9 * software and its documentation for any purpose, provided that the 10 * above copyright notice and the following two paragraphs appear in 11 * all copies of this software. 12 * 13 * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR 14 * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES 15 * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN 16 * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH 17 * DAMAGE. 18 * 19 * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, 20 * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND 21 * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS 22 * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO 23 * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. 24 * 25 * Red Hat Author(s): Behdad Esfahbod 26 * Google Author(s): Behdad Esfahbod, Roozbeh Pournader 27 */ 28 29 #include "hb-private.hh" 30 31 32 /* hb_script_t */ 33 34 static hb_tag_t 35 hb_ot_old_tag_from_script (hb_script_t script) 36 { 37 /* This seems to be accurate as of end of 2012. */ 38 39 switch ((hb_tag_t) script) { 40 case HB_SCRIPT_INVALID: return HB_OT_TAG_DEFAULT_SCRIPT; 41 42 /* KATAKANA and HIRAGANA both map to 'kana' */ 43 case HB_SCRIPT_HIRAGANA: return HB_TAG('k','a','n','a'); 44 45 /* Spaces at the end are preserved, unlike ISO 15924 */ 46 case HB_SCRIPT_LAO: return HB_TAG('l','a','o',' '); 47 case HB_SCRIPT_YI: return HB_TAG('y','i',' ',' '); 48 /* Unicode-5.0 additions */ 49 case HB_SCRIPT_NKO: return HB_TAG('n','k','o',' '); 50 /* Unicode-5.1 additions */ 51 case HB_SCRIPT_VAI: return HB_TAG('v','a','i',' '); 52 /* Unicode-5.2 additions */ 53 /* Unicode-6.0 additions */ 54 } 55 56 /* Else, just change first char to lowercase and return */ 57 return ((hb_tag_t) script) | 0x20000000u; 58 } 59 60 static hb_script_t 61 hb_ot_old_tag_to_script (hb_tag_t tag) 62 { 63 if (unlikely (tag == HB_OT_TAG_DEFAULT_SCRIPT)) 64 return HB_SCRIPT_INVALID; 65 66 /* This side of the conversion is fully algorithmic. */ 67 68 /* Any spaces at the end of the tag are replaced by repeating the last 69 * letter. Eg 'nko ' -> 'Nkoo' */ 70 if (unlikely ((tag & 0x0000FF00u) == 0x00002000u)) 71 tag |= (tag >> 8) & 0x0000FF00u; /* Copy second letter to third */ 72 if (unlikely ((tag & 0x000000FFu) == 0x00000020u)) 73 tag |= (tag >> 8) & 0x000000FFu; /* Copy third letter to fourth */ 97 98 static hb_script_t 99 hb_ot_new_tag_to_script (hb_tag_t tag) 100 { 101 switch (tag) { 102 case HB_TAG('b','n','g','2'): return HB_SCRIPT_BENGALI; 103 case HB_TAG('d','e','v','2'): return HB_SCRIPT_DEVANAGARI; 104 case HB_TAG('g','j','r','2'): return HB_SCRIPT_GUJARATI; 105 case HB_TAG('g','u','r','2'): return HB_SCRIPT_GURMUKHI; 106 case HB_TAG('k','n','d','2'): return HB_SCRIPT_KANNADA; 107 case HB_TAG('m','l','m','2'): return HB_SCRIPT_MALAYALAM; 108 case HB_TAG('o','r','y','2'): return HB_SCRIPT_ORIYA; 109 case HB_TAG('t','m','l','2'): return HB_SCRIPT_TAMIL; 110 case HB_TAG('t','e','l','2'): return HB_SCRIPT_TELUGU; 111 case HB_TAG('m','y','m','2'): return HB_SCRIPT_MYANMAR; 112 } 113 114 return HB_SCRIPT_UNKNOWN; 115 } 116 117 /* 118 * Complete list at: 119 * https://docs.microsoft.com/en-us/typography/opentype/spec/scripttags 120 * 121 * Most of the script tags are the same as the ISO 15924 tag but lowercased. 122 * So we just do that, and handle the exceptional cases in a switch. 123 */ 124 125 void 126 hb_ot_tags_from_script (hb_script_t script, 127 hb_tag_t *script_tag_1, 128 hb_tag_t *script_tag_2) 129 { 130 hb_tag_t new_tag; 131 132 *script_tag_2 = HB_OT_TAG_DEFAULT_SCRIPT; 133 *script_tag_1 = hb_ot_old_tag_from_script (script); 134 135 new_tag = hb_ot_new_tag_from_script (script); 136 if (unlikely (new_tag != HB_OT_TAG_DEFAULT_SCRIPT)) { 137 *script_tag_2 = *script_tag_1; 138 *script_tag_1 = new_tag; 139 } 140 } 141 142 hb_script_t 143 hb_ot_tag_to_script (hb_tag_t tag) 144 { 145 if (unlikely ((tag & 0x000000FFu) == '2')) 146 return hb_ot_new_tag_to_script (tag); 147 148 return hb_ot_old_tag_to_script (tag); 149 } 150 151 152 /* hb_language_t */ 153 154 typedef struct { 155 char language[4]; 156 hb_tag_t tag; 157 } LangTag; 158 159 /* 160 * Complete list at: 161 * https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags 162 * 163 * Generated by intersecting the OpenType language tag list from 164 * Draft OpenType 1.5 spec, with with the ISO 639-3 codes from 165 * 2008-08-04, matching on name, and finally adjusted manually. 166 * 167 * Updated on 2012-12-07 with more research into remaining codes. 168 * 169 * Updated on 2013-11-23 based on usage in SIL and Microsoft fonts, 170 * the new proposal from Microsoft, and latest ISO 639-3 names. 171 * 172 * Some items still missing. Those are commented out at the end. 173 * Keep sorted for bsearch. 174 * 175 * Updated as of 2015-05-06: OT1.7 on MS website has some newer 176 * items that we don't have here, eg. Zazaki. This is the new 177 * items in OpenType 1.7 (red items), most of which we have: 178 * https://docs.microsoft.com/en-us/typography/opentype/spec/languagetags 179 */ 180 181 static const LangTag ot_languages[] = { 182 {"aa", HB_TAG('A','F','R',' ')}, /* Afar */ 183 {"ab", HB_TAG('A','B','K',' ')}, /* Abkhazian */ 184 {"abq", HB_TAG('A','B','A',' ')}, /* Abaza */ 185 {"acf", HB_TAG('F','A','N',' ')}, /* French Antillean */ 186 {"ach", HB_TAG('A','C','H',' ')}, /* Acoli */ 187 {"acr", HB_TAG('A','C','R',' ')}, /* Achi */ 188 {"ada", HB_TAG('D','N','G',' ')}, /* Dangme */ 189 {"ady", HB_TAG('A','D','Y',' ')}, /* Adyghe */ 190 {"af", HB_TAG('A','F','K',' ')}, /* Afrikaans */ 191 {"ahg", HB_TAG('A','G','W',' ')}, /* Agaw */ 192 {"aii", HB_TAG('S','W','A',' ')}, /* Swadaya Aramaic */ 193 {"aio", HB_TAG('A','I','O',' ')}, /* Aiton */ 194 {"aiw", HB_TAG('A','R','I',' ')}, /* Aari */ 195 {"ak", HB_TAG('T','W','I',' ')}, /* Akan [macrolanguage] */ 196 {"aka", HB_TAG('A','K','A',' ')}, /* Akan */ 197 {"alt", HB_TAG('A','L','T',' ')}, /* [Southern] Altai */ 198 {"am", HB_TAG('A','M','H',' ')}, /* Amharic */ 199 {"amf", HB_TAG('H','B','N',' ')}, /* Hammer-Banna */ 200 {"amw", HB_TAG('S','Y','R',' ')}, /* Western Neo-Aramaic */ 201 {"an", HB_TAG('A','R','G',' ')}, /* Aragonese */ 202 {"ang", HB_TAG('A','N','G',' ')}, /* Old English (ca. 450-1100) */ 203 {"ar", HB_TAG('A','R','A',' ')}, /* Arabic [macrolanguage] */ 204 {"arb", HB_TAG('A','R','A',' ')}, /* Standard Arabic */ 205 {"arn", HB_TAG('M','A','P',' ')}, /* Mapudungun */ 206 {"ary", HB_TAG('M','O','R',' ')}, /* Moroccan Arabic */ 207 {"as", HB_TAG('A','S','M',' ')}, /* Assamese */ 208 {"ast", HB_TAG('A','S','T',' ')}, /* Asturian/Asturleonese/Bable/Leonese */ 209 {"ath", HB_TAG('A','T','H',' ')}, /* Athapaskan [family] */ 210 {"atj", HB_TAG('R','C','R',' ')}, /* R-Cree */ 211 {"atv", HB_TAG('A','L','T',' ')}, /* [Northern] Altai */ 212 {"av", HB_TAG('A','V','R',' ')}, /* Avaric */ 213 {"awa", HB_TAG('A','W','A',' ')}, /* Awadhi */ 214 {"ay", HB_TAG('A','Y','M',' ')}, /* Aymara [macrolanguage] */ 215 {"az", HB_TAG('A','Z','E',' ')}, /* Azerbaijani [macrolanguage] */ 216 {"azb", HB_TAG('A','Z','B',' ')}, /* South Azerbaijani */ 217 {"azj", HB_TAG('A','Z','E',' ')}, /* North Azerbaijani */ 218 {"ba", HB_TAG('B','S','H',' ')}, /* Bashkir */ 219 {"bad", HB_TAG('B','A','D','0')}, /* Banda */ 220 {"bai", HB_TAG('B','M','L',' ')}, /* Bamileke [family] */ 221 {"bal", HB_TAG('B','L','I',' ')}, /* Baluchi [macrolangauge] */ 222 {"ban", HB_TAG('B','A','N',' ')}, /* Balinese */ 223 {"bar", HB_TAG('B','A','R',' ')}, /* Bavarian */ 224 {"bbc", HB_TAG('B','B','C',' ')}, /* Batak Toba */ 225 {"bci", HB_TAG('B','A','U',' ')}, /* Baoulé */ 226 {"bcl", HB_TAG('B','I','K',' ')}, /* Central Bikol */ 227 {"bcq", HB_TAG('B','C','H',' ')}, /* Bench */ 228 {"bdy", HB_TAG('B','D','Y',' ')}, /* Bandjalang */ 229 {"be", HB_TAG('B','E','L',' ')}, /* Belarusian */ 230 {"bem", HB_TAG('B','E','M',' ')}, /* Bemba (Zambia) */ 231 {"ber", HB_TAG('B','E','R',' ')}, /* Berber [family] */ 232 {"bfq", HB_TAG('B','A','D',' ')}, /* Badaga */ 233 {"bft", HB_TAG('B','L','T',' ')}, /* Balti */ 234 {"bfu", HB_TAG('L','A','H',' ')}, /* Lahuli */ 235 {"bfy", HB_TAG('B','A','G',' ')}, /* Baghelkhandi */ 236 {"bg", HB_TAG('B','G','R',' ')}, /* Bulgarian */ 237 {"bgc", HB_TAG('B','G','C',' ')}, /* Haryanvi */ 238 {"bgq", HB_TAG('B','G','Q',' ')}, /* Bagri */ 239 {"bgr", HB_TAG('Q','I','N',' ')}, /* Bawm Chin */ 240 {"bhb", HB_TAG('B','H','I',' ')}, /* Bhili */ 241 {"bhk", HB_TAG('B','I','K',' ')}, /* Albay Bicolano (retired code) */ 242 {"bho", HB_TAG('B','H','O',' ')}, /* Bhojpuri */ 243 {"bi", HB_TAG('B','I','S',' ')}, /* Bislama */ 244 {"bik", HB_TAG('B','I','K',' ')}, /* Bikol [macrolanguage] */ 245 {"bin", HB_TAG('E','D','O',' ')}, /* Bini */ 246 {"bjj", HB_TAG('B','J','J',' ')}, /* Kanauji */ 247 {"bjt", HB_TAG('B','L','N',' ')}, /* Balanta-Ganja */ 248 {"bla", HB_TAG('B','K','F',' ')}, /* Blackfoot */ 249 {"ble", HB_TAG('B','L','N',' ')}, /* Balanta-Kentohe */ 250 {"blk", HB_TAG('B','L','K',' ')}, /* Pa'O/Pa'o Karen */ 251 {"bln", HB_TAG('B','I','K',' ')}, /* Southern Catanduanes Bikol */ 252 {"bm", HB_TAG('B','M','B',' ')}, /* Bambara */ 253 {"bn", HB_TAG('B','E','N',' ')}, /* Bengali */ 254 {"bo", HB_TAG('T','I','B',' ')}, /* Tibetan */ 255 {"bpy", HB_TAG('B','P','Y',' ')}, /* Bishnupriya */ 256 {"bqi", HB_TAG('L','R','C',' ')}, /* Bakhtiari */ 257 {"br", HB_TAG('B','R','E',' ')}, /* Breton */ 258 {"bra", HB_TAG('B','R','I',' ')}, /* Braj Bhasha */ 259 {"brh", HB_TAG('B','R','H',' ')}, /* Brahui */ 260 {"brx", HB_TAG('B','R','X',' ')}, /* Bodo (India) */ 261 {"bs", HB_TAG('B','O','S',' ')}, /* Bosnian */ 262 {"btb", HB_TAG('B','T','I',' ')}, /* Beti (Cameroon) */ 263 {"bto", HB_TAG('B','I','K',' ')}, /* Rinconada Bikol */ 264 {"bts", HB_TAG('B','T','S',' ')}, /* Batak Simalungun */ 265 {"bug", HB_TAG('B','U','G',' ')}, /* Buginese */ 266 {"bxr", HB_TAG('R','B','U',' ')}, /* Russian Buriat */ 267 {"byn", HB_TAG('B','I','L',' ')}, /* Bilen */ 268 {"ca", HB_TAG('C','A','T',' ')}, /* Catalan */ 269 {"cak", HB_TAG('C','A','K',' ')}, /* Kaqchikel */ 270 {"cbk", HB_TAG('C','B','K',' ')}, /* Chavacano */ 271 {"cbl", HB_TAG('Q','I','N',' ')}, /* Bualkhaw Chin */ 272 {"cco", HB_TAG('C','C','H','N')}, /* Chinantec */ 273 {"ce", HB_TAG('C','H','E',' ')}, /* Chechen */ 274 {"ceb", HB_TAG('C','E','B',' ')}, /* Cebuano */ 275 {"cfm", HB_TAG('H','A','L',' ')}, /* Halam/Falam Chin */ 276 {"cgg", HB_TAG('C','G','G',' ')}, /* Chiga */ 277 {"ch", HB_TAG('C','H','A',' ')}, /* Chamorro */ 278 {"chj", HB_TAG('C','C','H','N')}, /* Chinantec */ 279 {"chk", HB_TAG('C','H','K','0')}, /* Chuukese */ 280 {"cho", HB_TAG('C','H','O',' ')}, /* Choctaw */ 281 {"chp", HB_TAG('C','H','P',' ')}, /* Chipewyan */ 282 {"chq", HB_TAG('C','C','H','N')}, /* Chinantec */ 283 {"chr", HB_TAG('C','H','R',' ')}, /* Cherokee */ 284 {"chy", HB_TAG('C','H','Y',' ')}, /* Cheyenne */ 285 {"chz", HB_TAG('C','C','H','N')}, /* Chinantec */ 286 {"cja", HB_TAG('C','J','A',' ')}, /* Western Cham */ 287 {"cjm", HB_TAG('C','J','M',' ')}, /* Eastern Cham */ 288 {"cka", HB_TAG('Q','I','N',' ')}, /* Khumi Awa Chin */ 289 {"ckb", HB_TAG('K','U','R',' ')}, /* Central Kurdish (Sorani) */ 290 {"ckt", HB_TAG('C','H','K',' ')}, /* Chukchi */ 291 {"cld", HB_TAG('S','Y','R',' ')}, /* Chaldean Neo-Aramaic */ 292 {"cle", HB_TAG('C','C','H','N')}, /* Chinantec */ 293 {"cmr", HB_TAG('Q','I','N',' ')}, /* Mro-Khimi Chin */ 294 {"cnb", HB_TAG('Q','I','N',' ')}, /* Chinbon Chin */ 295 {"cnh", HB_TAG('Q','I','N',' ')}, /* Hakha Chin */ 296 {"cnk", HB_TAG('Q','I','N',' ')}, /* Khumi Chin */ 297 {"cnl", HB_TAG('C','C','H','N')}, /* Chinantec */ 298 {"cnt", HB_TAG('C','C','H','N')}, /* Chinantec */ 299 {"cnw", HB_TAG('Q','I','N',' ')}, /* Ngawn Chin */ 300 {"cop", HB_TAG('C','O','P',' ')}, /* Coptic */ 301 {"cpa", HB_TAG('C','C','H','N')}, /* Chinantec */ 302 {"cpp", HB_TAG('C','P','P',' ')}, /* Creoles */ 303 {"cr", HB_TAG('C','R','E',' ')}, /* Cree */ 304 {"cre", HB_TAG('Y','C','R',' ')}, /* Y-Cree */ 305 {"crh", HB_TAG('C','R','T',' ')}, /* Crimean Tatar */ 306 {"crj", HB_TAG('E','C','R',' ')}, /* [Southern] East Cree */ 307 {"crk", HB_TAG('W','C','R',' ')}, /* West-Cree */ 308 {"crl", HB_TAG('E','C','R',' ')}, /* [Northern] East Cree */ 309 {"crm", HB_TAG('M','C','R',' ')}, /* Moose Cree */ 310 {"crx", HB_TAG('C','R','R',' ')}, /* Carrier */ 311 {"cs", HB_TAG('C','S','Y',' ')}, /* Czech */ 312 {"csa", HB_TAG('C','C','H','N')}, /* Chinantec */ 313 {"csb", HB_TAG('C','S','B',' ')}, /* Kashubian */ 314 {"csh", HB_TAG('Q','I','N',' ')}, /* Asho Chin */ 315 {"cso", HB_TAG('C','C','H','N')}, /* Chinantec */ 316 {"csy", HB_TAG('Q','I','N',' ')}, /* Siyin Chin */ 317 {"ctd", HB_TAG('Q','I','N',' ')}, /* Tedim Chin */ 318 {"cte", HB_TAG('C','C','H','N')}, /* Chinantec */ 319 {"ctg", HB_TAG('C','T','G',' ')}, /* Chittagonian */ 320 {"ctl", HB_TAG('C','C','H','N')}, /* Chinantec */ 321 {"cts", HB_TAG('B','I','K',' ')}, /* Northern Catanduanes Bikol */ 322 {"cu", HB_TAG('C','S','L',' ')}, /* Church Slavic */ 323 {"cuc", HB_TAG('C','C','H','N')}, /* Chinantec */ 324 {"cuk", HB_TAG('C','U','K',' ')}, /* San Blas Kuna */ 325 {"cv", HB_TAG('C','H','U',' ')}, /* Chuvash */ 326 {"cvn", HB_TAG('C','C','H','N')}, /* Chinantec */ 327 {"cwd", HB_TAG('D','C','R',' ')}, /* Woods Cree */ 328 {"cy", HB_TAG('W','E','L',' ')}, /* Welsh */ 329 {"czt", HB_TAG('Q','I','N',' ')}, /* Zotung Chin */ 330 {"da", HB_TAG('D','A','N',' ')}, /* Danish */ 331 {"dao", HB_TAG('Q','I','N',' ')}, /* Daai Chin */ 332 {"dap", HB_TAG('N','I','S',' ')}, /* Nisi (India) */ 333 {"dar", HB_TAG('D','A','R',' ')}, /* Dargwa */ 334 {"dax", HB_TAG('D','A','X',' ')}, /* Dayi */ 335 {"de", HB_TAG('D','E','U',' ')}, /* German */ 336 {"dgo", HB_TAG('D','G','O',' ')}, /* Dogri */ 337 {"dhd", HB_TAG('M','A','W',' ')}, /* Dhundari */ 338 {"dhg", HB_TAG('D','H','G',' ')}, /* Dhangu */ 339 {"din", HB_TAG('D','N','K',' ')}, /* Dinka [macrolanguage] */ 340 {"diq", HB_TAG('D','I','Q',' ')}, /* Dimli */ 341 {"dje", HB_TAG('D','J','R',' ')}, /* Zarma */ 342 {"djr", HB_TAG('D','J','R','0')}, /* Djambarrpuyngu */ 343 {"dng", HB_TAG('D','U','N',' ')}, /* Dungan */ 344 {"dnj", HB_TAG('D','N','J',' ')}, /* Dan */ 345 {"doi", HB_TAG('D','G','R',' ')}, /* Dogri [macrolanguage] */ 346 {"dsb", HB_TAG('L','S','B',' ')}, /* Lower Sorbian */ 347 {"duj", HB_TAG('D','U','J',' ')}, /* Dhuwal */ 348 {"dv", HB_TAG('D','I','V',' ')}, /* Dhivehi/Divehi/Maldivian */ 349 {"dyu", HB_TAG('J','U','L',' ')}, /* Jula */ 350 {"dz", HB_TAG('D','Z','N',' ')}, /* Dzongkha */ 351 {"ee", HB_TAG('E','W','E',' ')}, /* Ewe */ 352 {"efi", HB_TAG('E','F','I',' ')}, /* Efik */ 353 {"ekk", HB_TAG('E','T','I',' ')}, /* Standard Estonian */ 354 {"el", HB_TAG('E','L','L',' ')}, /* Modern Greek (1453-) */ 355 {"emk", HB_TAG('M','N','K',' ')}, /* Eastern Maninkakan */ 356 {"en", HB_TAG('E','N','G',' ')}, /* English */ 357 {"enf", HB_TAG('F','N','E',' ')}, /* Forest Nenets */ 358 {"enh", HB_TAG('T','N','E',' ')}, /* Tundra Nenets */ 359 {"eo", HB_TAG('N','T','O',' ')}, /* Esperanto */ 360 {"eot", HB_TAG('B','T','I',' ')}, /* Beti (Côte d'Ivoire) */ 361 {"es", HB_TAG('E','S','P',' ')}, /* Spanish */ 362 {"esu", HB_TAG('E','S','U',' ')}, /* Central Yupik */ 363 {"et", HB_TAG('E','T','I',' ')}, /* Estonian [macrolanguage] */ 364 {"eu", HB_TAG('E','U','Q',' ')}, /* Basque */ 365 {"eve", HB_TAG('E','V','N',' ')}, /* Even */ 366 {"evn", HB_TAG('E','V','K',' ')}, /* Evenki */ 367 {"fa", HB_TAG('F','A','R',' ')}, /* Persian [macrolanguage] */ 368 {"fan", HB_TAG('F','A','N','0')}, /* Fang */ 369 {"fat", HB_TAG('F','A','T',' ')}, /* Fanti */ 370 {"ff", HB_TAG('F','U','L',' ')}, /* Fulah [macrolanguage] */ 371 {"fi", HB_TAG('F','I','N',' ')}, /* Finnish */ 372 {"fil", HB_TAG('P','I','L',' ')}, /* Filipino */ 373 {"fj", HB_TAG('F','J','I',' ')}, /* Fijian */ 374 {"flm", HB_TAG('H','A','L',' ')}, /* Halam/Falam Chin [retired ISO639 code] */ 375 {"fo", HB_TAG('F','O','S',' ')}, /* Faroese */ 376 {"fon", HB_TAG('F','O','N',' ')}, /* Fon */ 377 {"fr", HB_TAG('F','R','A',' ')}, /* French */ 378 {"frc", HB_TAG('F','R','C',' ')}, /* Cajun French */ 379 {"frp", HB_TAG('F','R','P',' ')}, /* Arpitan/Francoprovençal */ 380 {"fuf", HB_TAG('F','T','A',' ')}, /* Futa */ 381 {"fur", HB_TAG('F','R','L',' ')}, /* Friulian */ 382 {"fuv", HB_TAG('F','U','V',' ')}, /* Nigerian Fulfulde */ 383 {"fy", HB_TAG('F','R','I',' ')}, /* Western Frisian */ 384 {"ga", HB_TAG('I','R','I',' ')}, /* Irish */ 385 {"gaa", HB_TAG('G','A','D',' ')}, /* Ga */ 386 {"gag", HB_TAG('G','A','G',' ')}, /* Gagauz */ 387 {"gbm", HB_TAG('G','A','W',' ')}, /* Garhwali */ 388 {"gd", HB_TAG('G','A','E',' ')}, /* Scottish Gaelic */ 389 {"gez", HB_TAG('G','E','Z',' ')}, /* Ge'ez */ 390 {"ggo", HB_TAG('G','O','N',' ')}, /* Southern Gondi */ 391 {"gih", HB_TAG('G','I','H',' ')}, /* Githabul */ 392 {"gil", HB_TAG('G','I','L','0')}, /* Kiribati (Gilbertese) */ 393 {"gkp", HB_TAG('G','K','P',' ')}, /* Kpelle (Guinea) */ 394 {"gl", HB_TAG('G','A','L',' ')}, /* Galician */ 395 {"gld", HB_TAG('N','A','N',' ')}, /* Nanai */ 396 {"glk", HB_TAG('G','L','K',' ')}, /* Gilaki */ 397 {"gn", HB_TAG('G','U','A',' ')}, /* Guarani [macrolanguage] */ 398 {"gnn", HB_TAG('G','N','N',' ')}, /* Gumatj */ 399 {"gno", HB_TAG('G','O','N',' ')}, /* Northern Gondi */ 400 {"gog", HB_TAG('G','O','G',' ')}, /* Gogo */ 401 {"gon", HB_TAG('G','O','N',' ')}, /* Gondi [macrolanguage] */ 402 {"grt", HB_TAG('G','R','O',' ')}, /* Garo */ 403 {"gru", HB_TAG('S','O','G',' ')}, /* Sodo Gurage */ 404 {"gsw", HB_TAG('A','L','S',' ')}, /* Alsatian */ 405 {"gu", HB_TAG('G','U','J',' ')}, /* Gujarati */ 406 {"guc", HB_TAG('G','U','C',' ')}, /* Wayuu */ 407 {"guf", HB_TAG('G','U','F',' ')}, /* Gupapuyngu */ 408 {"guk", HB_TAG('G','M','Z',' ')}, /* Gumuz */ 409 /*{"guk", HB_TAG('G','U','K',' ')},*/ /* Gumuz (in SIL fonts) */ 410 {"guz", HB_TAG('G','U','Z',' ')}, /* Ekegusii/Gusii */ 411 {"gv", HB_TAG('M','N','X',' ')}, /* Manx */ 412 {"ha", HB_TAG('H','A','U',' ')}, /* Hausa */ 413 {"har", HB_TAG('H','R','I',' ')}, /* Harari */ 414 {"haw", HB_TAG('H','A','W',' ')}, /* Hawaiian */ 415 {"hay", HB_TAG('H','A','Y',' ')}, /* Haya */ 416 {"haz", HB_TAG('H','A','Z',' ')}, /* Hazaragi */ 417 {"he", HB_TAG('I','W','R',' ')}, /* Hebrew */ 418 {"hi", HB_TAG('H','I','N',' ')}, /* Hindi */ 419 {"hil", HB_TAG('H','I','L',' ')}, /* Hiligaynon */ 420 {"hlt", HB_TAG('Q','I','N',' ')}, /* Matu Chin */ 421 {"hmn", HB_TAG('H','M','N',' ')}, /* Hmong */ 422 {"hnd", HB_TAG('H','N','D',' ')}, /* [Southern] Hindko */ 423 {"hne", HB_TAG('C','H','H',' ')}, /* Chattisgarhi */ 424 {"hno", HB_TAG('H','N','D',' ')}, /* [Northern] Hindko */ 425 {"ho", HB_TAG('H','M','O',' ')}, /* Hiri Motu */ 426 {"hoc", HB_TAG('H','O',' ',' ')}, /* Ho */ 427 {"hoj", HB_TAG('H','A','R',' ')}, /* Harauti */ 428 {"hr", HB_TAG('H','R','V',' ')}, /* Croatian */ 429 {"hsb", HB_TAG('U','S','B',' ')}, /* Upper Sorbian */ 430 {"ht", HB_TAG('H','A','I',' ')}, /* Haitian/Haitian Creole */ 431 {"hu", HB_TAG('H','U','N',' ')}, /* Hungarian */ 432 {"hy", HB_TAG('H','Y','E',' ')}, /* Armenian */ 433 {"hz", HB_TAG('H','E','R',' ')}, /* Herero */ 434 {"ia", HB_TAG('I','N','A',' ')}, /* Interlingua (International Auxiliary Language Association) */ 435 {"iba", HB_TAG('I','B','A',' ')}, /* Iban */ 436 {"ibb", HB_TAG('I','B','B',' ')}, /* Ibibio */ 437 {"id", HB_TAG('I','N','D',' ')}, /* Indonesian */ 438 {"ie", HB_TAG('I','L','E',' ')}, /* Interlingue/Occidental */ 439 {"ig", HB_TAG('I','B','O',' ')}, /* Igbo */ 440 {"igb", HB_TAG('E','B','I',' ')}, /* Ebira */ 441 {"ii", HB_TAG('Y','I','M',' ')}, /* Yi Modern */ 442 {"ijc", HB_TAG('I','J','O',' ')}, /* Izon */ 443 {"ijo", HB_TAG('I','J','O',' ')}, /* Ijo [family] */ 444 {"ik", HB_TAG('I','P','K',' ')}, /* Inupiaq [macrolanguage] */ 445 {"ilo", HB_TAG('I','L','O',' ')}, /* Ilokano */ 446 {"inh", HB_TAG('I','N','G',' ')}, /* Ingush */ 447 {"io", HB_TAG('I','D','O',' ')}, /* Ido */ 448 {"is", HB_TAG('I','S','L',' ')}, /* Icelandic */ 449 {"it", HB_TAG('I','T','A',' ')}, /* Italian */ 450 {"iu", HB_TAG('I','N','U',' ')}, /* Inuktitut [macrolanguage] */ 451 {"ja", HB_TAG('J','A','N',' ')}, /* Japanese */ 452 {"jam", HB_TAG('J','A','M',' ')}, /* Jamaican Creole English */ 453 {"jbo", HB_TAG('J','B','O',' ')}, /* Lojban */ 454 {"jv", HB_TAG('J','A','V',' ')}, /* Javanese */ 455 {"ka", HB_TAG('K','A','T',' ')}, /* Georgian */ 456 {"kaa", HB_TAG('K','R','K',' ')}, /* Karakalpak */ 457 {"kab", HB_TAG('K','A','B','0')}, /* Kabyle */ 458 {"kam", HB_TAG('K','M','B',' ')}, /* Kamba (Kenya) */ 459 {"kar", HB_TAG('K','R','N',' ')}, /* Karen [family] */ 460 {"kat", HB_TAG('K','G','E',' ')}, /* Khutsuri Georgian */ 461 {"kbd", HB_TAG('K','A','B',' ')}, /* Kabardian */ 462 {"kde", HB_TAG('K','D','E',' ')}, /* Makonde */ 463 {"kdr", HB_TAG('K','R','M',' ')}, /* Karaim */ 464 {"kdt", HB_TAG('K','U','Y',' ')}, /* Kuy */ 465 {"kea", HB_TAG('K','E','A',' ')}, /* Kabuverdianu (Crioulo) */ 466 {"kek", HB_TAG('K','E','K',' ')}, /* Kekchi */ 467 {"kex", HB_TAG('K','K','N',' ')}, /* Kokni */ 468 {"kfa", HB_TAG('K','O','D',' ')}, /* Kodagu */ 469 {"kfr", HB_TAG('K','A','C',' ')}, /* Kachchi */ 470 {"kfx", HB_TAG('K','U','L',' ')}, /* Kulvi */ 471 {"kfy", HB_TAG('K','M','N',' ')}, /* Kumaoni */ 472 {"kg", HB_TAG('K','O','N',' ')}, /* Kongo [macrolanguage] */ 473 {"kha", HB_TAG('K','S','I',' ')}, /* Khasi */ 474 {"khb", HB_TAG('X','B','D',' ')}, /* Lü */ 475 {"kht", HB_TAG('K','H','N',' ')}, /* Khamti (Microsoft fonts) */ 476 /*{"kht", HB_TAG('K','H','T',' ')},*/ /* Khamti (OpenType spec and SIL fonts) */ 477 {"khw", HB_TAG('K','H','W',' ')}, /* Khowar */ 478 {"ki", HB_TAG('K','I','K',' ')}, /* Gikuyu/Kikuyu */ 479 {"kiu", HB_TAG('K','I','U',' ')}, /* Kirmanjki */ 480 {"kj", HB_TAG('K','U','A',' ')}, /* Kuanyama/Kwanyama */ 481 {"kjd", HB_TAG('K','J','D',' ')}, /* Southern Kiwai */ 482 {"kjh", HB_TAG('K','H','A',' ')}, /* Khakass */ 483 {"kjp", HB_TAG('K','J','P',' ')}, /* Pwo Eastern Karen */ 484 {"kk", HB_TAG('K','A','Z',' ')}, /* Kazakh */ 485 {"kl", HB_TAG('G','R','N',' ')}, /* Kalaallisut */ 486 {"kln", HB_TAG('K','A','L',' ')}, /* Kalenjin */ 487 {"km", HB_TAG('K','H','M',' ')}, /* Central Khmer */ 488 {"kmb", HB_TAG('M','B','N',' ')}, /* Kimbundu */ 489 {"kmw", HB_TAG('K','M','O',' ')}, /* Komo (Democratic Republic of Congo) */ 490 {"kn", HB_TAG('K','A','N',' ')}, /* Kannada */ 491 {"knn", HB_TAG('K','O','K',' ')}, /* Konkani */ 492 {"ko", HB_TAG('K','O','R',' ')}, /* Korean */ 493 {"koi", HB_TAG('K','O','P',' ')}, /* Komi-Permyak */ 494 {"kok", HB_TAG('K','O','K',' ')}, /* Konkani [macrolanguage] */ 495 {"kon", HB_TAG('K','O','N','0')}, /* Kongo */ 496 {"kos", HB_TAG('K','O','S',' ')}, /* Kosraean */ 497 {"kpe", HB_TAG('K','P','L',' ')}, /* Kpelle [macrolanguage] */ 498 {"kpv", HB_TAG('K','O','Z',' ')}, /* Komi-Zyrian */ 499 {"kpy", HB_TAG('K','Y','K',' ')}, /* Koryak */ 500 {"kqy", HB_TAG('K','R','T',' ')}, /* Koorete */ 501 {"kr", HB_TAG('K','N','R',' ')}, /* Kanuri [macrolanguage] */ 502 {"kri", HB_TAG('K','R','I',' ')}, /* Krio */ 503 {"krl", HB_TAG('K','R','L',' ')}, /* Karelian */ 504 {"kru", HB_TAG('K','U','U',' ')}, /* Kurukh */ 505 {"ks", HB_TAG('K','S','H',' ')}, /* Kashmiri */ 506 {"ksh", HB_TAG('K','S','H','0')}, /* Ripuarian, Kölsch */ 507 /*{"ksw", HB_TAG('K','R','N',' ')},*/ /* S'gaw Karen (Microsoft fonts?) */ 508 {"ksw", HB_TAG('K','S','W',' ')}, /* S'gaw Karen (OpenType spec and SIL fonts) */ 509 {"ktb", HB_TAG('K','E','B',' ')}, /* Kebena */ 510 {"ktu", HB_TAG('K','O','N',' ')}, /* Kikongo */ 511 {"ku", HB_TAG('K','U','R',' ')}, /* Kurdish [macrolanguage] */ 512 {"kum", HB_TAG('K','U','M',' ')}, /* Kumyk */ 513 {"kv", HB_TAG('K','O','M',' ')}, /* Komi [macrolanguage] */ 514 {"kvd", HB_TAG('K','U','I',' ')}, /* Kui (Indonesia) */ 515 {"kw", HB_TAG('C','O','R',' ')}, /* Cornish */ 516 {"kxc", HB_TAG('K','M','S',' ')}, /* Komso */ 517 {"kxu", HB_TAG('K','U','I',' ')}, /* Kui (India) */ 518 {"ky", HB_TAG('K','I','R',' ')}, /* Kirghiz/Kyrgyz */ 519 {"kyu", HB_TAG('K','Y','U',' ')}, /* Western Kayah */ 520 {"la", HB_TAG('L','A','T',' ')}, /* Latin */ 521 {"lad", HB_TAG('J','U','D',' ')}, /* Ladino */ 522 {"lb", HB_TAG('L','T','Z',' ')}, /* Luxembourgish */ 523 {"lbe", HB_TAG('L','A','K',' ')}, /* Lak */ 524 {"lbj", HB_TAG('L','D','K',' ')}, /* Ladakhi */ 525 {"lez", HB_TAG('L','E','Z',' ')}, /* Lezgi */ 526 {"lg", HB_TAG('L','U','G',' ')}, /* Ganda */ 527 {"li", HB_TAG('L','I','M',' ')}, /* Limburgan/Limburger/Limburgish */ 528 {"lif", HB_TAG('L','M','B',' ')}, /* Limbu */ 529 {"lij", HB_TAG('L','I','J',' ')}, /* Ligurian */ 530 {"lis", HB_TAG('L','I','S',' ')}, /* Lisu */ 531 {"ljp", HB_TAG('L','J','P',' ')}, /* Lampung Api */ 532 {"lki", HB_TAG('L','K','I',' ')}, /* Laki */ 533 {"lld", HB_TAG('L','A','D',' ')}, /* Ladin */ 534 {"lmn", HB_TAG('L','A','M',' ')}, /* Lambani */ 535 {"lmo", HB_TAG('L','M','O',' ')}, /* Lombard */ 536 {"ln", HB_TAG('L','I','N',' ')}, /* Lingala */ 537 {"lo", HB_TAG('L','A','O',' ')}, /* Lao */ 538 {"lom", HB_TAG('L','O','M',' ')}, /* Loma */ 539 {"lrc", HB_TAG('L','R','C',' ')}, /* Northern Luri */ 540 {"lt", HB_TAG('L','T','H',' ')}, /* Lithuanian */ 541 {"lu", HB_TAG('L','U','B',' ')}, /* Luba-Katanga */ 542 {"lua", HB_TAG('L','U','B',' ')}, /* Luba-Kasai */ 543 {"luo", HB_TAG('L','U','O',' ')}, /* Luo (Kenya and Tanzania) */ 544 {"lus", HB_TAG('M','I','Z',' ')}, /* Mizo */ 545 {"luy", HB_TAG('L','U','H',' ')}, /* Luyia/Oluluyia [macrolanguage] */ 546 {"luz", HB_TAG('L','R','C',' ')}, /* Southern Luri */ 547 {"lv", HB_TAG('L','V','I',' ')}, /* Latvian */ 548 {"lzz", HB_TAG('L','A','Z',' ')}, /* Laz */ 549 {"mad", HB_TAG('M','A','D',' ')}, /* Madurese */ 550 {"mag", HB_TAG('M','A','G',' ')}, /* Magahi */ 551 {"mai", HB_TAG('M','T','H',' ')}, /* Maithili */ 552 {"mak", HB_TAG('M','K','R',' ')}, /* Makasar */ 553 {"mam", HB_TAG('M','A','M',' ')}, /* Mam */ 554 {"man", HB_TAG('M','N','K',' ')}, /* Manding/Mandingo [macrolanguage] */ 555 {"mdc", HB_TAG('M','L','E',' ')}, /* Male (Papua New Guinea) */ 556 {"mdf", HB_TAG('M','O','K',' ')}, /* Moksha */ 557 {"mdr", HB_TAG('M','D','R',' ')}, /* Mandar */ 558 {"mdy", HB_TAG('M','L','E',' ')}, /* Male (Ethiopia) */ 559 {"men", HB_TAG('M','D','E',' ')}, /* Mende (Sierra Leone) */ 560 {"mer", HB_TAG('M','E','R',' ')}, /* Meru */ 561 {"mfe", HB_TAG('M','F','E',' ')}, /* Morisyen */ 562 {"mg", HB_TAG('M','L','G',' ')}, /* Malagasy [macrolanguage] */ 563 {"mh", HB_TAG('M','A','H',' ')}, /* Marshallese */ 564 {"mhr", HB_TAG('L','M','A',' ')}, /* Low Mari */ 565 {"mi", HB_TAG('M','R','I',' ')}, /* Maori */ 566 {"min", HB_TAG('M','I','N',' ')}, /* Minangkabau */ 567 {"mk", HB_TAG('M','K','D',' ')}, /* Macedonian */ 568 {"mku", HB_TAG('M','N','K',' ')}, /* Konyanka Maninka */ 569 {"mkw", HB_TAG('M','K','W',' ')}, /* Kituba (Congo) */ 570 {"ml", HB_TAG('M','L','R',' ')}, /* Malayalam */ 571 {"mlq", HB_TAG('M','N','K',' ')}, /* Western Maninkakan */ 572 {"mn", HB_TAG('M','N','G',' ')}, /* Mongolian [macrolanguage] */ 573 {"mnc", HB_TAG('M','C','H',' ')}, /* Manchu */ 574 {"mni", HB_TAG('M','N','I',' ')}, /* Manipuri */ 575 {"mnk", HB_TAG('M','N','D',' ')}, /* Mandinka */ 576 {"mns", HB_TAG('M','A','N',' ')}, /* Mansi */ 577 {"mnw", HB_TAG('M','O','N',' ')}, /* Mon */ 578 {"mo", HB_TAG('M','O','L',' ')}, /* Moldavian */ 579 {"moh", HB_TAG('M','O','H',' ')}, /* Mohawk */ 580 {"mos", HB_TAG('M','O','S',' ')}, /* Mossi */ 581 {"mpe", HB_TAG('M','A','J',' ')}, /* Majang */ 582 {"mr", HB_TAG('M','A','R',' ')}, /* Marathi */ 583 {"mrh", HB_TAG('Q','I','N',' ')}, /* Mara Chin */ 584 {"mrj", HB_TAG('H','M','A',' ')}, /* High Mari */ 585 {"ms", HB_TAG('M','L','Y',' ')}, /* Malay [macrolanguage] */ 586 {"msc", HB_TAG('M','N','K',' ')}, /* Sankaran Maninka */ 587 {"mt", HB_TAG('M','T','S',' ')}, /* Maltese */ 588 {"mtr", HB_TAG('M','A','W',' ')}, /* Mewari */ 589 {"mus", HB_TAG('M','U','S',' ')}, /* Creek */ 590 {"mve", HB_TAG('M','A','W',' ')}, /* Marwari (Pakistan) */ 591 {"mwk", HB_TAG('M','N','K',' ')}, /* Kita Maninkakan */ 592 {"mwl", HB_TAG('M','W','L',' ')}, /* Mirandese */ 593 {"mwr", HB_TAG('M','A','W',' ')}, /* Marwari [macrolanguage] */ 594 {"mww", HB_TAG('M','W','W',' ')}, /* Hmong Daw */ 595 {"my", HB_TAG('B','R','M',' ')}, /* Burmese */ 596 {"mym", HB_TAG('M','E','N',' ')}, /* Me'en */ 597 {"myn", HB_TAG('M','Y','N',' ')}, /* Mayan */ 598 {"myq", HB_TAG('M','N','K',' ')}, /* Forest Maninka (retired code) */ 599 {"myv", HB_TAG('E','R','Z',' ')}, /* Erzya */ 600 {"mzn", HB_TAG('M','Z','N',' ')}, /* Mazanderani */ 601 {"na", HB_TAG('N','A','U',' ')}, /* Nauru */ 602 {"nag", HB_TAG('N','A','G',' ')}, /* Naga-Assamese */ 603 {"nah", HB_TAG('N','A','H',' ')}, /* Nahuatl [family] */ 604 {"nap", HB_TAG('N','A','P',' ')}, /* Neapolitan */ 605 {"nb", HB_TAG('N','O','R',' ')}, /* Norwegian Bokmål */ 606 {"nco", HB_TAG('S','I','B',' ')}, /* Sibe */ 607 {"nd", HB_TAG('N','D','B',' ')}, /* [North] Ndebele */ 608 {"ndc", HB_TAG('N','D','C',' ')}, /* Ndau */ 609 {"nds", HB_TAG('N','D','S',' ')}, /* Low German/Low Saxon */ 610 {"ne", HB_TAG('N','E','P',' ')}, /* Nepali */ 611 {"new", HB_TAG('N','E','W',' ')}, /* Newari */ 612 {"ng", HB_TAG('N','D','G',' ')}, /* Ndonga */ 613 {"nga", HB_TAG('N','G','A',' ')}, /* Ngabaka */ 614 {"ngl", HB_TAG('L','M','W',' ')}, /* Lomwe */ 615 {"ngo", HB_TAG('S','X','T',' ')}, /* Sutu */ 616 {"niu", HB_TAG('N','I','U',' ')}, /* Niuean */ 617 {"niv", HB_TAG('G','I','L',' ')}, /* Gilyak */ 618 {"nl", HB_TAG('N','L','D',' ')}, /* Dutch */ 619 {"nn", HB_TAG('N','Y','N',' ')}, /* Norwegian Nynorsk */ 620 {"no", HB_TAG('N','O','R',' ')}, /* Norwegian [macrolanguage] */ 621 {"nod", HB_TAG('N','T','A',' ')}, /* Northern Thai */ 622 {"noe", HB_TAG('N','O','E',' ')}, /* Nimadi */ 623 {"nog", HB_TAG('N','O','G',' ')}, /* Nogai */ 624 {"nov", HB_TAG('N','O','V',' ')}, /* Novial */ 625 {"nqo", HB_TAG('N','K','O',' ')}, /* N'Ko */ 626 {"nr", HB_TAG('N','D','B',' ')}, /* [South] Ndebele */ 627 {"nsk", HB_TAG('N','A','S',' ')}, /* Naskapi */ 628 {"nso", HB_TAG('S','O','T',' ')}, /* [Northern] Sotho */ 629 {"nv", HB_TAG('N','A','V',' ')}, /* Navajo */ 630 {"ny", HB_TAG('C','H','I',' ')}, /* Chewa/Chichwa/Nyanja */ 631 {"nym", HB_TAG('N','Y','M',' ')}, /* Nyamwezi */ 632 {"nyn", HB_TAG('N','K','L',' ')}, /* Nyankole */ 633 {"oc", HB_TAG('O','C','I',' ')}, /* Occitan (post 1500) */ 634 {"oj", HB_TAG('O','J','B',' ')}, /* Ojibwa [macrolanguage] */ 635 {"ojs", HB_TAG('O','C','R',' ')}, /* Oji-Cree */ 636 {"okm", HB_TAG('K','O','H',' ')}, /* Korean Old Hangul */ 637 {"om", HB_TAG('O','R','O',' ')}, /* Oromo [macrolanguage] */ 638 {"or", HB_TAG('O','R','I',' ')}, /* Oriya */ 639 {"os", HB_TAG('O','S','S',' ')}, /* Ossetian */ 640 {"pa", HB_TAG('P','A','N',' ')}, /* Panjabi */ 641 {"pag", HB_TAG('P','A','G',' ')}, /* Pangasinan */ 642 {"pam", HB_TAG('P','A','M',' ')}, /* Kapampangan/Pampanga */ 643 {"pap", HB_TAG('P','A','P','0')}, /* Papiamento */ 644 {"pau", HB_TAG('P','A','U',' ')}, /* Palauan */ 645 {"pcc", HB_TAG('P','C','C',' ')}, /* Bouyei */ 646 {"pcd", HB_TAG('P','C','D',' ')}, /* Picard */ 647 {"pce", HB_TAG('P','L','G',' ')}, /* [Ruching] Palaung */ 648 {"pck", HB_TAG('Q','I','N',' ')}, /* Paite Chin */ 649 {"pdc", HB_TAG('P','D','C',' ')}, /* Pennsylvania German */ 650 {"pes", HB_TAG('F','A','R',' ')}, /* Iranian Persian */ 651 {"phk", HB_TAG('P','H','K',' ')}, /* Phake */ 652 {"pi", HB_TAG('P','A','L',' ')}, /* Pali */ 653 {"pih", HB_TAG('P','I','H',' ')}, /* Pitcairn-Norfolk */ 654 {"pl", HB_TAG('P','L','K',' ')}, /* Polish */ 655 {"pll", HB_TAG('P','L','G',' ')}, /* [Shwe] Palaung */ 656 {"plp", HB_TAG('P','A','P',' ')}, /* Palpa */ 657 {"pms", HB_TAG('P','M','S',' ')}, /* Piemontese */ 658 {"pnb", HB_TAG('P','N','B',' ')}, /* Western Panjabi */ 659 {"poh", HB_TAG('P','O','H',' ')}, /* Pocomchi */ 660 {"pon", HB_TAG('P','O','N',' ')}, /* Pohnpeian */ 661 {"prs", HB_TAG('D','R','I',' ')}, /* Afghan Persian/Dari */ 662 {"ps", HB_TAG('P','A','S',' ')}, /* Pashto/Pushto [macrolanguage] */ 663 {"pt", HB_TAG('P','T','G',' ')}, /* Portuguese */ 664 {"pwo", HB_TAG('P','W','O',' ')}, /* Pwo Western Karen */ 665 {"qu", HB_TAG('Q','U','Z',' ')}, /* Quechua [macrolanguage] */ 666 {"quc", HB_TAG('Q','U','C',' ')}, /* K'iche'/Quiché */ 667 {"quh", HB_TAG('Q','U','H',' ')}, /* Quechua (Bolivia) */ 668 {"quz", HB_TAG('Q','U','Z',' ')}, /* Cusco Quechua */ 669 {"qvi", HB_TAG('Q','V','I',' ')}, /* Quechua (Ecuador) */ 670 {"qwh", HB_TAG('Q','W','H',' ')}, /* Quechua (Peru) */ 671 {"raj", HB_TAG('R','A','J',' ')}, /* Rajasthani [macrolanguage] */ 672 {"rar", HB_TAG('R','A','R',' ')}, /* Rarotongan */ 673 {"rbb", HB_TAG('P','L','G',' ')}, /* Rumai Palaung */ 674 {"rej", HB_TAG('R','E','J',' ')}, /* Rejang */ 675 {"ria", HB_TAG('R','I','A',' ')}, /* Riang (India) */ 676 {"rif", HB_TAG('R','I','F',' ')}, /* Tarifit */ 677 {"ril", HB_TAG('R','I','A',' ')}, /* Riang (Myanmar) */ 678 {"rit", HB_TAG('R','I','T',' ')}, /* Ritarungo */ 679 {"rki", HB_TAG('A','R','K',' ')}, /* Rakhine */ 680 {"rkw", HB_TAG('R','K','W',' ')}, /* Arakwal */ 681 {"rm", HB_TAG('R','M','S',' ')}, /* Romansh */ 682 {"rmy", HB_TAG('R','M','Y',' ')}, /* Vlax Romani */ 683 {"rn", HB_TAG('R','U','N',' ')}, /* Rundi */ 684 {"ro", HB_TAG('R','O','M',' ')}, /* Romanian */ 685 {"rom", HB_TAG('R','O','Y',' ')}, /* Romany [macrolanguage] */ 686 {"rtm", HB_TAG('R','T','M',' ')}, /* Rotuman */ 687 {"ru", HB_TAG('R','U','S',' ')}, /* Russian */ 688 {"rue", HB_TAG('R','S','Y',' ')}, /* Rusyn */ 689 {"rup", HB_TAG('R','U','P',' ')}, /* Aromanian/Arumanian/Macedo-Romanian */ 690 {"rw", HB_TAG('R','U','A',' ')}, /* Kinyarwanda */ 691 {"rwr", HB_TAG('M','A','W',' ')}, /* Marwari (India) */ 692 {"sa", HB_TAG('S','A','N',' ')}, /* Sanskrit */ 693 {"sah", HB_TAG('Y','A','K',' ')}, /* Yakut */ 694 {"sam", HB_TAG('P','A','A',' ')}, /* Palestinian Aramaic */ 695 {"sas", HB_TAG('S','A','S',' ')}, /* Sasak */ 696 {"sat", HB_TAG('S','A','T',' ')}, /* Santali */ 697 {"sc", HB_TAG('S','R','D',' ')}, /* Sardinian [macrolanguage] */ 698 {"sck", HB_TAG('S','A','D',' ')}, /* Sadri */ 699 {"scn", HB_TAG('S','C','N',' ')}, /* Sicilian */ 700 {"sco", HB_TAG('S','C','O',' ')}, /* Scots */ 701 {"scs", HB_TAG('S','L','A',' ')}, /* [North] Slavey */ 702 {"sd", HB_TAG('S','N','D',' ')}, /* Sindhi */ 703 {"se", HB_TAG('N','S','M',' ')}, /* Northern Sami */ 704 {"seh", HB_TAG('S','N','A',' ')}, /* Sena */ 705 {"sel", HB_TAG('S','E','L',' ')}, /* Selkup */ 706 {"sez", HB_TAG('Q','I','N',' ')}, /* Senthang Chin */ 707 {"sg", HB_TAG('S','G','O',' ')}, /* Sango */ 708 {"sga", HB_TAG('S','G','A',' ')}, /* Old Irish (to 900) */ 709 {"sgs", HB_TAG('S','G','S',' ')}, /* Samogitian */ 710 {"sgw", HB_TAG('C','H','G',' ')}, /* Sebat Bet Gurage */ 711 /*{"sgw", HB_TAG('S','G','W',' ')},*/ /* Sebat Bet Gurage (in SIL fonts) */ 712 {"shi", HB_TAG('S','H','I',' ')}, /* Tachelhit */ 713 {"shn", HB_TAG('S','H','N',' ')}, /* Shan */ 714 {"si", HB_TAG('S','N','H',' ')}, /* Sinhala */ 715 {"sid", HB_TAG('S','I','D',' ')}, /* Sidamo */ 716 {"sjd", HB_TAG('K','S','M',' ')}, /* Kildin Sami */ 717 {"sk", HB_TAG('S','K','Y',' ')}, /* Slovak */ 718 {"skr", HB_TAG('S','R','K',' ')}, /* Seraiki */ 719 {"sl", HB_TAG('S','L','V',' ')}, /* Slovenian */ 720 {"sm", HB_TAG('S','M','O',' ')}, /* Samoan */ 721 {"sma", HB_TAG('S','S','M',' ')}, /* Southern Sami */ 722 {"smj", HB_TAG('L','S','M',' ')}, /* Lule Sami */ 723 {"smn", HB_TAG('I','S','M',' ')}, /* Inari Sami */ 724 {"sms", HB_TAG('S','K','S',' ')}, /* Skolt Sami */ 725 {"sn", HB_TAG('S','N','A','0')}, /* Shona */ 726 {"snk", HB_TAG('S','N','K',' ')}, /* Soninke */ 727 {"so", HB_TAG('S','M','L',' ')}, /* Somali */ 728 {"sop", HB_TAG('S','O','P',' ')}, /* Songe */ 729 {"sq", HB_TAG('S','Q','I',' ')}, /* Albanian [macrolanguage] */ 730 {"sr", HB_TAG('S','R','B',' ')}, /* Serbian */ 731 {"srr", HB_TAG('S','R','R',' ')}, /* Serer */ 732 {"ss", HB_TAG('S','W','Z',' ')}, /* Swati */ 733 {"st", HB_TAG('S','O','T',' ')}, /* [Southern] Sotho */ 734 {"stq", HB_TAG('S','T','Q',' ')}, /* Saterfriesisch */ 735 {"stv", HB_TAG('S','I','G',' ')}, /* Silt'e */ 736 {"su", HB_TAG('S','U','N',' ')}, /* Sundanese */ 737 {"suk", HB_TAG('S','U','K',' ')}, /* Sukama */ 738 {"suq", HB_TAG('S','U','R',' ')}, /* Suri */ 739 {"sv", HB_TAG('S','V','E',' ')}, /* Swedish */ 740 {"sva", HB_TAG('S','V','A',' ')}, /* Svan */ 741 {"sw", HB_TAG('S','W','K',' ')}, /* Swahili [macrolanguage] */ 742 {"swb", HB_TAG('C','M','R',' ')}, /* Comorian */ 743 {"swh", HB_TAG('S','W','K',' ')}, /* Kiswahili/Swahili */ 744 {"swv", HB_TAG('M','A','W',' ')}, /* Shekhawati */ 745 {"sxu", HB_TAG('S','X','U',' ')}, /* Upper Saxon */ 746 {"syc", HB_TAG('S','Y','R',' ')}, /* Classical Syriac */ 747 {"syl", HB_TAG('S','Y','L',' ')}, /* Sylheti */ 748 {"syr", HB_TAG('S','Y','R',' ')}, /* Syriac [macrolanguage] */ 749 {"szl", HB_TAG('S','Z','L',' ')}, /* Silesian */ 750 {"ta", HB_TAG('T','A','M',' ')}, /* Tamil */ 751 {"tab", HB_TAG('T','A','B',' ')}, /* Tabasaran */ 752 {"tcp", HB_TAG('Q','I','N',' ')}, /* Tawr Chin */ 753 {"tcy", HB_TAG('T','U','L',' ')}, /* Tulu */ 754 {"tcz", HB_TAG('Q','I','N',' ')}, /* Thado Chin */ 755 {"tdd", HB_TAG('T','D','D',' ')}, /* Tai Nüa */ 756 {"te", HB_TAG('T','E','L',' ')}, /* Telugu */ 757 {"tem", HB_TAG('T','M','N',' ')}, /* Temne */ 758 {"tet", HB_TAG('T','E','T',' ')}, /* Tetum */ 759 {"tg", HB_TAG('T','A','J',' ')}, /* Tajik */ 760 {"th", HB_TAG('T','H','A',' ')}, /* Thai */ 761 {"ti", HB_TAG('T','G','Y',' ')}, /* Tigrinya */ 762 {"tig", HB_TAG('T','G','R',' ')}, /* Tigre */ 763 {"tiv", HB_TAG('T','I','V',' ')}, /* Tiv */ 764 {"tk", HB_TAG('T','K','M',' ')}, /* Turkmen */ 765 {"tl", HB_TAG('T','G','L',' ')}, /* Tagalog */ 766 {"tmh", HB_TAG('T','M','H',' ')}, /* Tamashek */ 767 {"tn", HB_TAG('T','N','A',' ')}, /* Tswana */ 768 {"to", HB_TAG('T','G','N',' ')}, /* Tonga (Tonga Islands) */ 769 {"tod", HB_TAG('T','O','D','0')}, /* Toma */ 770 {"toi", HB_TAG('T','N','G',' ')}, /* Tonga */ 771 {"tpi", HB_TAG('T','P','I',' ')}, /* Tok Pisin */ 772 {"tr", HB_TAG('T','R','K',' ')}, /* Turkish */ 773 {"tru", HB_TAG('T','U','A',' ')}, /* Turoyo Aramaic */ 774 {"ts", HB_TAG('T','S','G',' ')}, /* Tsonga */ 775 {"tt", HB_TAG('T','A','T',' ')}, /* Tatar */ 776 {"tum", HB_TAG('T','U','M',' ')}, /* Tumbuka */ 777 {"tvl", HB_TAG('T','V','L',' ')}, /* Tuvalu */ 778 {"tw", HB_TAG('T','W','I',' ')}, /* Twi */ 779 {"ty", HB_TAG('T','H','T',' ')}, /* Tahitian */ 780 {"tyv", HB_TAG('T','U','V',' ')}, /* Tuvin */ 781 {"tyz", HB_TAG('T','Y','Z',' ')}, /* Tày */ 782 {"tzm", HB_TAG('T','Z','M',' ')}, /* Central Atlas Tamazight */ 783 {"tzo", HB_TAG('T','Z','O',' ')}, /* Tzotzil */ 784 {"udm", HB_TAG('U','D','M',' ')}, /* Udmurt */ 785 {"ug", HB_TAG('U','Y','G',' ')}, /* Uighur */ 786 {"uk", HB_TAG('U','K','R',' ')}, /* Ukrainian */ 787 {"umb", HB_TAG('U','M','B',' ')}, /* Umbundu */ 788 {"unr", HB_TAG('M','U','N',' ')}, /* Mundari */ 789 {"ur", HB_TAG('U','R','D',' ')}, /* Urdu */ 790 {"uz", HB_TAG('U','Z','B',' ')}, /* Uzbek [macrolanguage] */ 791 {"uzn", HB_TAG('U','Z','B',' ')}, /* Northern Uzbek */ 792 {"uzs", HB_TAG('U','Z','B',' ')}, /* Southern Uzbek */ 793 {"ve", HB_TAG('V','E','N',' ')}, /* Venda */ 794 {"vec", HB_TAG('V','E','C',' ')}, /* Venetian */ 795 {"vi", HB_TAG('V','I','T',' ')}, /* Vietnamese */ 796 {"vls", HB_TAG('F','L','E',' ')}, /* Vlaams */ 797 {"vmw", HB_TAG('M','A','K',' ')}, /* Makhuwa */ 798 {"vo", HB_TAG('V','O','L',' ')}, /* Volapük */ 799 {"vro", HB_TAG('V','R','O',' ')}, /* Võro */ 800 {"wa", HB_TAG('W','L','N',' ')}, /* Walloon */ 801 {"war", HB_TAG('W','A','R',' ')}, /* Waray (Philippines) */ 802 {"wbm", HB_TAG('W','A',' ',' ')}, /* Wa */ 803 {"wbr", HB_TAG('W','A','G',' ')}, /* Wagdi */ 804 {"wle", HB_TAG('S','I','G',' ')}, /* Wolane */ 805 {"wo", HB_TAG('W','L','F',' ')}, /* Wolof */ 806 {"wry", HB_TAG('M','A','W',' ')}, /* Merwari */ 807 {"wtm", HB_TAG('W','T','M',' ')}, /* Mewati */ 808 {"xal", HB_TAG('K','L','M',' ')}, /* Kalmyk */ 809 {"xan", HB_TAG('S','E','K',' ')}, /* Sekota */ 810 {"xh", HB_TAG('X','H','S',' ')}, /* Xhosa */ 811 {"xjb", HB_TAG('X','J','B',' ')}, /* Minjangbal */ 812 {"xog", HB_TAG('X','O','G',' ')}, /* Soga */ 813 {"xom", HB_TAG('K','M','O',' ')}, /* Komo (Sudan) */ 814 {"xpe", HB_TAG('X','P','E',' ')}, /* Kpelle (Liberia) */ 815 {"xsl", HB_TAG('S','S','L',' ')}, /* South Slavey */ 816 {"xst", HB_TAG('S','I','G',' ')}, /* Silt'e (retired code) */ 817 {"xwo", HB_TAG('T','O','D',' ')}, /* Written Oirat (Todo) */ 818 {"yao", HB_TAG('Y','A','O',' ')}, /* Yao */ 819 {"yap", HB_TAG('Y','A','P',' ')}, /* Yapese */ 820 {"yi", HB_TAG('J','I','I',' ')}, /* Yiddish [macrolanguage] */ 821 {"yo", HB_TAG('Y','B','A',' ')}, /* Yoruba */ 822 {"yos", HB_TAG('Q','I','N',' ')}, /* Yos, deprecated by IANA in favor of Zou [zom] */ 823 {"yso", HB_TAG('N','I','S',' ')}, /* Nisi (China) */ 824 {"za", HB_TAG('Z','H','A',' ')}, /* Chuang/Zhuang [macrolanguage] */ 825 {"zea", HB_TAG('Z','E','A',' ')}, /* Zeeuws */ 826 {"zgh", HB_TAG('Z','G','H',' ')}, /* Standard Morrocan Tamazigh */ 827 {"zne", HB_TAG('Z','N','D',' ')}, /* Zande */ 828 {"zom", HB_TAG('Q','I','N',' ')}, /* Zou */ 829 {"zu", HB_TAG('Z','U','L',' ')}, /* Zulu */ 830 {"zum", HB_TAG('L','R','C',' ')}, /* Kumzari */ 831 {"zza", HB_TAG('Z','Z','A',' ')}, /* Zazaki */ 832 833 /* The corresponding languages IDs for the following IDs are unclear, 834 * overlap, or are architecturally weird. Needs more research. */ 835 836 /*{"chp", HB_TAG('S','A','Y',' ')},*/ /* Sayisi */ 837 /*{"cwd", HB_TAG('T','C','R',' ')},*/ /* TH-Cree */ 838 /*{"emk", HB_TAG('E','M','K',' ')},*/ /* Eastern Maninkakan */ 839 /*{"krc", HB_TAG('B','A','L',' ')},*/ /* Balkar */ 840 /*{"??", HB_TAG('B','C','R',' ')},*/ /* Bible Cree */ 841 /*{"zh?", HB_TAG('C','H','N',' ')},*/ /* Chinese (seen in Microsoft fonts) */ 842 /*{"ar-Syrc?", HB_TAG('G','A','R',' ')},*/ /* Garshuni */ 843 /*{"hy?", HB_TAG('H','Y','E','0')},*/ /* Armenian East (ISO 639-3 hye according to Microsoft, but that’s equivalent to ISO 639-1 hy) */ 844 /*{"ga-Latg?/" HB_TAG('I','R','T',' ')},*/ /* Irish Traditional */ 845 /*{"krc", HB_TAG('K','A','R',' ')},*/ /* Karachay */ 846 /*{"ka-Geok?", HB_TAG('K','G','E',' ')},*/ /* Khutsuri Georgian */ 847 /*{"kca", HB_TAG('K','H','K',' ')},*/ /* Khanty-Kazim */ 848 /*{"kca", HB_TAG('K','H','S',' ')},*/ /* Khanty-Shurishkar */ 849 /*{"kca", HB_TAG('K','H','V',' ')},*/ /* Khanty-Vakhi */ 850 /*{"kqs, kss", HB_TAG('K','I','S',' ')},*/ /* Kisii */ 851 /*{"lua", HB_TAG('L','U','A',' ')},*/ /* Luba-Lulua */ 852 /*{"mlq", HB_TAG('M','L','N',' ')},*/ /* Malinke */ 853 /*{"nso", HB_TAG('N','S','O',' ')},*/ /* Sotho, Northern */ 854 /*{"??", HB_TAG('M','A','L',' ')},*/ /* Malayalam Traditional */ 855 /*{"csw", HB_TAG('N','C','R',' ')},*/ /* N-Cree */ 856 /*{"csw", HB_TAG('N','H','C',' ')},*/ /* Norway House Cree */ 857 /*{"el-polyton", HB_TAG('P','G','R',' ')},*/ /* Polytonic Greek */ 858 /*{"bgr, cnh, cnw, czt, sez, tcp, csy, ctd, flm, pck, tcz, zom, cmr, dao, hlt, cka, cnk, mrh, mwg, cbl, cnb, csh", HB_TAG('Q','I','N',' ')},*/ /* Chin */ 859 /*{"??", HB_TAG('Y','I','C',' ')},*/ /* Yi Classic */ 860 /*{"zh-Latn-pinyin", HB_TAG('Z','H','P',' ')},*/ /* Chinese Phonetic */ 861 }; 862 863 typedef struct { 864 char language[11]; 865 hb_tag_t tag; 866 } LangTagLong; 867 static const LangTagLong ot_languages_zh[] = { 868 /* Store longest-first, if one is a prefix of another. */ 869 {"zh-cn", HB_TAG('Z','H','S',' ')}, /* Chinese (China) */ 870 {"zh-hk", HB_TAG('Z','H','H',' ')}, /* Chinese (Hong Kong) */ 871 {"zh-mo", HB_TAG('Z','H','H',' ')}, /* Chinese (Macao) */ 872 {"zh-sg", HB_TAG('Z','H','S',' ')}, /* Chinese (Singapore) */ 873 {"zh-tw", HB_TAG('Z','H','T',' ')}, /* Chinese (Taiwan) */ 874 {"zh-hans", HB_TAG('Z','H','S',' ')}, /* Chinese (Simplified) */ 875 {"zh-hant-hk",HB_TAG('Z','H','H',' ')}, /* Chinese (Hong Kong) */ 876 {"zh-hant-mo",HB_TAG('Z','H','H',' ')}, /* Chinese (Macao) */ 877 {"zh-hant", HB_TAG('Z','H','T',' ')}, /* Chinese (Traditional) */ 878 }; 879 880 static int 881 lang_compare_first_component (const void *pa, 882 const void *pb) 883 { 884 const char *a = (const char *) pa; 885 const char *b = (const char *) pb; 886 unsigned int da, db; 887 const char *p; 888 889 p = strchr (a, '-'); 890 da = p ? (unsigned int) (p - a) : strlen (a); 891 892 p = strchr (b, '-'); 893 db = p ? (unsigned int) (p - b) : strlen (b); 894 895 return strncmp (a, b, MAX (da, db)); 896 } 897 898 static hb_bool_t 899 lang_matches (const char *lang_str, const char *spec) 900 { 901 unsigned int len = strlen (spec); 902 903 return strncmp (lang_str, spec, len) == 0 && 904 (lang_str[len] == '\0' || lang_str[len] == '-'); 905 } 906 907 hb_tag_t 908 hb_ot_tag_from_language (hb_language_t language) 909 { 910 const char *lang_str, *s; 911 912 if (language == HB_LANGUAGE_INVALID) 913 return HB_OT_TAG_DEFAULT_LANGUAGE; 914 915 lang_str = hb_language_to_string (language); 916 917 s = strstr (lang_str, "x-hbot"); 918 if (s) { 919 char tag[4]; 920 int i; 921 s += 6; 922 for (i = 0; i < 4 && ISALNUM (s[i]); i++) 923 tag[i] = TOUPPER (s[i]); 924 if (i) { 925 for (; i < 4; i++) 926 tag[i] = ' '; 927 return HB_TAG (tag[0], tag[1], tag[2], tag[3]); 928 } 929 } 930 931 /* 932 * "fonipa" is a variant tag in BCP-47, meaning the International Phonetic Alphabet. 933 * It can be applied to any language. 934 */ 935 if (strstr (lang_str, "-fonipa")) { 936 return HB_TAG('I','P','P','H'); /* Phonetic transcription—IPA conventions */ 937 } 938 939 /* 940 * "fonnapa" is a variant tag in BCP-47, meaning the North American Phonetic Alphabet 941 * also known as Americanist Phonetic Notation. It can be applied to any language. 942 */ 943 if (strstr (lang_str, "-fonnapa")) { 944 return HB_TAG('A','P','P','H'); /* Phonetic transcription—Americanist conventions */ 945 } 946 947 /* 948 * "Syre" is a BCP-47 script tag, meaning the Estrangela variant of the Syriac script. 949 * It can be applied to any language. 950 */ 951 if (strstr (lang_str, "-syre")) { 952 return HB_TAG('S','Y','R','E'); /* Estrangela Syriac */ 953 } 954 955 /* 956 * "Syrj" is a BCP-47 script tag, meaning the Western variant of the Syriac script. 957 * It can be applied to any language. 958 */ 959 if (strstr (lang_str, "-syrj")) { 960 return HB_TAG('S','Y','R','J'); /* Western Syriac */ 961 } 962 963 /* 964 * "Syrn" is a BCP-47 script tag, meaning the Eastern variant of the Syriac script. 965 * It can be applied to any language. 966 */ 967 if (strstr (lang_str, "-syrn")) { 968 return HB_TAG('S','Y','R','N'); /* Eastern Syriac */ 969 } 970 971 /* Find a language matching in the first component */ 972 { 973 const LangTag *lang_tag; 974 lang_tag = (LangTag *) bsearch (lang_str, ot_languages, 975 ARRAY_LENGTH (ot_languages), sizeof (LangTag), 976 lang_compare_first_component); 977 if (lang_tag) 978 return lang_tag->tag; 979 } 980 981 /* Otherwise, check the Chinese ones */ 982 if (0 == lang_compare_first_component (lang_str, "zh")) 983 { 984 unsigned int i; 985 986 for (i = 0; i < ARRAY_LENGTH (ot_languages_zh); i++) 987 { 988 const LangTagLong *lang_tag; 989 lang_tag = &ot_languages_zh[i]; 990 if (lang_matches (lang_str, lang_tag->language)) 991 return lang_tag->tag; 992 } 993 994 /* Otherwise just return 'ZHS ' */ 995 return HB_TAG('Z','H','S',' '); 996 } 997 998 s = strchr (lang_str, '-'); 999 if (!s) 1000 s = lang_str + strlen (lang_str); 1001 if (s - lang_str == 3) { 1002 /* Assume it's ISO-639-3 and upper-case and use it. */ 1003 return hb_tag_from_string (lang_str, s - lang_str) & ~0x20202000u; 1004 } 1005 1006 return HB_OT_TAG_DEFAULT_LANGUAGE; 1007 } 1008 1009 /** 1010 * hb_ot_tag_to_language: 1011 * 1012 * 1013 * 1014 * Return value: (transfer none): 1015 * 1016 * Since: 0.9.2 1017 **/ 1018 hb_language_t 1019 hb_ot_tag_to_language (hb_tag_t tag) 1020 { 1021 unsigned int i; 1022 1023 if (tag == HB_OT_TAG_DEFAULT_LANGUAGE) 1024 return nullptr; 1025 1026 /* struct LangTag has only room for 3-letter language tags. */ 1027 switch (tag) { 1028 case HB_TAG('A','P','P','H'): /* Phonetic transcription—Americanist conventions */ 1029 return hb_language_from_string ("und-fonnapa", -1); 1030 case HB_TAG('I','P','P','H'): /* Phonetic transcription—IPA conventions */ 1031 return hb_language_from_string ("und-fonipa", -1); 1032 case HB_TAG('S','Y','R',' '): /* Syriac [macrolanguage] */ 1033 return hb_language_from_string ("syr", -1); 1034 case HB_TAG('S','Y','R','E'): /* Estrangela Syriac */ 1035 return hb_language_from_string ("und-Syre", -1); 1036 case HB_TAG('S','Y','R','J'): /* Western Syriac */ 1037 return hb_language_from_string ("und-Syrj", -1); 1038 case HB_TAG('S','Y','R','N'): /* Eastern Syriac */ 1039 return hb_language_from_string ("und-Syrn", -1); 1040 } 1041 1042 for (i = 0; i < ARRAY_LENGTH (ot_languages); i++) 1043 if (ot_languages[i].tag == tag) 1044 return hb_language_from_string (ot_languages[i].language, -1); 1045 1046 /* If tag starts with ZH, it's Chinese */ 1047 if ((tag & 0xFFFF0000u) == 0x5A480000u) { 1048 switch (tag) { 1049 case HB_TAG('Z','H','H',' '): return hb_language_from_string ("zh-hk", -1); /* Hong Kong */ 1050 case HB_TAG('Z','H','S',' '): return hb_language_from_string ("zh-Hans", -1); /* Simplified */ 1051 case HB_TAG('Z','H','T',' '): return hb_language_from_string ("zh-Hant", -1); /* Traditional */ 1052 default: break; /* Fall through */ 1053 } 1054 } 1055 1056 /* Else return a custom language in the form of "x-hbotABCD" */ 1057 { 1058 unsigned char buf[11] = "x-hbot"; 1059 buf[6] = tag >> 24; 1060 buf[7] = (tag >> 16) & 0xFF; 1061 buf[8] = (tag >> 8) & 0xFF; 1062 buf[9] = tag & 0xFF; 1063 if (buf[9] == 0x20) 1064 buf[9] = '\0'; 1065 buf[10] = '\0'; 1066 return hb_language_from_string ((char *) buf, -1); 1067 } 1068 } 1069 1070 #ifdef MAIN 1071 static inline void 1072 test_langs_sorted (void) 1073 { 1074 for (unsigned int i = 1; i < ARRAY_LENGTH (ot_languages); i++) 1075 { 1076 int c = lang_compare_first_component (ot_languages[i-1].language, ot_languages[i].language); 1077 if (c >= 0) 1078 { 1079 fprintf (stderr, "ot_languages not sorted at index %d: %s %d %s\n", 1080 i, ot_languages[i-1].language, c, ot_languages[i].language); 1081 abort(); 1082 } 1083 } 1084 } 1085 1086 int 1087 main (void) 1088 { 1089 test_langs_sorted (); 1090 return 0; 1091 } 1092 1093 #endif | 9 * software and its documentation for any purpose, provided that the 10 * above copyright notice and the following two paragraphs appear in 11 * all copies of this software. 12 * 13 * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR 14 * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES 15 * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN 16 * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH 17 * DAMAGE. 18 * 19 * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, 20 * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND 21 * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS 22 * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO 23 * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. 24 * 25 * Red Hat Author(s): Behdad Esfahbod 26 * Google Author(s): Behdad Esfahbod, Roozbeh Pournader 27 */ 28 29 #include "hb.hh" 30 31 32 /* hb_script_t */ 33 34 static hb_tag_t 35 hb_ot_old_tag_from_script (hb_script_t script) 36 { 37 /* This seems to be accurate as of end of 2012. */ 38 39 switch ((hb_tag_t) script) 40 { 41 case HB_SCRIPT_INVALID: return HB_OT_TAG_DEFAULT_SCRIPT; 42 43 /* KATAKANA and HIRAGANA both map to 'kana' */ 44 case HB_SCRIPT_HIRAGANA: return HB_TAG('k','a','n','a'); 45 46 /* Spaces at the end are preserved, unlike ISO 15924 */ 47 case HB_SCRIPT_LAO: return HB_TAG('l','a','o',' '); 48 case HB_SCRIPT_YI: return HB_TAG('y','i',' ',' '); 49 /* Unicode-5.0 additions */ 50 case HB_SCRIPT_NKO: return HB_TAG('n','k','o',' '); 51 /* Unicode-5.1 additions */ 52 case HB_SCRIPT_VAI: return HB_TAG('v','a','i',' '); 53 } 54 55 /* Else, just change first char to lowercase and return */ 56 return ((hb_tag_t) script) | 0x20000000u; 57 } 58 59 static hb_script_t 60 hb_ot_old_tag_to_script (hb_tag_t tag) 61 { 62 if (unlikely (tag == HB_OT_TAG_DEFAULT_SCRIPT)) 63 return HB_SCRIPT_INVALID; 64 65 /* This side of the conversion is fully algorithmic. */ 66 67 /* Any spaces at the end of the tag are replaced by repeating the last 68 * letter. Eg 'nko ' -> 'Nkoo' */ 69 if (unlikely ((tag & 0x0000FF00u) == 0x00002000u)) 70 tag |= (tag >> 8) & 0x0000FF00u; /* Copy second letter to third */ 71 if (unlikely ((tag & 0x000000FFu) == 0x00000020u)) 72 tag |= (tag >> 8) & 0x000000FFu; /* Copy third letter to fourth */ 96 97 static hb_script_t 98 hb_ot_new_tag_to_script (hb_tag_t tag) 99 { 100 switch (tag) { 101 case HB_TAG('b','n','g','2'): return HB_SCRIPT_BENGALI; 102 case HB_TAG('d','e','v','2'): return HB_SCRIPT_DEVANAGARI; 103 case HB_TAG('g','j','r','2'): return HB_SCRIPT_GUJARATI; 104 case HB_TAG('g','u','r','2'): return HB_SCRIPT_GURMUKHI; 105 case HB_TAG('k','n','d','2'): return HB_SCRIPT_KANNADA; 106 case HB_TAG('m','l','m','2'): return HB_SCRIPT_MALAYALAM; 107 case HB_TAG('o','r','y','2'): return HB_SCRIPT_ORIYA; 108 case HB_TAG('t','m','l','2'): return HB_SCRIPT_TAMIL; 109 case HB_TAG('t','e','l','2'): return HB_SCRIPT_TELUGU; 110 case HB_TAG('m','y','m','2'): return HB_SCRIPT_MYANMAR; 111 } 112 113 return HB_SCRIPT_UNKNOWN; 114 } 115 116 void 117 hb_ot_tags_from_script (hb_script_t script, 118 hb_tag_t *script_tag_1, 119 hb_tag_t *script_tag_2) 120 { 121 unsigned int count = 2; 122 hb_tag_t tags[2]; 123 hb_ot_tags_from_script_and_language (script, HB_LANGUAGE_INVALID, &count, tags, nullptr, nullptr); 124 *script_tag_1 = count > 0 ? tags[0] : HB_OT_TAG_DEFAULT_SCRIPT; 125 *script_tag_2 = count > 1 ? tags[1] : HB_OT_TAG_DEFAULT_SCRIPT; 126 } 127 128 /* 129 * Complete list at: 130 * https://docs.microsoft.com/en-us/typography/opentype/spec/scripttags 131 * 132 * Most of the script tags are the same as the ISO 15924 tag but lowercased. 133 * So we just do that, and handle the exceptional cases in a switch. 134 */ 135 136 static void 137 hb_ot_all_tags_from_script (hb_script_t script, 138 unsigned int *count /* IN/OUT */, 139 hb_tag_t *tags /* OUT */) 140 { 141 unsigned int i = 0; 142 143 hb_tag_t new_tag = hb_ot_new_tag_from_script (script); 144 if (unlikely (new_tag != HB_OT_TAG_DEFAULT_SCRIPT)) 145 { 146 tags[i++] = new_tag | '3'; 147 if (*count > i) 148 tags[i++] = new_tag; 149 } 150 151 if (*count > i) 152 { 153 hb_tag_t old_tag = hb_ot_old_tag_from_script (script); 154 if (old_tag != HB_OT_TAG_DEFAULT_SCRIPT) 155 tags[i++] = old_tag; 156 } 157 158 *count = i; 159 } 160 161 hb_script_t 162 hb_ot_tag_to_script (hb_tag_t tag) 163 { 164 unsigned char digit = tag & 0x000000FFu; 165 if (unlikely (digit == '2' || digit == '3')) 166 return hb_ot_new_tag_to_script (tag & 0xFFFFFF32); 167 168 return hb_ot_old_tag_to_script (tag); 169 } 170 171 172 /* hb_language_t */ 173 174 static int 175 lang_compare_first_component (const void *pa, 176 const void *pb) 177 { 178 const char *a = (const char *) pa; 179 const char *b = (const char *) pb; 180 unsigned int da, db; 181 const char *p; 182 183 p = strchr (a, '-'); 184 da = p ? (unsigned int) (p - a) : strlen (a); 185 186 p = strchr (b, '-'); 187 db = p ? (unsigned int) (p - b) : strlen (b); 188 189 return strncmp (a, b, MAX (da, db)); 190 } 191 192 static bool 193 subtag_matches (const char *lang_str, 194 const char *limit, 195 const char *subtag) 196 { 197 do { 198 const char *s = strstr (lang_str, subtag); 199 if (!s || s >= limit) 200 return false; 201 if (!ISALNUM (s[strlen (subtag)])) 202 return true; 203 lang_str = s + strlen (subtag); 204 } while (true); 205 } 206 207 static hb_bool_t 208 lang_matches (const char *lang_str, const char *spec) 209 { 210 unsigned int len = strlen (spec); 211 212 return strncmp (lang_str, spec, len) == 0 && 213 (lang_str[len] == '\0' || lang_str[len] == '-'); 214 } 215 216 typedef struct { 217 char language[4]; 218 hb_tag_t tags[HB_OT_MAX_TAGS_PER_LANGUAGE]; 219 } LangTag; 220 221 #include "hb-ot-tag-table.hh" 222 223 /* The corresponding languages IDs for the following IDs are unclear, 224 * overlap, or are architecturally weird. Needs more research. */ 225 226 /*{"??", {HB_TAG('B','C','R',' ')}},*/ /* Bible Cree */ 227 /*{"zh?", {HB_TAG('C','H','N',' ')}},*/ /* Chinese (seen in Microsoft fonts) */ 228 /*{"ar-Syrc?", {HB_TAG('G','A','R',' ')}},*/ /* Garshuni */ 229 /*{"??", {HB_TAG('N','G','R',' ')}},*/ /* Nagari */ 230 /*{"??", {HB_TAG('Y','I','C',' ')}},*/ /* Yi Classic */ 231 /*{"zh?", {HB_TAG('Z','H','P',' ')}},*/ /* Chinese Phonetic */ 232 233 hb_tag_t 234 hb_ot_tag_from_language (hb_language_t language) 235 { 236 unsigned int count = 1; 237 hb_tag_t tags[1]; 238 hb_ot_tags_from_script_and_language (HB_SCRIPT_UNKNOWN, language, nullptr, nullptr, &count, tags); 239 return count > 0 ? tags[0] : HB_OT_TAG_DEFAULT_LANGUAGE; 240 } 241 242 static void 243 hb_ot_tags_from_language (const char *lang_str, 244 const char *limit, 245 unsigned int *count, 246 hb_tag_t *tags) 247 { 248 const char *s; 249 250 /* Check for matches of multiple subtags. */ 251 if (hb_ot_tags_from_complex_language (lang_str, limit, count, tags)) 252 return; 253 254 /* Find a language matching in the first component. */ 255 s = strchr (lang_str, '-'); 256 { 257 const LangTag *lang_tag; 258 if (s && limit - lang_str >= 6) 259 { 260 const char *extlang_end = strchr (s + 1, '-'); 261 /* If there is an extended language tag, use it. */ 262 if (3 == (extlang_end ? extlang_end - s - 1 : strlen (s + 1)) && 263 ISALPHA (s[1])) 264 lang_str = s + 1; 265 } 266 lang_tag = (LangTag *) bsearch (lang_str, ot_languages, 267 ARRAY_LENGTH (ot_languages), sizeof (LangTag), 268 lang_compare_first_component); 269 if (lang_tag) 270 { 271 unsigned int i; 272 for (i = 0; i < *count && lang_tag->tags[i] != HB_TAG_NONE; i++) 273 tags[i] = lang_tag->tags[i]; 274 *count = i; 275 return; 276 } 277 } 278 279 if (!s) 280 s = lang_str + strlen (lang_str); 281 if (s - lang_str == 3) { 282 /* Assume it's ISO-639-3 and upper-case and use it. */ 283 tags[0] = hb_tag_from_string (lang_str, s - lang_str) & ~0x20202000u; 284 *count = 1; 285 return; 286 } 287 288 *count = 0; 289 } 290 291 static bool 292 parse_private_use_subtag (const char *private_use_subtag, 293 unsigned int *count, 294 hb_tag_t *tags, 295 const char *prefix, 296 unsigned char (*normalize) (unsigned char)) 297 { 298 if (private_use_subtag && count && tags && *count) 299 { 300 const char *s = strstr (private_use_subtag, prefix); 301 if (s) 302 { 303 char tag[4]; 304 int i; 305 s += strlen (prefix); 306 for (i = 0; i < 4 && ISALNUM (s[i]); i++) 307 tag[i] = normalize (s[i]); 308 if (i) 309 { 310 for (; i < 4; i++) 311 tag[i] = ' '; 312 tags[0] = HB_TAG (tag[0], tag[1], tag[2], tag[3]); 313 if ((tags[0] & 0xDFDFDFDF) == HB_OT_TAG_DEFAULT_SCRIPT) 314 tags[0] ^= ~0xDFDFDFDF; 315 *count = 1; 316 return false; 317 } 318 } 319 } 320 return true; 321 } 322 323 /** 324 * hb_ot_tags_from_script_and_language: 325 * @script: an #hb_script_t to convert. 326 * @language: an #hb_language_t to convert. 327 * @script_count: (allow-none): maximum number of script tags to retrieve (IN) 328 * and actual number of script tags retrieved (OUT) 329 * @script_tags: (out) (allow-none): array of size at least @script_count to store the 330 * script tag results 331 * @language_count: (allow-none): maximum number of language tags to retrieve 332 * (IN) and actual number of language tags retrieved (OUT) 333 * @language_tags: (out) (allow-none): array of size at least @language_count to store 334 * the language tag results 335 * 336 * Converts an #hb_script_t and an #hb_language_t to script and language tags. 337 * 338 * Since: 2.0.0 339 **/ 340 void 341 hb_ot_tags_from_script_and_language (hb_script_t script, 342 hb_language_t language, 343 unsigned int *script_count /* IN/OUT */, 344 hb_tag_t *script_tags /* OUT */, 345 unsigned int *language_count /* IN/OUT */, 346 hb_tag_t *language_tags /* OUT */) 347 { 348 bool needs_script = true; 349 350 if (language == HB_LANGUAGE_INVALID) 351 { 352 if (language_count && language_tags && *language_count) 353 *language_count = 0; 354 } 355 else 356 { 357 const char *lang_str, *s, *limit, *private_use_subtag; 358 bool needs_language; 359 360 lang_str = hb_language_to_string (language); 361 limit = nullptr; 362 private_use_subtag = nullptr; 363 if (lang_str[0] == 'x' && lang_str[1] == '-') 364 { 365 private_use_subtag = lang_str; 366 } else { 367 for (s = lang_str + 1; *s; s++) 368 { 369 if (s[-1] == '-' && s[1] == '-') 370 { 371 if (s[0] == 'x') 372 { 373 private_use_subtag = s; 374 if (!limit) 375 limit = s - 1; 376 break; 377 } else if (!limit) 378 { 379 limit = s - 1; 380 } 381 } 382 } 383 if (!limit) 384 limit = s; 385 } 386 387 needs_script = parse_private_use_subtag (private_use_subtag, script_count, script_tags, "-hbsc", TOLOWER); 388 needs_language = parse_private_use_subtag (private_use_subtag, language_count, language_tags, "-hbot", TOUPPER); 389 390 if (needs_language && language_count && language_tags && *language_count) 391 hb_ot_tags_from_language (lang_str, limit, language_count, language_tags); 392 } 393 394 if (needs_script && script_count && script_tags && *script_count) 395 hb_ot_all_tags_from_script (script, script_count, script_tags); 396 } 397 398 /** 399 * hb_ot_tag_to_language: 400 * 401 * 402 * 403 * Return value: (transfer none): 404 * 405 * Since: 0.9.2 406 **/ 407 hb_language_t 408 hb_ot_tag_to_language (hb_tag_t tag) 409 { 410 unsigned int i; 411 412 if (tag == HB_OT_TAG_DEFAULT_LANGUAGE) 413 return nullptr; 414 415 { 416 hb_language_t disambiguated_tag = hb_ot_ambiguous_tag_to_language (tag); 417 if (disambiguated_tag != HB_LANGUAGE_INVALID) 418 return disambiguated_tag; 419 } 420 421 for (i = 0; i < ARRAY_LENGTH (ot_languages); i++) 422 if (ot_languages[i].tags[0] == tag) 423 return hb_language_from_string (ot_languages[i].language, -1); 424 425 /* Else return a custom language in the form of "x-hbotABCD" */ 426 { 427 unsigned char buf[11] = "x-hbot"; 428 buf[6] = tag >> 24; 429 buf[7] = (tag >> 16) & 0xFF; 430 buf[8] = (tag >> 8) & 0xFF; 431 buf[9] = tag & 0xFF; 432 if (buf[9] == 0x20) 433 buf[9] = '\0'; 434 buf[10] = '\0'; 435 return hb_language_from_string ((char *) buf, -1); 436 } 437 } 438 439 /** 440 * hb_ot_tags_to_script_and_language: 441 * @script_tag: a script tag 442 * @language_tag: a language tag 443 * @script: (allow-none): the #hb_script_t corresponding to @script_tag (OUT). 444 * @language: (allow-none): the #hb_language_t corresponding to @script_tag and 445 * @language_tag (OUT). 446 * 447 * Converts a script tag and a language tag to an #hb_script_t and an 448 * #hb_language_t. 449 * 450 * Since: 2.0.0 451 **/ 452 void 453 hb_ot_tags_to_script_and_language (hb_tag_t script_tag, 454 hb_tag_t language_tag, 455 hb_script_t *script /* OUT */, 456 hb_language_t *language /* OUT */) 457 { 458 hb_script_t script_out = hb_ot_tag_to_script (script_tag); 459 if (script) 460 *script = script_out; 461 if (language) 462 { 463 unsigned int script_count = 1; 464 hb_tag_t primary_script_tag[1]; 465 hb_ot_tags_from_script_and_language (script_out, 466 HB_LANGUAGE_INVALID, 467 &script_count, 468 primary_script_tag, 469 nullptr, nullptr); 470 *language = hb_ot_tag_to_language (language_tag); 471 if (script_count == 0 || primary_script_tag[0] != script_tag) 472 { 473 unsigned char *buf; 474 const char *lang_str = hb_language_to_string (*language); 475 size_t len = strlen (lang_str); 476 buf = (unsigned char *) malloc (len + 11); 477 if (unlikely (!buf)) 478 { 479 *language = nullptr; 480 } 481 else 482 { 483 memcpy (buf, lang_str, len); 484 if (lang_str[0] != 'x' || lang_str[1] != '-') { 485 buf[len++] = '-'; 486 buf[len++] = 'x'; 487 } 488 buf[len++] = '-'; 489 buf[len++] = 'h'; 490 buf[len++] = 'b'; 491 buf[len++] = 's'; 492 buf[len++] = 'c'; 493 buf[len++] = script_tag >> 24; 494 buf[len++] = (script_tag >> 16) & 0xFF; 495 buf[len++] = (script_tag >> 8) & 0xFF; 496 buf[len++] = script_tag & 0xFF; 497 *language = hb_language_from_string ((char *) buf, len); 498 free (buf); 499 } 500 } 501 } 502 } 503 504 #ifdef MAIN 505 static inline void 506 test_langs_sorted () 507 { 508 for (unsigned int i = 1; i < ARRAY_LENGTH (ot_languages); i++) 509 { 510 int c = lang_compare_first_component (ot_languages[i-1].language, ot_languages[i].language); 511 if (c >= 0) 512 { 513 fprintf (stderr, "ot_languages not sorted at index %d: %s %d %s\n", 514 i, ot_languages[i-1].language, c, ot_languages[i].language); 515 abort(); 516 } 517 } 518 } 519 520 int 521 main () 522 { 523 test_langs_sorted (); 524 return 0; 525 } 526 527 #endif |