1 /* Conversion from and to CP1258. 2 Copyright (C) 1998-2022 Free Software Foundation, Inc. 3 This file is part of the GNU C Library. 4 5 The GNU C Library is free software; you can redistribute it and/or 6 modify it under the terms of the GNU Lesser General Public 7 License as published by the Free Software Foundation; either 8 version 2.1 of the License, or (at your option) any later version. 9 10 The GNU C Library is distributed in the hope that it will be useful, 11 but WITHOUT ANY WARRANTY; without even the implied warranty of 12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 Lesser General Public License for more details. 14 15 You should have received a copy of the GNU Lesser General Public 16 License along with the GNU C Library; if not, see 17 <https://www.gnu.org/licenses/>. */ 18 19 #include <dlfcn.h> 20 #include <stdint.h> 21 #include <assert.h> 22 #include <stdlib.h> 23 24 #define NELEMS(arr) (sizeof (arr) / sizeof (arr[0])) 25 26 /* Definitions used in the body of the `gconv' function. */ 27 #define CHARSET_NAME "CP1258//" 28 #define FROM_LOOP from_cp1258 29 #define TO_LOOP to_cp1258 30 #define DEFINE_INIT 1 31 #define DEFINE_FINI 1 32 #define ONE_DIRECTION 0 33 #define FROM_LOOP_MIN_NEEDED_FROM 1 34 #define FROM_LOOP_MAX_NEEDED_FROM 1 35 #define FROM_LOOP_MIN_NEEDED_TO 4 36 #define FROM_LOOP_MAX_NEEDED_TO 4 37 #define TO_LOOP_MIN_NEEDED_FROM 4 38 #define TO_LOOP_MAX_NEEDED_FROM 4 39 #define TO_LOOP_MIN_NEEDED_TO 1 40 #define TO_LOOP_MAX_NEEDED_TO 2 41 #define PREPARE_LOOP \ 42 int saved_state; \ 43 int *statep = &data->__statep->__count; 44 #define EXTRA_LOOP_ARGS , statep 45 46 47 /* Since we might have to reset input pointer we must be able to save 48 and restore the state. */ 49 #define SAVE_RESET_STATE(Save) \ 50 if (Save) \ 51 saved_state = *statep; \ 52 else \ 53 *statep = saved_state 54 55 56 /* During CP1258 to UCS4 conversion, the COUNT element of the state 57 contains the last UCS4 character, shifted by 3 bits. */ 58 59 60 /* Since this is a stateful encoding we have to provide code which resets 61 the output state to the initial state. This has to be done during the 62 flushing. */ 63 #define EMIT_SHIFT_TO_INIT \ 64 if (data->__statep->__count != 0) \ 65 { \ 66 if (FROM_DIRECTION) \ 67 { \ 68 if (__glibc_likely (outbuf + 4 <= outend)) \ 69 { \ 70 /* Write out the last character. */ \ 71 *((uint32_t *) outbuf) = data->__statep->__count >> 3; \ 72 outbuf += sizeof (uint32_t); \ 73 data->__statep->__count = 0; \ 74 } \ 75 else \ 76 /* We don't have enough room in the output buffer. */ \ 77 status = __GCONV_FULL_OUTPUT; \ 78 } \ 79 else \ 80 /* We don't use shift states in the TO_DIRECTION. */ \ 81 data->__statep->__count = 0; \ 82 } 83 84 85 /* First define the conversion function from CP1258 to UCS4. */ 86 87 static const uint16_t to_ucs4[128] = 88 { 89 /* 0x80 */ 90 0x20AC, 0, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021, 91 0x02C6, 0x2030, 0, 0x2039, 0x0152, 0, 0, 0, 92 /* 0x90 */ 93 0, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, 94 0x02DC, 0x2122, 0, 0x203A, 0x0153, 0, 0, 0x0178, 95 /* 0xA0 */ 96 0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, 97 0x00A8, 0x00A9, 0x00AA, 0x00AB, 0x00AC, 0x00AD, 0x00AE, 0x00AF, 98 /* 0xB0 */ 99 0x00B0, 0x00B1, 0x00B2, 0x00B3, 0x00B4, 0x00B5, 0x00B6, 0x00B7, 100 0x00B8, 0x00B9, 0x00BA, 0x00BB, 0x00BC, 0x00BD, 0x00BE, 0x00BF, 101 /* 0xC0 */ 102 0x00C0, 0x00C1, 0x00C2, 0x0102, 0x00C4, 0x00C5, 0x00C6, 0x00C7, 103 0x00C8, 0x00C9, 0x00CA, 0x00CB, 0x0300, 0x00CD, 0x00CE, 0x00CF, 104 /* 0xD0 */ 105 0x0110, 0x00D1, 0x0309, 0x00D3, 0x00D4, 0x01A0, 0x00D6, 0x00D7, 106 0x00D8, 0x00D9, 0x00DA, 0x00DB, 0x00DC, 0x01AF, 0x0303, 0x00DF, 107 /* 0xE0 */ 108 0x00E0, 0x00E1, 0x00E2, 0x0103, 0x00E4, 0x00E5, 0x00E6, 0x00E7, 109 0x00E8, 0x00E9, 0x00EA, 0x00EB, 0x0301, 0x00ED, 0x00EE, 0x00EF, 110 /* 0xF0 */ 111 0x0111, 0x00F1, 0x0323, 0x00F3, 0x00F4, 0x01A1, 0x00F6, 0x00F7, 112 0x00F8, 0x00F9, 0x00FA, 0x00FB, 0x00FC, 0x01B0, 0x20AB, 0x00FF, 113 }; 114 115 /* CP1258 contains five combining characters: 116 0x0300, 0x0301, 0x0303, 0x0309, 0x0323. */ 117 118 /* Composition tables for each of the relevant combining characters. */ 119 static const struct 120 { 121 uint16_t base; 122 uint16_t composed; 123 } comp_table_data[] = 124 { 125 #define COMP_TABLE_IDX_0300 0 126 #define COMP_TABLE_LEN_0300 31 127 { 0x0041, 0x00C0 }, 128 { 0x0045, 0x00C8 }, 129 { 0x0049, 0x00CC }, 130 { 0x004E, 0x01F8 }, 131 { 0x004F, 0x00D2 }, 132 { 0x0055, 0x00D9 }, 133 { 0x0057, 0x1E80 }, 134 { 0x0059, 0x1EF2 }, 135 { 0x0061, 0x00E0 }, 136 { 0x0065, 0x00E8 }, 137 { 0x0069, 0x00EC }, 138 { 0x006E, 0x01F9 }, 139 { 0x006F, 0x00F2 }, 140 { 0x0075, 0x00F9 }, 141 { 0x0077, 0x1E81 }, 142 { 0x0079, 0x1EF3 }, 143 { 0x00A8, 0x1FED }, 144 { 0x00C2, 0x1EA6 }, 145 { 0x00CA, 0x1EC0 }, 146 { 0x00D4, 0x1ED2 }, 147 { 0x00DC, 0x01DB }, 148 { 0x00E2, 0x1EA7 }, 149 { 0x00EA, 0x1EC1 }, 150 { 0x00F4, 0x1ED3 }, 151 { 0x00FC, 0x01DC }, 152 { 0x0102, 0x1EB0 }, 153 { 0x0103, 0x1EB1 }, 154 /*{ 0x0112, 0x1E14 },*/ 155 /*{ 0x0113, 0x1E15 },*/ 156 /*{ 0x014C, 0x1E50 },*/ 157 /*{ 0x014D, 0x1E51 },*/ 158 { 0x01A0, 0x1EDC }, 159 { 0x01A1, 0x1EDD }, 160 { 0x01AF, 0x1EEA }, 161 { 0x01B0, 0x1EEB }, 162 #define COMP_TABLE_IDX_0301 (COMP_TABLE_IDX_0300 + COMP_TABLE_LEN_0300) 163 #define COMP_TABLE_LEN_0301 59 164 { 0x0041, 0x00C1 }, 165 { 0x0043, 0x0106 }, 166 { 0x0045, 0x00C9 }, 167 { 0x0047, 0x01F4 }, 168 { 0x0049, 0x00CD }, 169 { 0x004B, 0x1E30 }, 170 { 0x004C, 0x0139 }, 171 { 0x004D, 0x1E3E }, 172 { 0x004E, 0x0143 }, 173 { 0x004F, 0x00D3 }, 174 { 0x0050, 0x1E54 }, 175 { 0x0052, 0x0154 }, 176 { 0x0053, 0x015A }, 177 { 0x0055, 0x00DA }, 178 { 0x0057, 0x1E82 }, 179 { 0x0059, 0x00DD }, 180 { 0x005A, 0x0179 }, 181 { 0x0061, 0x00E1 }, 182 { 0x0063, 0x0107 }, 183 { 0x0065, 0x00E9 }, 184 { 0x0067, 0x01F5 }, 185 { 0x0069, 0x00ED }, 186 { 0x006B, 0x1E31 }, 187 { 0x006C, 0x013A }, 188 { 0x006D, 0x1E3F }, 189 { 0x006E, 0x0144 }, 190 { 0x006F, 0x00F3 }, 191 { 0x0070, 0x1E55 }, 192 { 0x0072, 0x0155 }, 193 { 0x0073, 0x015B }, 194 { 0x0075, 0x00FA }, 195 { 0x0077, 0x1E83 }, 196 { 0x0079, 0x00FD }, 197 { 0x007A, 0x017A }, 198 { 0x00A8, 0x0385 }, /* prefer U+0385 over U+1FEE */ 199 { 0x00C2, 0x1EA4 }, 200 { 0x00C5, 0x01FA }, 201 { 0x00C6, 0x01FC }, 202 { 0x00C7, 0x1E08 }, 203 { 0x00CA, 0x1EBE }, 204 { 0x00CF, 0x1E2E }, 205 { 0x00D4, 0x1ED0 }, 206 /*{ 0x00D5, 0x1E4C },*/ 207 { 0x00D8, 0x01FE }, 208 { 0x00DC, 0x01D7 }, 209 { 0x00E2, 0x1EA5 }, 210 { 0x00E5, 0x01FB }, 211 { 0x00E6, 0x01FD }, 212 { 0x00E7, 0x1E09 }, 213 { 0x00EA, 0x1EBF }, 214 { 0x00EF, 0x1E2F }, 215 { 0x00F4, 0x1ED1 }, 216 /*{ 0x00F5, 0x1E4D },*/ 217 { 0x00F8, 0x01FF }, 218 { 0x00FC, 0x01D8 }, 219 { 0x0102, 0x1EAE }, 220 { 0x0103, 0x1EAF }, 221 /*{ 0x0112, 0x1E16 },*/ 222 /*{ 0x0113, 0x1E17 },*/ 223 /*{ 0x014C, 0x1E52 },*/ 224 /*{ 0x014D, 0x1E53 },*/ 225 /*{ 0x0168, 0x1E78 },*/ 226 /*{ 0x0169, 0x1E79 },*/ 227 { 0x01A0, 0x1EDA }, 228 { 0x01A1, 0x1EDB }, 229 { 0x01AF, 0x1EE8 }, 230 { 0x01B0, 0x1EE9 }, 231 #define COMP_TABLE_IDX_0303 (COMP_TABLE_IDX_0301 + COMP_TABLE_LEN_0301) 232 #define COMP_TABLE_LEN_0303 34 233 { 0x0041, 0x00C3 }, 234 { 0x0045, 0x1EBC }, 235 { 0x0049, 0x0128 }, 236 { 0x004E, 0x00D1 }, 237 { 0x004F, 0x00D5 }, 238 { 0x0055, 0x0168 }, 239 { 0x0056, 0x1E7C }, 240 { 0x0059, 0x1EF8 }, 241 { 0x0061, 0x00E3 }, 242 { 0x0065, 0x1EBD }, 243 { 0x0069, 0x0129 }, 244 { 0x006E, 0x00F1 }, 245 { 0x006F, 0x00F5 }, 246 { 0x0075, 0x0169 }, 247 { 0x0076, 0x1E7D }, 248 { 0x0079, 0x1EF9 }, 249 { 0x00C2, 0x1EAA }, 250 { 0x00CA, 0x1EC4 }, 251 { 0x00D3, 0x1E4C }, 252 { 0x00D4, 0x1ED6 }, 253 { 0x00D6, 0x1E4E }, 254 { 0x00DA, 0x1E78 }, 255 { 0x00E2, 0x1EAB }, 256 { 0x00EA, 0x1EC5 }, 257 { 0x00F3, 0x1E4D }, 258 { 0x00F4, 0x1ED7 }, 259 { 0x00F6, 0x1E4F }, 260 { 0x00FA, 0x1E79 }, 261 { 0x0102, 0x1EB4 }, 262 { 0x0103, 0x1EB5 }, 263 { 0x01A0, 0x1EE0 }, 264 { 0x01A1, 0x1EE1 }, 265 { 0x01AF, 0x1EEE }, 266 { 0x01B0, 0x1EEF }, 267 #define COMP_TABLE_IDX_0309 (COMP_TABLE_IDX_0303 + COMP_TABLE_LEN_0303) 268 #define COMP_TABLE_LEN_0309 24 269 { 0x0041, 0x1EA2 }, 270 { 0x0045, 0x1EBA }, 271 { 0x0049, 0x1EC8 }, 272 { 0x004F, 0x1ECE }, 273 { 0x0055, 0x1EE6 }, 274 { 0x0059, 0x1EF6 }, 275 { 0x0061, 0x1EA3 }, 276 { 0x0065, 0x1EBB }, 277 { 0x0069, 0x1EC9 }, 278 { 0x006F, 0x1ECF }, 279 { 0x0075, 0x1EE7 }, 280 { 0x0079, 0x1EF7 }, 281 { 0x00C2, 0x1EA8 }, 282 { 0x00CA, 0x1EC2 }, 283 { 0x00D4, 0x1ED4 }, 284 { 0x00E2, 0x1EA9 }, 285 { 0x00EA, 0x1EC3 }, 286 { 0x00F4, 0x1ED5 }, 287 { 0x0102, 0x1EB2 }, 288 { 0x0103, 0x1EB3 }, 289 { 0x01A0, 0x1EDE }, 290 { 0x01A1, 0x1EDF }, 291 { 0x01AF, 0x1EEC }, 292 { 0x01B0, 0x1EED }, 293 #define COMP_TABLE_IDX_0323 (COMP_TABLE_IDX_0309 + COMP_TABLE_LEN_0309) 294 #define COMP_TABLE_LEN_0323 50 295 { 0x0041, 0x1EA0 }, 296 { 0x0042, 0x1E04 }, 297 { 0x0044, 0x1E0C }, 298 { 0x0045, 0x1EB8 }, 299 { 0x0048, 0x1E24 }, 300 { 0x0049, 0x1ECA }, 301 { 0x004B, 0x1E32 }, 302 { 0x004C, 0x1E36 }, 303 { 0x004D, 0x1E42 }, 304 { 0x004E, 0x1E46 }, 305 { 0x004F, 0x1ECC }, 306 { 0x0052, 0x1E5A }, 307 { 0x0053, 0x1E62 }, 308 { 0x0054, 0x1E6C }, 309 { 0x0055, 0x1EE4 }, 310 { 0x0056, 0x1E7E }, 311 { 0x0057, 0x1E88 }, 312 { 0x0059, 0x1EF4 }, 313 { 0x005A, 0x1E92 }, 314 { 0x0061, 0x1EA1 }, 315 { 0x0062, 0x1E05 }, 316 { 0x0064, 0x1E0D }, 317 { 0x0065, 0x1EB9 }, 318 { 0x0068, 0x1E25 }, 319 { 0x0069, 0x1ECB }, 320 { 0x006B, 0x1E33 }, 321 { 0x006C, 0x1E37 }, 322 { 0x006D, 0x1E43 }, 323 { 0x006E, 0x1E47 }, 324 { 0x006F, 0x1ECD }, 325 { 0x0072, 0x1E5B }, 326 { 0x0073, 0x1E63 }, 327 { 0x0074, 0x1E6D }, 328 { 0x0075, 0x1EE5 }, 329 { 0x0076, 0x1E7F }, 330 { 0x0077, 0x1E89 }, 331 { 0x0079, 0x1EF5 }, 332 { 0x007A, 0x1E93 }, 333 { 0x00C2, 0x1EAC }, 334 { 0x00CA, 0x1EC6 }, 335 { 0x00D4, 0x1ED8 }, 336 { 0x00E2, 0x1EAD }, 337 { 0x00EA, 0x1EC7 }, 338 { 0x00F4, 0x1ED9 }, 339 { 0x0102, 0x1EB6 }, 340 { 0x0103, 0x1EB7 }, 341 { 0x01A0, 0x1EE2 }, 342 { 0x01A1, 0x1EE3 }, 343 { 0x01AF, 0x1EF0 }, 344 { 0x01B0, 0x1EF1 }, 345 #define COMP_TABLE_IDX_END (COMP_TABLE_IDX_0323 + COMP_TABLE_LEN_0323) 346 }; 347 /* Compile-time verification of table size. */ 348 typedef int verify1[(NELEMS (comp_table_data) == COMP_TABLE_IDX_END) - 1]; 349 350 static const struct 351 { 352 unsigned int idx; 353 unsigned int len; 354 } comp_table[5] = 355 { 356 { COMP_TABLE_IDX_0300, COMP_TABLE_LEN_0300 }, 357 { COMP_TABLE_IDX_0301, COMP_TABLE_LEN_0301 }, 358 { COMP_TABLE_IDX_0303, COMP_TABLE_LEN_0303 }, 359 { COMP_TABLE_IDX_0309, COMP_TABLE_LEN_0309 }, 360 { COMP_TABLE_IDX_0323, COMP_TABLE_LEN_0323 } 361 }; 362 363 #define MIN_NEEDED_INPUT FROM_LOOP_MIN_NEEDED_FROM 364 #define MAX_NEEDED_INPUT FROM_LOOP_MAX_NEEDED_FROM 365 #define MIN_NEEDED_OUTPUT FROM_LOOP_MIN_NEEDED_TO 366 #define MAX_NEEDED_OUTPUT FROM_LOOP_MAX_NEEDED_TO 367 #define LOOPFCT FROM_LOOP 368 #define BODY \ 369 { \ 370 uint32_t ch = *inptr; \ 371 uint32_t last_ch; \ 372 int must_buffer_ch; \ 373 \ 374 if (ch >= 0x80) \ 375 { \ 376 ch = to_ucs4[ch - 0x80]; \ 377 if (__glibc_unlikely (ch == L'\0')) \ 378 { \ 379 /* This is an illegal character. */ \ 380 STANDARD_FROM_LOOP_ERR_HANDLER (1); \ 381 } \ 382 } \ 383 \ 384 /* Determine whether there is a buffered character pending. */ \ 385 last_ch = *statep >> 3; \ 386 \ 387 /* We have to buffer ch if it is a possible match in comp_table_data. */ \ 388 must_buffer_ch = (ch >= 0x0041 && ch <= 0x01b0); \ 389 \ 390 if (last_ch) \ 391 { \ 392 if (ch >= 0x0300 && ch < 0x0340) \ 393 { \ 394 /* See whether last_ch and ch can be combined. */ \ 395 unsigned int i, i1, i2; \ 396 \ 397 switch (ch) \ 398 { \ 399 case 0x0300: \ 400 i = 0; \ 401 break; \ 402 case 0x0301: \ 403 i = 1; \ 404 break; \ 405 case 0x0303: \ 406 i = 2; \ 407 break; \ 408 case 0x0309: \ 409 i = 3; \ 410 break; \ 411 case 0x0323: \ 412 i = 4; \ 413 break; \ 414 default: \ 415 abort (); \ 416 } \ 417 \ 418 i1 = comp_table[i].idx; \ 419 i2 = i1 + comp_table[i].len - 1; \ 420 \ 421 if (last_ch >= comp_table_data[i1].base \ 422 && last_ch <= comp_table_data[i2].base) \ 423 { \ 424 for (;;) \ 425 { \ 426 i = (i1 + i2) >> 1; \ 427 if (last_ch == comp_table_data[i].base) \ 428 break; \ 429 if (last_ch < comp_table_data[i].base) \ 430 { \ 431 if (i1 == i) \ 432 goto not_combining; \ 433 i2 = i; \ 434 } \ 435 else \ 436 { \ 437 if (i1 != i) \ 438 i1 = i; \ 439 else \ 440 { \ 441 i = i2; \ 442 if (last_ch == comp_table_data[i].base) \ 443 break; \ 444 goto not_combining; \ 445 } \ 446 } \ 447 } \ 448 last_ch = comp_table_data[i].composed; \ 449 /* Output the combined character. */ \ 450 put32 (outptr, last_ch); \ 451 outptr += 4; \ 452 *statep = 0; \ 453 ++inptr; \ 454 continue; \ 455 } \ 456 } \ 457 \ 458 not_combining: \ 459 /* Output the buffered character. */ \ 460 put32 (outptr, last_ch); \ 461 outptr += 4; \ 462 *statep = 0; \ 463 \ 464 /* If we don't have enough room to output ch as well, then deal \ 465 with it in another round. */ \ 466 if (!must_buffer_ch && __builtin_expect (outptr + 4 > outend, 0)) \ 467 continue; \ 468 } \ 469 \ 470 if (must_buffer_ch) \ 471 *statep = ch << 3; \ 472 else \ 473 { \ 474 put32 (outptr, ch); \ 475 outptr += 4; \ 476 } \ 477 ++inptr; \ 478 } 479 #define LOOP_NEED_FLAGS 480 #define EXTRA_LOOP_DECLS , int *statep 481 #define ONEBYTE_BODY \ 482 { \ 483 uint32_t ch; \ 484 \ 485 if (c < 0x80) \ 486 ch = c; \ 487 else \ 488 { \ 489 ch = to_ucs4[c - 0x80]; \ 490 if (ch == L'\0') \ 491 return WEOF; \ 492 } \ 493 if (ch >= 0x0041 && ch <= 0x01b0) \ 494 return WEOF; \ 495 return ch; \ 496 } 497 #include <iconv/loop.c> 498 499 500 /* Next, define the conversion function from UCS4 to CP1258. */ 501 502 static const unsigned char from_ucs4[] = 503 { 504 #define FROM_IDX_00 0 505 0xc4, 0xc5, 0xc6, 0xc7, /* 0x00c4-0x00c7 */ 506 0xc8, 0xc9, 0xca, 0xcb, 0x00, 0xcd, 0xce, 0xcf, /* 0x00c8-0x00cf */ 507 0x00, 0xd1, 0x00, 0xd3, 0xd4, 0x00, 0xd6, 0xd7, /* 0x00d0-0x00d7 */ 508 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0x00, 0x00, 0xdf, /* 0x00d8-0x00df */ 509 0xe0, 0xe1, 0xe2, 0x00, 0xe4, 0xe5, 0xe6, 0xe7, /* 0x00e0-0x00e7 */ 510 0xe8, 0xe9, 0xea, 0xeb, 0x00, 0xed, 0xee, 0xef, /* 0x00e8-0x00ef */ 511 0x00, 0xf1, 0x00, 0xf3, 0xf4, 0x00, 0xf6, 0xf7, /* 0x00f0-0x00f7 */ 512 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0x00, 0x00, 0xff, /* 0x00f8-0x00ff */ 513 0x00, 0x00, 0xc3, 0xe3, 0x00, 0x00, 0x00, 0x00, /* 0x0100-0x0107 */ 514 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0108-0x010f */ 515 0xd0, 0xf0, /* 0x0110-0x0111 */ 516 #define FROM_IDX_01 (FROM_IDX_00 + 78) 517 0x8c, 0x9c, 0x00, 0x00, 0x00, 0x00, /* 0x0152-0x0157 */ 518 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0158-0x015f */ 519 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0160-0x0167 */ 520 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0168-0x016f */ 521 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0170-0x0177 */ 522 0x9f, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0178-0x017f */ 523 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0180-0x0187 */ 524 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0188-0x018f */ 525 0x00, 0x00, 0x83, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0190-0x0197 */ 526 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0198-0x019f */ 527 0xd5, 0xf5, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x01a0-0x01a7 */ 528 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xdd, /* 0x01a8-0x01af */ 529 0xfd, /* 0x01b0-0x01b0 */ 530 #define FROM_IDX_02 (FROM_IDX_01 + 95) 531 0x88, 0x00, /* 0x02c6-0x02c7 */ 532 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x02c8-0x02cf */ 533 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x02d0-0x02d7 */ 534 0x00, 0x00, 0x00, 0x00, 0x98, /* 0x02d8-0x02dc */ 535 #define FROM_IDX_03 (FROM_IDX_02 + 23) 536 0xcc, 0xec, 0x00, 0xde, 0x00, 0x00, 0x00, 0x00, /* 0x0300-0x0307 */ 537 0x00, 0xd2, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0308-0x030f */ 538 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0310-0x0317 */ 539 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0318-0x031f */ 540 0x00, 0x00, 0x00, 0xf2, /* 0x0320-0x0323 */ 541 #define FROM_IDX_20 (FROM_IDX_03 + 36) 542 0x96, 0x97, 0x00, 0x00, 0x00, /* 0x2013-0x2017 */ 543 0x91, 0x92, 0x82, 0x00, 0x93, 0x94, 0x84, 0x00, /* 0x2018-0x201f */ 544 0x86, 0x87, 0x95, 0x00, 0x00, 0x00, 0x85, 0x00, /* 0x2020-0x2027 */ 545 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x2028-0x202f */ 546 0x89, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x2030-0x2037 */ 547 0x00, 0x8b, 0x9b, /* 0x2038-0x203a */ 548 #define FROM_IDX_FF (FROM_IDX_20 + 40) 549 }; 550 /* Compile-time verification of table size. */ 551 typedef int verify2[(NELEMS (from_ucs4) == FROM_IDX_FF) - 1]; 552 553 /* Decomposition table for the relevant Unicode characters. */ 554 static const struct 555 { 556 uint16_t composed; 557 uint32_t base:8; 558 uint32_t comb1:8; 559 } decomp_table[] = 560 { 561 { 0x00c0, 0x41, 0xcc }, 562 { 0x00c1, 0x41, 0xec }, 563 { 0x00c3, 0x41, 0xde }, 564 { 0x00c8, 0x45, 0xcc }, 565 { 0x00c9, 0x45, 0xec }, 566 { 0x00cc, 0x49, 0xcc }, 567 { 0x00cd, 0x49, 0xec }, 568 { 0x00d1, 0x4e, 0xde }, 569 { 0x00d2, 0x4f, 0xcc }, 570 { 0x00d3, 0x4f, 0xec }, 571 { 0x00d5, 0x4f, 0xde }, 572 { 0x00d9, 0x55, 0xcc }, 573 { 0x00da, 0x55, 0xec }, 574 { 0x00dd, 0x59, 0xec }, 575 { 0x00e0, 0x61, 0xcc }, 576 { 0x00e1, 0x61, 0xec }, 577 { 0x00e3, 0x61, 0xde }, 578 { 0x00e8, 0x65, 0xcc }, 579 { 0x00e9, 0x65, 0xec }, 580 { 0x00ec, 0x69, 0xcc }, 581 { 0x00ed, 0x69, 0xec }, 582 { 0x00f1, 0x6e, 0xde }, 583 { 0x00f2, 0x6f, 0xcc }, 584 { 0x00f3, 0x6f, 0xec }, 585 { 0x00f5, 0x6f, 0xde }, 586 { 0x00f9, 0x75, 0xcc }, 587 { 0x00fa, 0x75, 0xec }, 588 { 0x00fd, 0x79, 0xec }, 589 { 0x0106, 0x43, 0xec }, 590 { 0x0107, 0x63, 0xec }, 591 { 0x0128, 0x49, 0xde }, 592 { 0x0129, 0x69, 0xde }, 593 { 0x0139, 0x4c, 0xec }, 594 { 0x013a, 0x6c, 0xec }, 595 { 0x0143, 0x4e, 0xec }, 596 { 0x0144, 0x6e, 0xec }, 597 { 0x0154, 0x52, 0xec }, 598 { 0x0155, 0x72, 0xec }, 599 { 0x015a, 0x53, 0xec }, 600 { 0x015b, 0x73, 0xec }, 601 { 0x0168, 0x55, 0xde }, 602 { 0x0169, 0x75, 0xde }, 603 { 0x0179, 0x5a, 0xec }, 604 { 0x017a, 0x7a, 0xec }, 605 { 0x01d7, 0xdc, 0xec }, 606 { 0x01d8, 0xfc, 0xec }, 607 { 0x01db, 0xdc, 0xcc }, 608 { 0x01dc, 0xfc, 0xcc }, 609 { 0x01f4, 0x47, 0xec }, 610 { 0x01f5, 0x67, 0xec }, 611 { 0x01f8, 0x4e, 0xcc }, 612 { 0x01f9, 0x6e, 0xcc }, 613 { 0x01fa, 0xc5, 0xec }, 614 { 0x01fb, 0xe5, 0xec }, 615 { 0x01fc, 0xc6, 0xec }, 616 { 0x01fd, 0xe6, 0xec }, 617 { 0x01fe, 0xd8, 0xec }, 618 { 0x01ff, 0xf8, 0xec }, 619 { 0x0385, 0xa8, 0xec }, 620 { 0x1e04, 0x42, 0xf2 }, 621 { 0x1e05, 0x62, 0xf2 }, 622 { 0x1e08, 0xc7, 0xec }, 623 { 0x1e09, 0xe7, 0xec }, 624 { 0x1e0c, 0x44, 0xf2 }, 625 { 0x1e0d, 0x64, 0xf2 }, 626 { 0x1e24, 0x48, 0xf2 }, 627 { 0x1e25, 0x68, 0xf2 }, 628 { 0x1e2e, 0xcf, 0xec }, 629 { 0x1e2f, 0xef, 0xec }, 630 { 0x1e30, 0x4b, 0xec }, 631 { 0x1e31, 0x6b, 0xec }, 632 { 0x1e32, 0x4b, 0xf2 }, 633 { 0x1e33, 0x6b, 0xf2 }, 634 { 0x1e36, 0x4c, 0xf2 }, 635 { 0x1e37, 0x6c, 0xf2 }, 636 { 0x1e3e, 0x4d, 0xec }, 637 { 0x1e3f, 0x6d, 0xec }, 638 { 0x1e42, 0x4d, 0xf2 }, 639 { 0x1e43, 0x6d, 0xf2 }, 640 { 0x1e46, 0x4e, 0xf2 }, 641 { 0x1e47, 0x6e, 0xf2 }, 642 { 0x1e4c, 0xd3, 0xde }, /*{ 0x1e4c, 0x00d5, 0xec }, { 0x1e4c, 0x004f, 1, 0xde },*/ 643 { 0x1e4d, 0xf3, 0xde }, /*{ 0x1e4d, 0x00f5, 0xec }, { 0x1e4d, 0x006f, 1, 0xde },*/ 644 { 0x1e4e, 0xd6, 0xde }, 645 { 0x1e4f, 0xf6, 0xde }, 646 { 0x1e54, 0x50, 0xec }, 647 { 0x1e55, 0x70, 0xec }, 648 { 0x1e5a, 0x52, 0xf2 }, 649 { 0x1e5b, 0x72, 0xf2 }, 650 { 0x1e62, 0x53, 0xf2 }, 651 { 0x1e63, 0x73, 0xf2 }, 652 { 0x1e6c, 0x54, 0xf2 }, 653 { 0x1e6d, 0x74, 0xf2 }, 654 { 0x1e78, 0xda, 0xde }, /*{ 0x1e78, 0x0168, 0xec }, { 0x1e78, 0x0055, 1, 0xde },*/ 655 { 0x1e79, 0xfa, 0xde }, /*{ 0x1e79, 0x0169, 0xec }, { 0x1e79, 0x0075, 1, 0xde },*/ 656 { 0x1e7c, 0x56, 0xde }, 657 { 0x1e7d, 0x76, 0xde }, 658 { 0x1e7e, 0x56, 0xf2 }, 659 { 0x1e7f, 0x76, 0xf2 }, 660 { 0x1e80, 0x57, 0xcc }, 661 { 0x1e81, 0x77, 0xcc }, 662 { 0x1e82, 0x57, 0xec }, 663 { 0x1e83, 0x77, 0xec }, 664 { 0x1e88, 0x57, 0xf2 }, 665 { 0x1e89, 0x77, 0xf2 }, 666 { 0x1e92, 0x5a, 0xf2 }, 667 { 0x1e93, 0x7a, 0xf2 }, 668 { 0x1ea0, 0x41, 0xf2 }, 669 { 0x1ea1, 0x61, 0xf2 }, 670 { 0x1ea2, 0x41, 0xd2 }, 671 { 0x1ea3, 0x61, 0xd2 }, 672 { 0x1ea4, 0xc2, 0xec }, 673 { 0x1ea5, 0xe2, 0xec }, 674 { 0x1ea6, 0xc2, 0xcc }, 675 { 0x1ea7, 0xe2, 0xcc }, 676 { 0x1ea8, 0xc2, 0xd2 }, 677 { 0x1ea9, 0xe2, 0xd2 }, 678 { 0x1eaa, 0xc2, 0xde }, 679 { 0x1eab, 0xe2, 0xde }, 680 { 0x1eac, 0xc2, 0xf2 }, 681 { 0x1ead, 0xe2, 0xf2 }, 682 { 0x1eae, 0xc3, 0xec }, 683 { 0x1eaf, 0xe3, 0xec }, 684 { 0x1eb0, 0xc3, 0xcc }, 685 { 0x1eb1, 0xe3, 0xcc }, 686 { 0x1eb2, 0xc3, 0xd2 }, 687 { 0x1eb3, 0xe3, 0xd2 }, 688 { 0x1eb4, 0xc3, 0xde }, 689 { 0x1eb5, 0xe3, 0xde }, 690 { 0x1eb6, 0xc3, 0xf2 }, 691 { 0x1eb7, 0xe3, 0xf2 }, 692 { 0x1eb8, 0x45, 0xf2 }, 693 { 0x1eb9, 0x65, 0xf2 }, 694 { 0x1eba, 0x45, 0xd2 }, 695 { 0x1ebb, 0x65, 0xd2 }, 696 { 0x1ebc, 0x45, 0xde }, 697 { 0x1ebd, 0x65, 0xde }, 698 { 0x1ebe, 0xca, 0xec }, 699 { 0x1ebf, 0xea, 0xec }, 700 { 0x1ec0, 0xca, 0xcc }, 701 { 0x1ec1, 0xea, 0xcc }, 702 { 0x1ec2, 0xca, 0xd2 }, 703 { 0x1ec3, 0xea, 0xd2 }, 704 { 0x1ec4, 0xca, 0xde }, 705 { 0x1ec5, 0xea, 0xde }, 706 { 0x1ec6, 0xca, 0xf2 }, 707 { 0x1ec7, 0xea, 0xf2 }, 708 { 0x1ec8, 0x49, 0xd2 }, 709 { 0x1ec9, 0x69, 0xd2 }, 710 { 0x1eca, 0x49, 0xf2 }, 711 { 0x1ecb, 0x69, 0xf2 }, 712 { 0x1ecc, 0x4f, 0xf2 }, 713 { 0x1ecd, 0x6f, 0xf2 }, 714 { 0x1ece, 0x4f, 0xd2 }, 715 { 0x1ecf, 0x6f, 0xd2 }, 716 { 0x1ed0, 0xd4, 0xec }, 717 { 0x1ed1, 0xf4, 0xec }, 718 { 0x1ed2, 0xd4, 0xcc }, 719 { 0x1ed3, 0xf4, 0xcc }, 720 { 0x1ed4, 0xd4, 0xd2 }, 721 { 0x1ed5, 0xf4, 0xd2 }, 722 { 0x1ed6, 0xd4, 0xde }, 723 { 0x1ed7, 0xf4, 0xde }, 724 { 0x1ed8, 0xd4, 0xf2 }, 725 { 0x1ed9, 0xf4, 0xf2 }, 726 { 0x1eda, 0xd5, 0xec }, 727 { 0x1edb, 0xf5, 0xec }, 728 { 0x1edc, 0xd5, 0xcc }, 729 { 0x1edd, 0xf5, 0xcc }, 730 { 0x1ede, 0xd5, 0xd2 }, 731 { 0x1edf, 0xf5, 0xd2 }, 732 { 0x1ee0, 0xd5, 0xde }, 733 { 0x1ee1, 0xf5, 0xde }, 734 { 0x1ee2, 0xd5, 0xf2 }, 735 { 0x1ee3, 0xf5, 0xf2 }, 736 { 0x1ee4, 0x55, 0xf2 }, 737 { 0x1ee5, 0x75, 0xf2 }, 738 { 0x1ee6, 0x55, 0xd2 }, 739 { 0x1ee7, 0x75, 0xd2 }, 740 { 0x1ee8, 0xdd, 0xec }, 741 { 0x1ee9, 0xfd, 0xec }, 742 { 0x1eea, 0xdd, 0xcc }, 743 { 0x1eeb, 0xfd, 0xcc }, 744 { 0x1eec, 0xdd, 0xd2 }, 745 { 0x1eed, 0xfd, 0xd2 }, 746 { 0x1eee, 0xdd, 0xde }, 747 { 0x1eef, 0xfd, 0xde }, 748 { 0x1ef0, 0xdd, 0xf2 }, 749 { 0x1ef1, 0xfd, 0xf2 }, 750 { 0x1ef2, 0x59, 0xcc }, 751 { 0x1ef3, 0x79, 0xcc }, 752 { 0x1ef4, 0x59, 0xf2 }, 753 { 0x1ef5, 0x79, 0xf2 }, 754 { 0x1ef6, 0x59, 0xd2 }, 755 { 0x1ef7, 0x79, 0xd2 }, 756 { 0x1ef8, 0x59, 0xde }, 757 { 0x1ef9, 0x79, 0xde }, 758 { 0x1fed, 0xa8, 0xcc }, 759 { 0x1fee, 0xa8, 0xec }, 760 }; 761 762 #define MIN_NEEDED_INPUT TO_LOOP_MIN_NEEDED_FROM 763 #define MAX_NEEDED_INPUT TO_LOOP_MAX_NEEDED_FROM 764 #define MIN_NEEDED_OUTPUT TO_LOOP_MIN_NEEDED_TO 765 #define MAX_NEEDED_OUTPUT TO_LOOP_MAX_NEEDED_TO 766 #define LOOPFCT TO_LOOP 767 #define BODY \ 768 { \ 769 uint32_t ch = get32 (inptr); \ 770 \ 771 if (ch < 0x0080 || (ch >= 0x00a0 && ch < 0x00c3)) \ 772 { \ 773 *outptr++ = ch; \ 774 inptr += 4; \ 775 } \ 776 else \ 777 { \ 778 unsigned char res; \ 779 \ 780 if (ch >= 0x00c4 && ch < 0x0112) \ 781 res = from_ucs4[ch - 0x00c4 + FROM_IDX_00]; \ 782 else if (ch >= 0x0152 && ch < 0x01b1) \ 783 res = from_ucs4[ch - 0x0152 + FROM_IDX_01]; \ 784 else if (ch >= 0x02c6 && ch < 0x02dd) \ 785 res = from_ucs4[ch - 0x02c6 + FROM_IDX_02]; \ 786 else if (ch >= 0x0300 && ch < 0x0324) \ 787 res = from_ucs4[ch - 0x0300 + FROM_IDX_03]; \ 788 else if (ch >= 0x0340 && ch < 0x0342) /* Vietnamese tone marks */ \ 789 res = from_ucs4[ch - 0x0340 + FROM_IDX_03]; \ 790 else if (ch >= 0x2013 && ch < 0x203b) \ 791 res = from_ucs4[ch - 0x2013 + FROM_IDX_20]; \ 792 else if (ch == 0x20ab) \ 793 res = 0xfe; \ 794 else if (ch == 0x20ac) \ 795 res = 0x80; \ 796 else if (ch == 0x2122) \ 797 res = 0x99; \ 798 else \ 799 { \ 800 UNICODE_TAG_HANDLER (ch, 4); \ 801 res = 0; \ 802 } \ 803 \ 804 if (__glibc_likely (res != 0)) \ 805 { \ 806 *outptr++ = res; \ 807 inptr += 4; \ 808 } \ 809 else \ 810 { \ 811 /* Try canonical decomposition. */ \ 812 unsigned int i1, i2; \ 813 \ 814 i1 = 0; \ 815 i2 = sizeof (decomp_table) / sizeof (decomp_table[0]) - 1; \ 816 if (ch >= decomp_table[i1].composed \ 817 && ch <= decomp_table[i2].composed) \ 818 { \ 819 unsigned int i; \ 820 \ 821 for (;;) \ 822 { \ 823 i = (i1 + i2) >> 1; \ 824 if (ch == decomp_table[i].composed) \ 825 break; \ 826 if (ch < decomp_table[i].composed) \ 827 { \ 828 if (i1 == i) \ 829 goto failed; \ 830 i2 = i; \ 831 } \ 832 else \ 833 { \ 834 if (i1 != i) \ 835 i1 = i; \ 836 else \ 837 { \ 838 i = i2; \ 839 if (ch == decomp_table[i].composed) \ 840 break; \ 841 goto failed; \ 842 } \ 843 } \ 844 } \ 845 \ 846 /* See whether we have room for two bytes. */ \ 847 if (__glibc_unlikely (outptr + 1 >= outend)) \ 848 { \ 849 result = __GCONV_FULL_OUTPUT; \ 850 break; \ 851 } \ 852 \ 853 /* Found a canonical decomposition. */ \ 854 *outptr++ = decomp_table[i].base; \ 855 *outptr++ = decomp_table[i].comb1; \ 856 inptr += 4; \ 857 continue; \ 858 } \ 859 \ 860 failed: \ 861 /* This is an illegal character. */ \ 862 STANDARD_TO_LOOP_ERR_HANDLER (4); \ 863 } \ 864 } \ 865 } 866 #define LOOP_NEED_FLAGS 867 #define EXTRA_LOOP_DECLS , int *statep 868 #include <iconv/loop.c> 869 870 871 /* Now define the toplevel functions. */ 872 #include <iconv/skeleton.c> 873