1 /* 2 * Copyright (C) 1999-2001, 2005 Free Software Foundation, Inc. 3 * This file is part of the GNU LIBICONV Library. 4 * 5 * The GNU LIBICONV Library is free software; you can redistribute it 6 * and/or modify it under the terms of the GNU Library General Public 7 * License as published by the Free Software Foundation; either version 2 8 * of the License, or (at your option) any later version. 9 * 10 * The GNU LIBICONV Library is distributed in the hope that it will be 11 * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of 12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 * Library General Public License for more details. 14 * 15 * You should have received a copy of the GNU Library General Public 16 * License along with the GNU LIBICONV Library; see the file COPYING.LIB. 17 * If not, write to the Free Software Foundation, Inc., 51 Franklin Street, 18 * Fifth Floor, Boston, MA 02110-1301, USA. 19 */ 20 21 /* 22 * GB18030 23 */ 24 25 /* 26 * GB18030, as specified in the GB18030 standard, is an extension of GBK. 27 * 28 * In what follows, page numbers refer to the GB18030 standard (second 29 * printing). 30 * 31 * 32 * It consists of the following parts: 33 * 34 * One-byte range: 35 * ASCII p. 2 0x{00..7F} 36 * 37 * Two-byte range: 38 * GBK part 1 p. 10..12 0x{A1..A9}{A1..FE} 39 * GBK part 2 p. 13..36 0x{B0..F7}{A1..FE} 40 * GBK part 3 p. 37..52 0x{81..A0}{40..7E,80..FE} 41 * GBK part 4 p. 53..81 0x{AA..FE}{40..7E,80..A0} 42 * GBK part 5 p. 82 0x{A8..A9}{40..7E,80..A0} 43 * UDA part 1 p. 83..84 0x{AA..AF}{A1..FE} U+E000..U+E233 44 * UDA part 2 p. 85..87 0x{F8..FE}{A1..FE} U+E234..U+E4C5 45 * UDA part 3 p. 88..90 0x{A1..A7}{40..7E,80..A0} U+E4C6..U+E765 46 * 47 * Four-byte range: 48 * BMP rest p. 94..283 0x{81..84}{30..39}{81..FE}{30..39} 49 * rest of U+0080..U+FFFF 50 * Planes 1-16 p. 5 0x{90..FE}{30..39}{81..FE}{30..39} 51 * U+10000..U+10FFFF 52 * 53 * To GBK part 1 were added: 54 * 1. 0xA2E3, 0xA8BF. 55 * 2. Characters mapped to the Unicode PUA 56 * 0xA2AB..0xA2B0 U+E766..U+E76B 57 * 0xA2E4 U+E76D 58 * 0xA2EF..0xA2F0 U+E76E..U+E76F 59 * 0xA2FD..0xA2FE U+E770..U+E771 60 * 0xA4F4..0xA4FE U+E772..U+E77C 61 * 0xA5F7..0xA5FE U+E77D..U+E784 62 * 0xA6B9..0xA6C0 U+E785..U+E78C 63 * 0xA6D9..0xA6DF [glyphs here!!] U+E78D..U+E793 64 * 0xA6EC..0xA6ED [glyphs here!!] U+E794..U+E795 65 * 0xA6F3 [glyphs here!!] U+E796 66 * 0xA6F6..0xA6FE U+E797..U+E79F 67 * 0xA7C2..0xA7D0 U+E7A0..U+E7AE 68 * 0xA7F2..0xA7FE U+E7AF..U+E7BB 69 * 0xA8BC [glyphs here!!] U+E7C7 70 * 0xA8C1..0xA8C4 U+E7C9..U+E7CC 71 * 0xA8EA..0xA8FE U+E7CD..U+E7E1 72 * 0xA9A1..0xA9A3 U+E7FE..U+E800 73 * 0xA9F0..0xA9FE U+E801..U+E80F 74 * 75 * To GBK part 2 were added: 76 * 3. Characters mapped to the Unicode PUA 77 * 0xD7FA..0xD7FE U+E810..0xE814 78 * 79 * To GBK part 3 nothing was added. 80 * 81 * To GBK part 4 were added: 82 * 4. 0xFE{50,54..58,5A..60,62..65,68..6B,6E..75,77..7D,80..8F,92..9F}. 83 * 5. Characters mapped to the Unicode PUA 84 * 0xFE51..0xFE53 [glyphs here!!] U+E816..U+E818 85 * 0xFE59 [glyphs here!!] U+E81E 86 * 0xFE61 [glyphs here!!] U+E826 87 * 0xFE66..0xFE67 [glyphs here!!] U+E82B..U+E82C 88 * 0xFE6C..0xFE6D [glyphs here!!] U+E831..U+E832 89 * 0xFE76 [glyphs here!!] U+E83B 90 * 0xFE7E [glyphs here!!] U+E843 91 * 0xFE90..0xFE91 [glyphs here!!] U+E854..U+E855 92 * 0xFEA0 [glyphs here!!] U+E864 93 * 94 * To GBK part 5 were added: 95 * 6. 0xA98A..0xA995. 96 * 7. Characters mapped to the Unicode PUA 97 * 0xA896..0xA8A0 U+E7BC..U+E7C6 98 * 0xA958 U+E7E2 99 * 0xA95B U+E7E3 100 * 0xA95D..0xA95F U+E7E4..U+E7E6 101 * 0xA997..0xA9A0 U+E7F4..U+E7FD 102 * 103 * UDA part 1 contains the user-defined characters, mapped to the Unicode PUA 104 * U+E000..U+E233 in ascending order. 105 * 106 * UDA part 2 contains the user-defined characters, mapped to the Unicode PUA 107 * U+E234..U+E4C5 in ascending order. 108 * 109 * UDA part 3 contains the user-defined characters, mapped to the Unicode PUA 110 * U+E4C6..U+E765 in ascending order. 111 * 112 * The four-byte range 0x{81..84}{30..39}{81..FE}{30..39} 113 * contains the rest of the Unicode BMP in ascending order. 114 * Start: 0x81308130 = 0x0080 115 * End: 0x8431A439 = 0xFFFF 116 * 117 * The four-byte range 0x{90..E3}{30..39}{81..FE}{30..39} 118 * contains the remaining 16 Unicode planes in Unicode order. 119 * Start: 0x90308130 = 0x010000 120 * End: 0xE3329A35 = 0x10FFFF 121 * 122 * 123 * Unassigned Unicode characters are mapped. For example, 124 * U+173F = 0x8134BF35 (p. 120) 125 * U+2EFF = 0x81398B31 (p. 148) 126 * U+FFFE = 0x8431A438 (p. 283) 127 * 128 * 129 * The Unicode PUA (U+E000..U+F8FF) is mapped as follows: 130 * p. 83..84 0x{AA..AF}{A1..FE} U+E000..U+E233 131 * p. 85..87 0x{F8..FE}{A1..FE} U+E234..U+E4C5 132 * p. 88..90 0x{A1..A7}{40..7E,80..A0} U+E4C6..U+E765 133 * p. 10 0xA2AB..0xA2B0 U+E766..U+E76B 134 * p. 255 0x8336C739 U+E76C 135 * p. 10 0xA2E4 U+E76D 136 * p. 10 0xA2EF..0xA2F0 U+E76E..U+E76F 137 * p. 10 0xA2FD..0xA2FE U+E770..U+E771 138 * p. 11 0xA4F4..0xA4FE U+E772..U+E77C 139 * p. 11 0xA5F7..0xA5FE U+E77D..U+E784 140 * p. 11 0xA6B9..0xA6C0 U+E785..U+E78C 141 * p. 11 0xA6D9..0xA6DF [glyphs here!!] U+E78D..U+E793 142 * p. 11 0xA6EC..0xA6ED [glyphs here!!] U+E794..U+E795 143 * p. 11 0xA6F3 [glyphs here!!] U+E796 144 * p. 11 0xA6F6..0xA6FE U+E797..U+E79F 145 * p. 12 0xA7C2..0xA7D0 U+E7A0..U+E7AE 146 * p. 12 0xA7F2..0xA7FE U+E7AF..U+E7BB 147 * p. 82 0xA896..0xA8A0 U+E7BC..U+E7C6 148 * p. 12 0xA8BC [glyphs here!!] U+E7C7 149 * p. 255 0x8336C830 U+E7C8 150 * p. 12 0xA8C1..0xA8C4 U+E7C9..U+E7CC 151 * p. 12 0xA8EA..0xA8FE U+E7CD..U+E7E1 152 * p. 82 0xA958 U+E7E2 153 * p. 82 0xA95B U+E7E3 154 * p. 82 0xA95D..0xA95F U+E7E4..U+E7E6 155 * p. 255 0x8336C831..0x8336C933 U+E7E7..U+E7F3 156 * p. 82 0xA997..0xA9A0 U+E7F4..U+E7FD 157 * p. 12 0xA9A1..0xA9A3 U+E7FE..U+E800 158 * p. 12 0xA9F0..0xA9FE U+E801..U+E80F 159 * p. 26 0xD7FA..0xD7FE U+E810..0xE814 160 * p. 255 0x8336C934 U+E815 161 * p. 81 0xFE51..0xFE53 [glyphs here!!] U+E816..U+E818 162 * p. 255 0x8336C935..0x8336C939 U+E819..U+E81D 163 * p. 81 0xFE59 [glyphs here!!] U+E81E 164 * p. 255 0x8336CA30..0x8336CA36 U+E81F..U+E825 165 * p. 81 0xFE61 [glyphs here!!] U+E826 166 * p. 255 0x8336CA37..0x8336CB30 U+E827..U+E82A 167 * p. 81 0xFE66..0xFE67 [glyphs here!!] U+E82B..U+E82C 168 * p. 255 0x8336CB31..0x8336CB34 U+E82D..U+E830 169 * p. 81 0xFE6C..0xFE6D [glyphs here!!] U+E831..U+E832 170 * p. 255 0x8336CB35..0x8336CC32 U+E833..U+E83A 171 * p. 81 0xFE76 [glyphs here!!] U+E83B 172 * p. 255 0x8336CC33..0x8336CC39 U+E83C..U+E842 173 * p. 81 0xFE7E [glyphs here!!] U+E843 174 * p. 255 0x8336CD30..0x8336CE35 U+E844..U+E853 175 * p. 81 0xFE90..0xFE91 [glyphs here!!] U+E854..U+E855 176 * p. 255 0x8336CE36..0x8336CF39 U+E856..U+E863 177 * p. 81 0xFEA0 [glyphs here!!] U+E864 178 * p. 255..276 0x8336D030..0x84308130 U+E865..U+F8FF 179 * 180 * 181 * The Unicode surrogate area (U+D800..U+DFFF) is not mapped. (p. 255) 182 * 183 */ 184 185 #include "gb18030ext.h" 186 #include "gb18030uni.h" 187 188 static int 189 gb18030_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n) 190 { 191 int ret; 192 193 /* Code set 0 (ASCII) */ 194 if (*s < 0x80) 195 return ascii_mbtowc(conv,pwc,s,n); 196 197 /* Code set 1 (GBK extended) */ 198 ret = gbk_mbtowc(conv,pwc,s,n); 199 if (ret != RET_ILSEQ) 200 return ret; 201 202 ret = gb18030ext_mbtowc(conv,pwc,s,n); 203 if (ret != RET_ILSEQ) 204 return ret; 205 206 /* Code set 2 (remainder of Unicode U+0000..U+FFFF), including 207 User-defined characters, two-byte part of range U+E766..U+E864 */ 208 ret = gb18030uni_mbtowc(conv,pwc,s,n); 209 if (ret != RET_ILSEQ) 210 return ret; 211 /* User-defined characters range U+E000..U+E765 */ 212 { 213 unsigned char c1 = s[0]; 214 if ((c1 >= 0xaa && c1 <= 0xaf) || (c1 >= 0xf8 && c1 <= 0xfe)) { 215 if (n >= 2) { 216 unsigned char c2 = s[1]; 217 if (c2 >= 0xa1 && c2 <= 0xfe) { 218 *pwc = 0xe000 + 94 * (c1 >= 0xf8 ? c1 - 0xf2 : c1 - 0xaa) + (c2 - 0xa1); 219 return 2; 220 } 221 } else 222 return RET_TOOFEW(0); 223 } else if (c1 >= 0xa1 && c1 <= 0xa7) { 224 if (n >= 2) { 225 unsigned char c2 = s[1]; 226 if (c2 >= 0x40 && c2 <= 0xa1 && c2 != 0x7f) { 227 *pwc = 0xe4c6 + 96 * (c1 - 0xa1) + c2 - (c2 >= 0x80 ? 0x41 : 0x40); 228 return 2; 229 } 230 } else 231 return RET_TOOFEW(0); 232 } 233 } 234 235 /* Code set 3 (Unicode U+10000..U+10FFFF) */ 236 { 237 unsigned char c1 = s[0]; 238 if (c1 >= 0x90 && c1 <= 0xe3) { 239 if (n >= 2) { 240 unsigned char c2 = s[1]; 241 if (c2 >= 0x30 && c2 <= 0x39) { 242 if (n >= 3) { 243 unsigned char c3 = s[2]; 244 if (c3 >= 0x81 && c3 <= 0xfe) { 245 if (n >= 4) { 246 unsigned char c4 = s[3]; 247 if (c4 >= 0x30 && c4 <= 0x39) { 248 unsigned int i = (((c1 - 0x90) * 10 + (c2 - 0x30)) * 126 + (c3 - 0x81)) * 10 + (c4 - 0x30); 249 if (i >= 0 && i < 0x100000) { 250 *pwc = (ucs4_t) (0x10000 + i); 251 return 4; 252 } 253 } 254 return RET_ILSEQ; 255 } 256 return RET_TOOFEW(0); 257 } 258 return RET_ILSEQ; 259 } 260 return RET_TOOFEW(0); 261 } 262 return RET_ILSEQ; 263 } 264 return RET_TOOFEW(0); 265 } 266 return RET_ILSEQ; 267 } 268 } 269 270 static const unsigned short gb18030_pua2charset[32*3] = { 271 /* Unicode range GB18030 range */ 272 0xe766, 0xe76b, 0xa2ab, /*.. 0xa2b0, */ 273 0xe76d, 0xe76d, 0xa2e4, 274 0xe76e, 0xe76f, 0xa2ef, /*.. 0xa2f0, */ 275 0xe770, 0xe771, 0xa2fd, /*.. 0xa2fe, */ 276 0xe772, 0xe77c, 0xa4f4, /*.. 0xa4fe, */ 277 0xe77d, 0xe784, 0xa5f7, /*.. 0xa5fe, */ 278 0xe785, 0xe78c, 0xa6b9, /*.. 0xa6c0, */ 279 0xe78d, 0xe793, 0xa6d9, /*.. 0xa6df, */ 280 0xe794, 0xe795, 0xa6ec, /*.. 0xa6ed, */ 281 0xe796, 0xe796, 0xa6f3, 282 0xe797, 0xe79f, 0xa6f6, /*.. 0xa6fe, */ 283 0xe7a0, 0xe7ae, 0xa7c2, /*.. 0xa7d0, */ 284 0xe7af, 0xe7bb, 0xa7f2, /*.. 0xa7fe, */ 285 0xe7bc, 0xe7c6, 0xa896, /*.. 0xa8a0, */ 286 0xe7c7, 0xe7c7, 0xa8bc, 287 0xe7c9, 0xe7cc, 0xa8c1, /*.. 0xa8c4, */ 288 0xe7cd, 0xe7e1, 0xa8ea, /*.. 0xa8fe, */ 289 0xe7e2, 0xe7e2, 0xa958, 290 0xe7e3, 0xe7e3, 0xa95b, 291 0xe7e4, 0xe7e6, 0xa95d, /*.. 0xa95f, */ 292 0xe7f4, 0xe800, 0xa997, /*.. 0xa9a3, */ 293 0xe801, 0xe80f, 0xa9f0, /*.. 0xa9fe, */ 294 0xe810, 0xe814, 0xd7fa, /*.. 0xd7fe, */ 295 0xe816, 0xe818, 0xfe51, /*.. 0xfe53, */ 296 0xe81e, 0xe81e, 0xfe59, 297 0xe826, 0xe826, 0xfe61, 298 0xe82b, 0xe82c, 0xfe66, /*.. 0xfe67, */ 299 0xe831, 0xe832, 0xfe6c, /*.. 0xfe6d, */ 300 0xe83b, 0xe83b, 0xfe76, 301 0xe843, 0xe843, 0xfe7e, 302 0xe854, 0xe855, 0xfe90, /*.. 0xfe91, */ 303 0xe864, 0xe864, 0xfea0, 304 }; 305 306 static int 307 gb18030_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) 308 { 309 int ret; 310 311 /* Code set 0 (ASCII) */ 312 ret = ascii_wctomb(conv,r,wc,n); 313 if (ret != RET_ILUNI) 314 return ret; 315 316 /* Code set 1 (GBK extended) */ 317 ret = gbk_wctomb(conv,r,wc,n); 318 if (ret != RET_ILUNI) 319 return ret; 320 321 ret = gb18030ext_wctomb(conv,r,wc,n); 322 if (ret != RET_ILUNI) 323 return ret; 324 325 /* Code set 2 (remainder of Unicode U+0000..U+FFFF) */ 326 if (wc >= 0xe000 && wc <= 0xe864) { 327 if (n >= 2) { 328 if (wc < 0xe766) { 329 /* User-defined characters range U+E000..U+E765 */ 330 if (wc < 0xe4c6) { 331 unsigned int i = wc - 0xe000; 332 r[1] = (i % 94) + 0xa1; i = i / 94; 333 r[0] = (i < 6 ? i + 0xaa : i + 0xf2); 334 return 2; 335 } else { 336 unsigned int i = wc - 0xe4c6; 337 r[0] = (i / 96) + 0xa1; i = i % 96; 338 r[1] = i + (i >= 0x3f ? 0x41 : 0x40); 339 return 2; 340 } 341 } else { 342 /* User-defined characters, two-byte part of range U+E766..U+E864 */ 343 unsigned int k1 = 0; 344 unsigned int k2 = 32; 345 /* Invariant: We know that if wc occurs in Unicode interval in 346 gb18030_pua2charset, it does so at a k with k1 <= k < k2. */ 347 while (k1 < k2) { 348 unsigned int k = (k1 + k2) / 2; 349 if (wc < gb18030_pua2charset[k*3+0]) 350 k2 = k; 351 else if (wc > gb18030_pua2charset[k*3+1]) 352 k1 = k + 1; 353 else { 354 unsigned short c = 355 gb18030_pua2charset[k*3+2] + (wc - gb18030_pua2charset[k*3+0]); 356 r[0] = (c >> 8); 357 r[1] = (c & 0xff); 358 return 2; 359 } 360 } 361 } 362 } else 363 return RET_TOOSMALL; 364 } 365 ret = gb18030uni_wctomb(conv,r,wc,n); 366 if (ret != RET_ILUNI) 367 return ret; 368 369 /* Code set 3 (Unicode U+10000..U+10FFFF) */ 370 if (n >= 4) { 371 if (wc >= 0x10000 && wc < 0x110000) { 372 unsigned int i = wc - 0x10000; 373 r[3] = (i % 10) + 0x30; i = i / 10; 374 r[2] = (i % 126) + 0x81; i = i / 126; 375 r[1] = (i % 10) + 0x30; i = i / 10; 376 r[0] = i + 0x90; 377 return 4; 378 } 379 return RET_ILUNI; 380 } 381 return RET_TOOSMALL; 382 } 383