILCID_POSIX_MAP(br),    /*  br  Breton                    0x7e */
715    ILCID_POSIX_MAP(ca),    /*  ca  Catalan                   0x03 */
716    ILCID_POSIX_MAP(chr),   /*  chr Cherokee                  0x5c */
717    ILCID_POSIX_MAP(co),    /*  co  Corsican                  0x83 */
718    ILCID_POSIX_MAP(cs),    /*  cs  Czech                     0x05 */
719    ILCID_POSIX_MAP(cy),    /*  cy  Welsh                     0x52 */
720    ILCID_POSIX_MAP(da),    /*  da  Danish                    0x06 */
721    ILCID_POSIX_MAP(de),    /*  de  German                    0x07 */
722    ILCID_POSIX_MAP(dv),    /*  dv  Divehi                    0x65 */
723    ILCID_POSIX_MAP(el),    /*  el  Greek                     0x08 */
724    ILCID_POSIX_MAP(en),    /*  en  English                   0x09 */
725    ILCID_POSIX_MAP(en_US_POSIX), /*    invariant             0x7f */
726    ILCID_POSIX_MAP(es),    /*  es  Spanish                   0x0a */
727    ILCID_POSIX_MAP(et),    /*  et  Estonian                  0x25 */
728    ILCID_POSIX_MAP(eu),    /*  eu  Basque                    0x2d */
729    ILCID_POSIX_MAP(fa),    /*  fa  Persian/Farsi             0x29 */
730    ILCID_POSIX_MAP(fa_AF), /*  fa  Persian/Dari              0x8c */
731    ILCID_POSIX_MAP(fi),    /*  fi  Finnish                   0x0b */
732    ILCID_POSIX_MAP(fil),   /*  fil Filipino                  0x64 */
733    ILCID_POSIX_MAP(fo),    /*  fo  Faroese                   0x38 */
734    ILCID_POSIX_MAP(fr),    /*  fr  French                    0x0c */
735    ILCID_POSIX_MAP(fuv),   /*  fuv Fulfulde - Nigeria        0x67 */
736    ILCID_POSIX_MAP(fy),    /*  fy  Frisian                   0x62 */
737    ILCID_POSIX_MAP(ga),    /*  *   Gaelic (Ireland,Scotland) 0x3c */
738    ILCID_POSIX_MAP(gd),    /*  gd  Gaelic (United Kingdom)   0x91 */
739    ILCID_POSIX_MAP(gl),    /*  gl  Galician                  0x56 */
740    ILCID_POSIX_MAP(gn),    /*  gn  Guarani                   0x74 */
741    ILCID_POSIX_MAP(gsw),   /*  gsw Alemanic/Alsatian/Swiss German 0x84 */
742    ILCID_POSIX_MAP(gu),    /*  gu  Gujarati                  0x47 */
743    ILCID_POSIX_MAP(ha),    /*  ha  Hausa                     0x68 */
744    ILCID_POSIX_MAP(haw),   /*  haw Hawaiian                  0x75 */
745    ILCID_POSIX_MAP(he),    /*  he  Hebrew (formerly iw)      0x0d */
746    ILCID_POSIX_MAP(hi),    /*  hi  Hindi                     0x39 */
747    ILCID_POSIX_MAP(hr),    /*  *   Croatian and others       0x1a */
748    ILCID_POSIX_MAP(hu),    /*  hu  Hungarian                 0x0e */
749    ILCID_POSIX_MAP(hy),    /*  hy  Armenian                  0x2b */
750    ILCID_POSIX_MAP(ibb),   /*  ibb Ibibio - Nigeria          0x69 */
751    ILCID_POSIX_MAP(id),    /*  id  Indonesian (formerly in)  0x21 */
752    ILCID_POSIX_MAP(ig),    /*  ig  Igbo                      0x70 */
753    ILCID_POSIX_MAP(ii),    /*  ii  Sichuan Yi                0x78 */
754    ILCID_POSIX_MAP(is),    /*  is  Icelandic                 0x0f */
755    ILCID_POSIX_MAP(it),    /*  it  Italian                   0x10 */
756    ILCID_POSIX_MAP(iu),    /*  iu  Inuktitut                 0x5d */
757    ILCID_POSIX_MAP(iw),    /*  iw  Hebrew                    0x0d */
758    ILCID_POSIX_MAP(ja),    /*  ja  Japanese                  0x11 */
759    ILCID_POSIX_MAP(ka),    /*  ka  Georgian                  0x37 */
760    ILCID_POSIX_MAP(kk),    /*  kk  Kazakh                    0x3f */
761    ILCID_POSIX_MAP(kl),    /*  kl  Kalaallisut               0x6f */
762    ILCID_POSIX_MAP(km),    /*  km  Khmer                     0x53 */
763    ILCID_POSIX_MAP(kn),    /*  kn  Kannada                   0x4b */
764    ILCID_POSIX_MAP(ko),    /*  ko  Korean                    0x12 */
765    ILCID_POSIX_MAP(kok),   /*  kok Konkani                   0x57 */
766    ILCID_POSIX_MAP(kr),    /*  kr  Kanuri                    0x71 */
767    ILCID_POSIX_MAP(ks),    /*  ks  Kashmiri                  0x60 */
768    ILCID_POSIX_MAP(ky),    /*  ky  Kyrgyz                    0x40 */
769    ILCID_POSIX_MAP(lb),    /*  lb  Luxembourgish             0x6e */
770    ILCID_POSIX_MAP(la),    /*  la  Latin                     0x76 */
771    ILCID_POSIX_MAP(lo),    /*  lo  Lao                       0x54 */
772    ILCID_POSIX_MAP(lt),    /*  lt  Lithuanian                0x27 */
773    ILCID_POSIX_MAP(lv),    /*  lv  Latvian, Lettish          0x26 */
774    ILCID_POSIX_MAP(mi),    /*  mi  Maori                     0x81 */
775    ILCID_POSIX_MAP(mk),    /*  mk  Macedonian                0x2f */
776    ILCID_POSIX_MAP(ml),    /*  ml  Malayalam                 0x4c */
777    ILCID_POSIX_MAP(mn),    /*  mn  Mongolian                 0x50 */
778    ILCID_POSIX_MAP(mni),   /*  mni Manipuri                  0x58 */
779    ILCID_POSIX_MAP(moh),   /*  moh Mohawk                    0x7c */
780    ILCID_POSIX_MAP(mr),    /*  mr  Marathi                   0x4e */
781    ILCID_POSIX_MAP(ms),    /*  ms  Malay                     0x3e */
782    ILCID_POSIX_MAP(mt),    /*  mt  Maltese                   0x3a */
783    ILCID_POSIX_MAP(my),    /*  my  Burmese                   0x55 */
784/*    ILCID_POSIX_MAP(nb),    //  no  Norwegian                 0x14 */
785    ILCID_POSIX_MAP(ne),    /*  ne  Nepali                    0x61 */
786    ILCID_POSIX_MAP(nl),    /*  nl  Dutch                     0x13 */
787/*    ILCID_POSIX_MAP(nn),    //  no  Norwegian                 0x14 */
788    ILCID_POSIX_MAP(no),    /*  *   Norwegian                 0x14 */
789    ILCID_POSIX_MAP(nso),   /*  nso Sotho, Northern (Sepedi dialect) 0x6c */
790    ILCID_POSIX_MAP(oc),    /*  oc  Occitan                   0x82 */
791    ILCID_POSIX_MAP(om),    /*  om  Oromo                     0x72 */
792    ILCID_POSIX_MAP(or_IN), /*  or  Oriya                     0x48 */
793    ILCID_POSIX_MAP(pa),    /*  pa  Punjabi                   0x46 */
794    ILCID_POSIX_MAP(pap),   /*  pap Papiamentu                0x79 */
795    ILCID_POSIX_MAP(pl),    /*  pl  Polish                    0x15 */
796    ILCID_POSIX_MAP(ps),    /*  ps  Pashto                    0x63 */
797    ILCID_POSIX_MAP(pt),    /*  pt  Portuguese                0x16 */
798    ILCID_POSIX_MAP(qu),    /*  qu  Quechua                   0x6B */
799    ILCID_POSIX_MAP(qut),   /*  qut K'iche                    0x86 */
800    ILCID_POSIX_MAP(rm),    /*  rm  Raeto-Romance/Romansh     0x17 */
801    ILCID_POSIX_MAP(ro),    /*  ro  Romanian                  0x18 */
802    ILCID_POSIX_MAP(root),  /*  root                          0x00 */
803    ILCID_POSIX_MAP(ru),    /*  ru  Russian                   0x19 */
804    ILCID_POSIX_MAP(rw),    /*  rw  Kinyarwanda               0x87 */
805    ILCID_POSIX_MAP(sa),    /*  sa  Sanskrit                  0x4f */
806    ILCID_POSIX_MAP(sah),   /*  sah Yakut                     0x85 */
807    ILCID_POSIX_MAP(sd),    /*  sd  Sindhi                    0x59 */
808    ILCID_POSIX_MAP(se),    /*  se  Sami                      0x3b */
809/*    ILCID_POSIX_MAP(sh),    //  sh  Serbo-Croatian            0x1a */
810    ILCID_POSIX_MAP(si),    /*  si  Sinhalese                 0x5b */
811    ILCID_POSIX_MAP(sk),    /*  sk  Slovak                    0x1b */
812    ILCID_POSIX_MAP(sl),    /*  sl  Slovenian                 0x24 */
813    ILCID_POSIX_MAP(so),    /*  so  Somali                    0x77 */
814    ILCID_POSIX_MAP(sq),    /*  sq  Albanian                  0x1c */
815/*    ILCID_POSIX_MAP(sr),    //  sr  Serbian                   0x1a */
816    ILCID_POSIX_MAP(st),    /*  st  Sutu                      0x30 */
817    ILCID_POSIX_MAP(sv),    /*  sv  Swedish                   0x1d */
818    ILCID_POSIX_MAP(sw),    /*  sw  Swahili                   0x41 */
819    ILCID_POSIX_MAP(syr),   /*  syr Syriac                    0x5A */
820    ILCID_POSIX_MAP(ta),    /*  ta  Tamil                     0x49 */
821    ILCID_POSIX_MAP(te),    /*  te  Telugu                    0x4a */
822    ILCID_POSIX_MAP(tg),    /*  tg  Tajik                     0x28 */
823    ILCID_POSIX_MAP(th),    /*  th  Thai                      0x1e */
824    ILCID_POSIX_MAP(ti),    /*  ti  Tigrigna                  0x73 */
825    ILCID_POSIX_MAP(tk),    /*  tk  Turkmen                   0x42 */
826    ILCID_POSIX_MAP(tn),    /*  tn  Tswana                    0x32 */
827    ILCID_POSIX_MAP(tr),    /*  tr  Turkish                   0x1f */
828    ILCID_POSIX_MAP(ts),    /*  ts  Tsonga                    0x31 */
829    ILCID_POSIX_MAP(tt),    /*  tt  Tatar                     0x44 */
830    ILCID_POSIX_MAP(tzm),   /*  tzm Tamazight                 0x5f */
831    ILCID_POSIX_MAP(ug),    /*  ug  Uighur                    0x80 */
832    ILCID_POSIX_MAP(uk),    /*  uk  Ukrainian                 0x22 */
833    ILCID_POSIX_MAP(ur),    /*  ur  Urdu                      0x20 */
834    ILCID_POSIX_MAP(uz),    /*  uz  Uzbek                     0x43 */
835    ILCID_POSIX_MAP(ve),    /*  ve  Venda                     0x33 */
836    ILCID_POSIX_MAP(vi),    /*  vi  Vietnamese                0x2a */
837    ILCID_POSIX_MAP(wen),   /*  wen Sorbian                   0x2e */
838    ILCID_POSIX_MAP(wo),    /*  wo  Wolof                     0x88 */
839    ILCID_POSIX_MAP(xh),    /*  xh  Xhosa                     0x34 */
840    ILCID_POSIX_MAP(yi),    /*  yi  Yiddish                   0x3d */
841    ILCID_POSIX_MAP(yo),    /*  yo  Yoruba                    0x6a */
842    ILCID_POSIX_MAP(zh),    /*  zh  Chinese                   0x04 */
843    ILCID_POSIX_MAP(zu),    /*  zu  Zulu                      0x35 */
844};
845
846static const uint32_t gLocaleCount = sizeof(gPosixIDmap)/sizeof(ILcidPosixMap);
847
848/**
849 * Do not call this function. It is called by hostID.
850 * The function is not private because this struct must stay as a C struct,
851 * and this is an internal class.
852 */
853static int32_t
854idCmp(const char* id1, const char* id2)
855{
856    int32_t diffIdx = 0;
857    while (*id1 == *id2 && *id1 != 0) {
858        diffIdx++;
859        id1++;
860        id2++;
861    }
862    return diffIdx;
863}
864
865/**
866 * Searches for a Windows LCID
867 *
868 * @param posixid the Posix style locale id.
869 * @param status gets set to U_ILLEGAL_ARGUMENT_ERROR when the Posix ID has
870 *               no equivalent Windows LCID.
871 * @return the LCID
872 */
873static uint32_t
874getHostID(const ILcidPosixMap *this_0, const char* posixID, UErrorCode* status)
875{
876    int32_t bestIdx = 0;
877    int32_t bestIdxDiff = 0;
878    int32_t posixIDlen = (int32_t)uprv_strlen(posixID);
879    uint32_t idx;
880
881    for (idx = 0; idx < this_0->numRegions; idx++ ) {
882        int32_t sameChars = idCmp(posixID, this_0->regionMaps[idx].posixID);
883        if (sameChars > bestIdxDiff && this_0->regionMaps[idx].posixID[sameChars] == 0) {
884            if (posixIDlen == sameChars) {
885                /* Exact match */
886                return this_0->regionMaps[idx].hostID;
887            }
888            bestIdxDiff = sameChars;
889            bestIdx = idx;
890        }
891    }
892    /* We asked for something unusual, like en_ZZ, and we try to return the number for the same language. */
893    /* We also have to make sure that sid and si and similar string subsets don't match. */
894    if ((posixID[bestIdxDiff] == '_' || posixID[bestIdxDiff] == '@')
895        && this_0->regionMaps[bestIdx].posixID[bestIdxDiff] == 0)
896    {
897        *status = U_USING_FALLBACK_WARNING;
898        return this_0->regionMaps[bestIdx].hostID;
899    }
900
901    /*no match found */
902    *status = U_ILLEGAL_ARGUMENT_ERROR;
903    return this_0->regionMaps->hostID;
904}
905
906static const char*
907getPosixID(const ILcidPosixMap *this_0, uint32_t hostID)
908{
909    uint32_t i;
910    for (i = 0; i <= this_0->numRegions; i++)
911    {
912        if (this_0->regionMaps[i].hostID == hostID)
913        {
914            return this_0->regionMaps[i].posixID;
915        }
916    }
917
918    /* If you get here, then no matching region was found,
919       so return the language id with the wild card region. */
920    return this_0->regionMaps[0].posixID;
921}
922
923/*
924//////////////////////////////////////
925//
926// LCID --> POSIX
927//
928/////////////////////////////////////
929*/
930#ifdef USE_WINDOWS_LOCALE_API
931/*
932 * Change the tag separator from '-' to '_'
933 */
934#define FIX_LOCALE_ID_TAG_SEPARATOR(buffer, len, i) \
935    for(i = 0; i < len; i++) \
936        if (buffer[i] == '-') buffer[i] = '_';
937
938/*
939 * Various language tags needs to be changed:
940 * quz -> qu
941 * prs -> fa
942 */
943#define FIX_LANGUAGE_ID_TAG(buffer, len) \
944    if (len >= 3) { \
945        if (buffer[0] == 'q' && buffer[1] == 'u' && buffer[2] == 'z') {\
946            buffer[2] = 0; \
947            uprv_strcat(buffer, buffer+3); \
948        } else if (buffer[0] == 'p' && buffer[1] == 'r' && buffer[2] == 's') {\
949            buffer[0] = 'f'; buffer[1] = 'a'; buffer[2] = 0; \
950            uprv_strcat(buffer, buffer+3); \
951        } \
952    }
953
954static char gPosixFromLCID[ULOC_FULLNAME_CAPACITY];
955#endif
956U_CAPI const char *
957uprv_convertToPosix(uint32_t hostid, UErrorCode* status)
958{
959    uint16_t langID;
960    uint32_t localeIndex;
961#ifdef USE_WINDOWS_LOCALE_API
962    int32_t ret = 0;
963
964    uprv_memset(gPosixFromLCID, 0, sizeof(gPosixFromLCID));
965
966    ret = GetLocaleInfoA(hostid, LOCALE_SNAME, (LPSTR)gPosixFromLCID, sizeof(gPosixFromLCID));
967    if (ret > 1) {
968        FIX_LOCALE_ID_TAG_SEPARATOR(gPosixFromLCID, (uint32_t)ret, localeIndex)
969        FIX_LANGUAGE_ID_TAG(gPosixFromLCID, ret)
970
971        return gPosixFromLCID;
972    }
973#endif
974    langID = LANGUAGE_LCID(hostid);
975
976    for (localeIndex = 0; localeIndex < gLocaleCount; localeIndex++)
977    {
978        if (langID == gPosixIDmap[localeIndex].regionMaps->hostID)
979        {
980            return getPosixID(&gPosixIDmap[localeIndex], hostid);
981        }
982    }
983
984    /* no match found */
985    *status = U_ILLEGAL_ARGUMENT_ERROR;
986    return NULL;
987}
988
989/*
990//////////////////////////////////////
991//
992// POSIX --> LCID
993// This should only be called from uloc_getLCID.
994// The locale ID must be in canonical form.
995// langID is separate so that this file doesn't depend on the uloc_* API.
996//
997/////////////////////////////////////
998*/
999
1000U_CAPI uint32_t
1001uprv_convertToLCID(const char *langID, const char* posixID, UErrorCode* status)
1002{
1003
1004    uint32_t   low    = 0;
1005    uint32_t   high   = gLocaleCount;
1006    uint32_t   mid;
1007    uint32_t   oldmid = 0;
1008    int32_t    compVal;
1009
1010    uint32_t   value         = 0;
1011    uint32_t   fallbackValue = (uint32_t)-1;
1012    UErrorCode myStatus;
1013    uint32_t   idx;
1014
1015    /* Check for incomplete id. */
1016    if (!langID || !posixID || uprv_strlen(langID) < 2 || uprv_strlen(posixID) < 2) {
1017        return 0;
1018    }
1019
1020    /*Binary search for the map entry for normal cases */
1021
1022    while (high > low)  /*binary search*/{
1023
1024        mid = (high+low) >> 1; /*Finds median*/
1025
1026        if (mid == oldmid)
1027            break;
1028
1029        compVal = uprv_strcmp(langID, gPosixIDmap[mid].regionMaps->posixID);
1030        if (compVal < 0){
1031            high = mid;
1032        }
1033        else if (compVal > 0){
1034            low = mid;
1035        }
1036        else /*we found it*/{
1037            return getHostID(&gPosixIDmap[mid], posixID, status);
1038        }
1039        oldmid = mid;
1040    }
1041
1042    /*
1043     * Sometimes we can't do a binary search on posixID because some LCIDs
1044     * go to different locales.  We hit one of those special cases.
1045     */
1046    for (idx = 0; idx < gLocaleCount; idx++ ) {
1047        myStatus = U_ZERO_ERROR;
1048        value = getHostID(&gPosixIDmap[idx], posixID, &myStatus);
1049        if (myStatus == U_ZERO_ERROR) {
1050            return value;
1051        }
1052        else if (myStatus == U_USING_FALLBACK_WARNING) {
1053            fallbackValue = value;
1054        }
1055    }
1056
1057    if (fallbackValue != (uint32_t)-1) {
1058        *status = U_USING_FALLBACK_WARNING;
1059        return fallbackValue;
1060    }
1061
1062    /* no match found */
1063    *status = U_ILLEGAL_ARGUMENT_ERROR;
1064    return 0;   /* return international (root) */
1065}
1066
1067