1 /*- 2 * Copyright 2013 Garrett D'Amore <garrett@damore.org> 3 * Copyright 2011 Nexenta Systems, Inc. All rights reserved. 4 * Copyright (c) 2002-2004 Tim J. Robbins. All rights reserved. 5 * Copyright (c) 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * Paul Borman at Krystal Technologies. 10 * 11 * Copyright (c) 2011 The FreeBSD Foundation 12 * All rights reserved. 13 * Portions of this software were developed by David Chisnall 14 * under sponsorship from the FreeBSD Foundation. 15 * 16 * Redistribution and use in source and binary forms, with or without 17 * modification, are permitted provided that the following conditions 18 * are met: 19 * 1. Redistributions of source code must retain the above copyright 20 * notice, this list of conditions and the following disclaimer. 21 * 2. Redistributions in binary form must reproduce the above copyright 22 * notice, this list of conditions and the following disclaimer in the 23 * documentation and/or other materials provided with the distribution. 24 * 3. Neither the name of the University nor the names of its contributors 25 * may be used to endorse or promote products derived from this software 26 * without specific prior written permission. 27 * 28 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 29 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 30 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 31 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 32 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 33 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 34 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 35 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 37 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 38 * SUCH DAMAGE. 39 */ 40 41 #if defined(LIBC_SCCS) && !defined(lint) 42 static char sccsid[] = "@(#)euc.c 8.1 (Berkeley) 6/4/93"; 43 #endif /* LIBC_SCCS and not lint */ 44 #include <sys/param.h> 45 __FBSDID("$FreeBSD$"); 46 47 #include <errno.h> 48 #include <limits.h> 49 #include <runetype.h> 50 #include <stdlib.h> 51 #include <string.h> 52 #include <wchar.h> 53 #include "mblocal.h" 54 55 extern int __mb_sb_limit; 56 57 static size_t _EUC_mbrtowc_impl(wchar_t * __restrict, const char * __restrict, 58 size_t, mbstate_t * __restrict, uint8_t, uint8_t, uint8_t, uint8_t); 59 static size_t _EUC_wcrtomb_impl(char * __restrict, wchar_t, 60 mbstate_t * __restrict, uint8_t, uint8_t, uint8_t, uint8_t); 61 62 static size_t _EUC_CN_mbrtowc(wchar_t * __restrict, const char * __restrict, 63 size_t, mbstate_t * __restrict); 64 static size_t _EUC_JP_mbrtowc(wchar_t * __restrict, const char * __restrict, 65 size_t, mbstate_t * __restrict); 66 static size_t _EUC_KR_mbrtowc(wchar_t * __restrict, const char * __restrict, 67 size_t, mbstate_t * __restrict); 68 static size_t _EUC_TW_mbrtowc(wchar_t * __restrict, const char * __restrict, 69 size_t, mbstate_t * __restrict); 70 71 static size_t _EUC_CN_wcrtomb(char * __restrict, wchar_t, 72 mbstate_t * __restrict); 73 static size_t _EUC_JP_wcrtomb(char * __restrict, wchar_t, 74 mbstate_t * __restrict); 75 static size_t _EUC_KR_wcrtomb(char * __restrict, wchar_t, 76 mbstate_t * __restrict); 77 static size_t _EUC_TW_wcrtomb(char * __restrict, wchar_t, 78 mbstate_t * __restrict); 79 80 static size_t _EUC_CN_mbsnrtowcs(wchar_t * __restrict, 81 const char ** __restrict, size_t, size_t, 82 mbstate_t * __restrict); 83 static size_t _EUC_JP_mbsnrtowcs(wchar_t * __restrict, 84 const char ** __restrict, size_t, size_t, 85 mbstate_t * __restrict); 86 static size_t _EUC_KR_mbsnrtowcs(wchar_t * __restrict, 87 const char ** __restrict, size_t, size_t, 88 mbstate_t * __restrict); 89 static size_t _EUC_TW_mbsnrtowcs(wchar_t * __restrict, 90 const char ** __restrict, size_t, size_t, 91 mbstate_t * __restrict); 92 93 static size_t _EUC_CN_wcsnrtombs(char * __restrict, 94 const wchar_t ** __restrict, size_t, size_t, 95 mbstate_t * __restrict); 96 static size_t _EUC_JP_wcsnrtombs(char * __restrict, 97 const wchar_t ** __restrict, size_t, size_t, 98 mbstate_t * __restrict); 99 static size_t _EUC_KR_wcsnrtombs(char * __restrict, 100 const wchar_t ** __restrict, size_t, size_t, 101 mbstate_t * __restrict); 102 static size_t _EUC_TW_wcsnrtombs(char * __restrict, 103 const wchar_t ** __restrict, size_t, size_t, 104 mbstate_t * __restrict); 105 106 static int _EUC_mbsinit(const mbstate_t *); 107 108 typedef struct { 109 wchar_t ch; 110 int set; 111 int want; 112 } _EucState; 113 114 static int 115 _EUC_mbsinit(const mbstate_t *ps) 116 { 117 118 return (ps == NULL || ((const _EucState *)ps)->want == 0); 119 } 120 121 /* 122 * EUC-CN uses CS0, CS1 and CS2 (4 bytes). 123 */ 124 int 125 _EUC_CN_init(struct xlocale_ctype *l, _RuneLocale *rl) 126 { 127 l->__mbrtowc = _EUC_CN_mbrtowc; 128 l->__wcrtomb = _EUC_CN_wcrtomb; 129 l->__mbsnrtowcs = _EUC_CN_mbsnrtowcs; 130 l->__wcsnrtombs = _EUC_CN_wcsnrtombs; 131 l->__mbsinit = _EUC_mbsinit; 132 133 l->runes = rl; 134 l->__mb_cur_max = 4; 135 l->__mb_sb_limit = 256; 136 return (0); 137 } 138 139 static size_t 140 _EUC_CN_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, 141 size_t n, mbstate_t * __restrict ps) 142 { 143 return (_EUC_mbrtowc_impl(pwc, s, n, ps, SS2, 4, 0, 0)); 144 } 145 146 static size_t 147 _EUC_CN_mbsnrtowcs(wchar_t * __restrict dst, 148 const char ** __restrict src, 149 size_t nms, size_t len, mbstate_t * __restrict ps) 150 { 151 return (__mbsnrtowcs_std(dst, src, nms, len, ps, _EUC_CN_mbrtowc)); 152 } 153 154 static size_t 155 _EUC_CN_wcrtomb(char * __restrict s, wchar_t wc, 156 mbstate_t * __restrict ps) 157 { 158 return (_EUC_wcrtomb_impl(s, wc, ps, SS2, 4, 0, 0)); 159 } 160 161 static size_t 162 _EUC_CN_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src, 163 size_t nwc, size_t len, mbstate_t * __restrict ps) 164 { 165 return (__wcsnrtombs_std(dst, src, nwc, len, ps, _EUC_CN_wcrtomb)); 166 } 167 168 /* 169 * EUC-KR uses only CS0 and CS1. 170 */ 171 int 172 _EUC_KR_init(struct xlocale_ctype *l, _RuneLocale *rl) 173 { 174 l->__mbrtowc = _EUC_KR_mbrtowc; 175 l->__wcrtomb = _EUC_KR_wcrtomb; 176 l->__mbsnrtowcs = _EUC_KR_mbsnrtowcs; 177 l->__wcsnrtombs = _EUC_KR_wcsnrtombs; 178 l->__mbsinit = _EUC_mbsinit; 179 180 l->runes = rl; 181 l->__mb_cur_max = 2; 182 l->__mb_sb_limit = 128; 183 return (0); 184 } 185 186 static size_t 187 _EUC_KR_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, 188 size_t n, mbstate_t * __restrict ps) 189 { 190 return (_EUC_mbrtowc_impl(pwc, s, n, ps, 0, 0, 0, 0)); 191 } 192 193 static size_t 194 _EUC_KR_mbsnrtowcs(wchar_t * __restrict dst, 195 const char ** __restrict src, 196 size_t nms, size_t len, mbstate_t * __restrict ps) 197 { 198 return (__mbsnrtowcs_std(dst, src, nms, len, ps, _EUC_KR_mbrtowc)); 199 } 200 201 static size_t 202 _EUC_KR_wcrtomb(char * __restrict s, wchar_t wc, 203 mbstate_t * __restrict ps) 204 { 205 return (_EUC_wcrtomb_impl(s, wc, ps, 0, 0, 0, 0)); 206 } 207 208 static size_t 209 _EUC_KR_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src, 210 size_t nwc, size_t len, mbstate_t * __restrict ps) 211 { 212 return (__wcsnrtombs_std(dst, src, nwc, len, ps, _EUC_KR_wcrtomb)); 213 } 214 215 /* 216 * EUC-JP uses CS0, CS1, CS2, and CS3. 217 */ 218 int 219 _EUC_JP_init(struct xlocale_ctype *l, _RuneLocale *rl) 220 { 221 l->__mbrtowc = _EUC_JP_mbrtowc; 222 l->__wcrtomb = _EUC_JP_wcrtomb; 223 l->__mbsnrtowcs = _EUC_JP_mbsnrtowcs; 224 l->__wcsnrtombs = _EUC_JP_wcsnrtombs; 225 l->__mbsinit = _EUC_mbsinit; 226 227 l->runes = rl; 228 l->__mb_cur_max = 3; 229 l->__mb_sb_limit = 196; 230 return (0); 231 } 232 233 static size_t 234 _EUC_JP_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, 235 size_t n, mbstate_t * __restrict ps) 236 { 237 return (_EUC_mbrtowc_impl(pwc, s, n, ps, SS2, 2, SS3, 3)); 238 } 239 240 static size_t 241 _EUC_JP_mbsnrtowcs(wchar_t * __restrict dst, 242 const char ** __restrict src, 243 size_t nms, size_t len, mbstate_t * __restrict ps) 244 { 245 return (__mbsnrtowcs_std(dst, src, nms, len, ps, _EUC_JP_mbrtowc)); 246 } 247 248 static size_t 249 _EUC_JP_wcrtomb(char * __restrict s, wchar_t wc, 250 mbstate_t * __restrict ps) 251 { 252 return (_EUC_wcrtomb_impl(s, wc, ps, SS2, 2, SS3, 3)); 253 } 254 255 static size_t 256 _EUC_JP_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src, 257 size_t nwc, size_t len, mbstate_t * __restrict ps) 258 { 259 return (__wcsnrtombs_std(dst, src, nwc, len, ps, _EUC_JP_wcrtomb)); 260 } 261 262 /* 263 * EUC-TW uses CS0, CS1, and CS2. 264 */ 265 int 266 _EUC_TW_init(struct xlocale_ctype *l, _RuneLocale *rl) 267 { 268 l->__mbrtowc = _EUC_TW_mbrtowc; 269 l->__wcrtomb = _EUC_TW_wcrtomb; 270 l->__mbsnrtowcs = _EUC_TW_mbsnrtowcs; 271 l->__wcsnrtombs = _EUC_TW_wcsnrtombs; 272 l->__mbsinit = _EUC_mbsinit; 273 274 l->runes = rl; 275 l->__mb_cur_max = 4; 276 l->__mb_sb_limit = 256; 277 return (0); 278 } 279 280 static size_t 281 _EUC_TW_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, 282 size_t n, mbstate_t * __restrict ps) 283 { 284 return (_EUC_mbrtowc_impl(pwc, s, n, ps, SS2, 4, 0, 0)); 285 } 286 287 static size_t 288 _EUC_TW_mbsnrtowcs(wchar_t * __restrict dst, 289 const char ** __restrict src, 290 size_t nms, size_t len, mbstate_t * __restrict ps) 291 { 292 return (__mbsnrtowcs_std(dst, src, nms, len, ps, _EUC_TW_mbrtowc)); 293 } 294 295 static size_t 296 _EUC_TW_wcrtomb(char * __restrict s, wchar_t wc, 297 mbstate_t * __restrict ps) 298 { 299 return (_EUC_wcrtomb_impl(s, wc, ps, SS2, 4, 0, 0)); 300 } 301 302 static size_t 303 _EUC_TW_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src, 304 size_t nwc, size_t len, mbstate_t * __restrict ps) 305 { 306 return (__wcsnrtombs_std(dst, src, nwc, len, ps, _EUC_TW_wcrtomb)); 307 } 308 309 /* 310 * Common EUC code. 311 */ 312 313 static size_t 314 _EUC_mbrtowc_impl(wchar_t * __restrict pwc, const char * __restrict s, 315 size_t n, mbstate_t * __restrict ps, 316 uint8_t cs2, uint8_t cs2width, uint8_t cs3, uint8_t cs3width) 317 { 318 _EucState *es; 319 int i, want; 320 wchar_t wc = 0; 321 unsigned char ch, chs; 322 323 es = (_EucState *)ps; 324 325 if (es->want < 0 || es->want > MB_CUR_MAX) { 326 errno = EINVAL; 327 return ((size_t)-1); 328 } 329 330 if (s == NULL) { 331 s = ""; 332 n = 1; 333 pwc = NULL; 334 } 335 336 if (n == 0) 337 /* Incomplete multibyte sequence */ 338 return ((size_t)-2); 339 340 if (es->want == 0) { 341 /* Fast path for plain ASCII (CS0) */ 342 if (((ch = (unsigned char)*s) & 0x80) == 0) { 343 if (pwc != NULL) 344 *pwc = ch; 345 return (ch != '\0' ? 1 : 0); 346 } 347 348 if (ch >= 0xa1) { 349 /* CS1 */ 350 want = 2; 351 } else if (ch == cs2) { 352 want = cs2width; 353 } else if (ch == cs3) { 354 want = cs3width; 355 } else { 356 errno = EILSEQ; 357 return ((size_t)-1); 358 } 359 360 361 es->want = want; 362 es->ch = 0; 363 } else { 364 want = es->want; 365 wc = es->ch; 366 } 367 368 for (i = 0; i < MIN(want, n); i++) { 369 wc <<= 8; 370 chs = *s; 371 wc |= chs; 372 s++; 373 } 374 if (i < want) { 375 /* Incomplete multibyte sequence */ 376 es->want = want - i; 377 es->ch = wc; 378 errno = EILSEQ; 379 return ((size_t)-2); 380 } 381 if (pwc != NULL) 382 *pwc = wc; 383 es->want = 0; 384 return (wc == L'\0' ? 0 : want); 385 } 386 387 static size_t 388 _EUC_wcrtomb_impl(char * __restrict s, wchar_t wc, 389 mbstate_t * __restrict ps, 390 uint8_t cs2, uint8_t cs2width, uint8_t cs3, uint8_t cs3width) 391 { 392 _EucState *es; 393 int i, len; 394 wchar_t nm; 395 396 es = (_EucState *)ps; 397 398 if (es->want != 0) { 399 errno = EINVAL; 400 return ((size_t)-1); 401 } 402 403 if (s == NULL) 404 /* Reset to initial shift state (no-op) */ 405 return (1); 406 407 if ((wc & ~0x7f) == 0) { 408 /* Fast path for plain ASCII (CS0) */ 409 *s = (char)wc; 410 return (1); 411 } 412 413 /* Determine the "length" */ 414 if ((unsigned)wc > 0xffffff) { 415 len = 4; 416 } else if ((unsigned)wc > 0xffff) { 417 len = 3; 418 } else if ((unsigned)wc > 0xff) { 419 len = 2; 420 } else { 421 len = 1; 422 } 423 424 if (len > MB_CUR_MAX) { 425 errno = EILSEQ; 426 return ((size_t)-1); 427 } 428 429 /* This first check excludes CS1, which is implicitly valid. */ 430 if ((wc < 0xa100) || (wc > 0xffff)) { 431 /* Check for valid CS2 or CS3 */ 432 nm = (wc >> ((len - 1) * 8)); 433 if (nm == cs2) { 434 if (len != cs2width) { 435 errno = EILSEQ; 436 return ((size_t)-1); 437 } 438 } else if (nm == cs3) { 439 if (len != cs3width) { 440 errno = EILSEQ; 441 return ((size_t)-1); 442 } 443 } else { 444 errno = EILSEQ; 445 return ((size_t)-1); 446 } 447 } 448 449 /* Stash the bytes, least significant last */ 450 for (i = len - 1; i >= 0; i--) { 451 s[i] = (wc & 0xff); 452 wc >>= 8; 453 } 454 return (len); 455 } 456