1 /* $NetBSD: citrus_utf8.c,v 1.16 2007/03/06 16:13:58 tnozaki Exp $ */ 2 3 /*- 4 * Copyright (c)2002 Citrus Project, 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 26 * SUCH DAMAGE. 27 */ 28 29 /*- 30 * Copyright (c) 1993 31 * The Regents of the University of California. All rights reserved. 32 * 33 * This code is derived from software contributed to Berkeley by 34 * Paul Borman at Krystal Technologies. 35 * 36 * Redistribution and use in source and binary forms, with or without 37 * modification, are permitted provided that the following conditions 38 * are met: 39 * 1. Redistributions of source code must retain the above copyright 40 * notice, this list of conditions and the following disclaimer. 41 * 2. Redistributions in binary form must reproduce the above copyright 42 * notice, this list of conditions and the following disclaimer in the 43 * documentation and/or other materials provided with the distribution. 44 * 3. Neither the name of the University nor the names of its contributors 45 * may be used to endorse or promote products derived from this software 46 * without specific prior written permission. 47 * 48 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 49 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 50 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 51 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 52 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 53 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 54 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 55 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 56 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 57 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 58 * SUCH DAMAGE. 59 */ 60 61 #include <sys/cdefs.h> 62 #if defined(LIBC_SCCS) && !defined(lint) 63 __RCSID("$NetBSD: citrus_utf8.c,v 1.16 2007/03/06 16:13:58 tnozaki Exp $"); 64 #endif /* LIBC_SCCS and not lint */ 65 66 #include <assert.h> 67 #include <errno.h> 68 #include <string.h> 69 #include <stdio.h> 70 #include <stdlib.h> 71 #include <stddef.h> 72 #include <locale.h> 73 #include <wchar.h> 74 #include <sys/types.h> 75 #include <limits.h> 76 77 #include "citrus_namespace.h" 78 #include "citrus_types.h" 79 #include "citrus_module.h" 80 #include "citrus_ctype.h" 81 #include "citrus_stdenc.h" 82 #include "citrus_utf8.h" 83 84 85 /* ---------------------------------------------------------------------- 86 * private stuffs used by templates 87 */ 88 89 static int _UTF8_count_array[256]; 90 static int const *_UTF8_count = NULL; 91 92 static const u_int32_t _UTF8_range[] = { 93 0, /*dummy*/ 94 0x00000000, 0x00000080, 0x00000800, 0x00010000, 95 0x00200000, 0x04000000, 0x80000000, 96 }; 97 98 typedef struct { 99 char ch[6]; 100 int chlen; 101 } _UTF8State; 102 103 typedef struct { 104 } _UTF8EncodingInfo; 105 106 typedef struct { 107 _UTF8EncodingInfo ei; 108 struct { 109 /* for future multi-locale facility */ 110 _UTF8State s_mblen; 111 _UTF8State s_mbrlen; 112 _UTF8State s_mbrtowc; 113 _UTF8State s_mbtowc; 114 _UTF8State s_mbsrtowcs; 115 _UTF8State s_wcrtomb; 116 _UTF8State s_wcsrtombs; 117 _UTF8State s_wctomb; 118 } states; 119 } _UTF8CTypeInfo; 120 121 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei) 122 #define _CEI_TO_STATE(_ei_, _func_) (_ei_)->states.s_##_func_ 123 124 #define _FUNCNAME(m) _citrus_UTF8_##m 125 #define _ENCODING_INFO _UTF8EncodingInfo 126 #define _CTYPE_INFO _UTF8CTypeInfo 127 #define _ENCODING_STATE _UTF8State 128 #define _ENCODING_MB_CUR_MAX(_ei_) 6 129 #define _ENCODING_IS_STATE_DEPENDENT 0 130 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0 131 132 133 static __inline void 134 _UTF8_init_count(void) 135 { 136 int i; 137 if (!_UTF8_count) { 138 memset(_UTF8_count_array, 0, sizeof(_UTF8_count_array)); 139 for (i = 0; i <= 0x7f; i++) 140 _UTF8_count_array[i] = 1; 141 for (i = 0xc0; i <= 0xdf; i++) 142 _UTF8_count_array[i] = 2; 143 for (i = 0xe0; i <= 0xef; i++) 144 _UTF8_count_array[i] = 3; 145 for (i = 0xf0; i <= 0xf7; i++) 146 _UTF8_count_array[i] = 4; 147 for (i = 0xf8; i <= 0xfb; i++) 148 _UTF8_count_array[i] = 5; 149 for (i = 0xfc; i <= 0xfd; i++) 150 _UTF8_count_array[i] = 6; 151 _UTF8_count = _UTF8_count_array; 152 } 153 } 154 155 static int 156 _UTF8_findlen(wchar_t v) 157 { 158 int i; 159 u_int32_t c; 160 161 c = (u_int32_t)v; /*XXX*/ 162 for (i = 1; i < sizeof(_UTF8_range) / sizeof(_UTF8_range[0]) - 1; i++) 163 if (c >= _UTF8_range[i] && c < _UTF8_range[i + 1]) 164 return i; 165 166 return -1; /*out of range*/ 167 } 168 169 static __inline int 170 _UTF8_surrogate(wchar_t wc) 171 { 172 return wc >= 0xd800 && wc <= 0xdfff; 173 } 174 175 static __inline void 176 /*ARGSUSED*/ 177 _citrus_UTF8_init_state(_UTF8EncodingInfo *ei, _UTF8State *s) 178 { 179 s->chlen = 0; 180 } 181 182 static __inline void 183 /*ARGSUSED*/ 184 _citrus_UTF8_pack_state(_UTF8EncodingInfo *ei, void *pspriv, 185 const _UTF8State *s) 186 { 187 memcpy(pspriv, (const void *)s, sizeof(*s)); 188 } 189 190 static __inline void 191 /*ARGSUSED*/ 192 _citrus_UTF8_unpack_state(_UTF8EncodingInfo *ei, _UTF8State *s, 193 const void *pspriv) 194 { 195 memcpy((void *)s, pspriv, sizeof(*s)); 196 } 197 198 static int 199 _citrus_UTF8_mbrtowc_priv(_UTF8EncodingInfo *ei, wchar_t *pwc, const char **s, 200 size_t n, _UTF8State *psenc, size_t *nresult) 201 { 202 wchar_t wchar; 203 const char *s0; 204 int c; 205 int i; 206 207 _DIAGASSERT(nresult != 0); 208 _DIAGASSERT(s != NULL); 209 _DIAGASSERT(psenc != NULL); 210 211 s0 = *s; 212 213 if (s0 == NULL) { 214 _citrus_UTF8_init_state(ei, psenc); 215 *nresult = 0; /* state independent */ 216 return 0; 217 } 218 219 /* make sure we have the first byte in the buffer */ 220 if (psenc->chlen == 0) { 221 if (n-- < 1) 222 goto restart; 223 psenc->ch[psenc->chlen++] = *s0++; 224 } 225 226 c = _UTF8_count[psenc->ch[0] & 0xff]; 227 if (c < 1 || c < psenc->chlen) 228 goto ilseq; 229 230 if (c == 1) 231 wchar = psenc->ch[0] & 0xff; 232 else { 233 while (psenc->chlen < c) { 234 if (n-- < 1) 235 goto restart; 236 psenc->ch[psenc->chlen++] = *s0++; 237 } 238 wchar = psenc->ch[0] & (0x7f >> c); 239 for (i = 1; i < c; i++) { 240 if ((psenc->ch[i] & 0xc0) != 0x80) 241 goto ilseq; 242 wchar <<= 6; 243 wchar |= (psenc->ch[i] & 0x3f); 244 } 245 if (_UTF8_surrogate(wchar) || _UTF8_findlen(wchar) != c) 246 goto ilseq; 247 } 248 if (pwc != NULL) 249 *pwc = wchar; 250 *nresult = (wchar == 0) ? 0 : s0 - *s; 251 *s = s0; 252 psenc->chlen = 0; 253 254 return 0; 255 256 ilseq: 257 *nresult = (size_t)-1; 258 return EILSEQ; 259 260 restart: 261 *s = s0; 262 *nresult = (size_t)-2; 263 return 0; 264 } 265 266 static int 267 _citrus_UTF8_wcrtomb_priv(_UTF8EncodingInfo *ei, char *s, size_t n, wchar_t wc, 268 _UTF8State *psenc, size_t *nresult) 269 { 270 int cnt, i, ret; 271 wchar_t c; 272 273 _DIAGASSERT(nresult != 0); 274 _DIAGASSERT(s != NULL); 275 276 if (_UTF8_surrogate(wc)) { 277 ret = EILSEQ; 278 goto err; 279 } 280 cnt = _UTF8_findlen(wc); 281 if (cnt <= 0 || cnt > 6) { 282 /* invalid UCS4 value */ 283 ret = EILSEQ; 284 goto err; 285 } 286 if (n < cnt) { 287 /* bound check failure */ 288 ret = E2BIG; 289 goto err; 290 } 291 292 c = wc; 293 if (s) { 294 for (i = cnt - 1; i > 0; i--) { 295 s[i] = 0x80 | (c & 0x3f); 296 c >>= 6; 297 } 298 s[0] = c; 299 if (cnt == 1) 300 s[0] &= 0x7f; 301 else { 302 s[0] &= (0x7f >> cnt); 303 s[0] |= ((0xff00 >> cnt) & 0xff); 304 } 305 } 306 307 *nresult = (size_t)cnt; 308 return 0; 309 310 err: 311 *nresult = (size_t)-1; 312 return ret; 313 } 314 315 static __inline int 316 /*ARGSUSED*/ 317 _citrus_UTF8_stdenc_wctocs(_UTF8EncodingInfo * __restrict ei, 318 _csid_t * __restrict csid, 319 _index_t * __restrict idx, 320 wchar_t wc) 321 { 322 323 _DIAGASSERT(csid != NULL && idx != NULL); 324 325 *csid = 0; 326 *idx = (_citrus_index_t)wc; 327 328 return (0); 329 } 330 331 static __inline int 332 /*ARGSUSED*/ 333 _citrus_UTF8_stdenc_cstowc(_UTF8EncodingInfo * __restrict ei, 334 wchar_t * __restrict wc, 335 _csid_t csid, _index_t idx) 336 { 337 338 _DIAGASSERT(wc != NULL); 339 340 if (csid != 0) 341 return (EILSEQ); 342 343 *wc = (wchar_t)idx; 344 345 return (0); 346 } 347 348 static __inline int 349 /*ARGSUSED*/ 350 _citrus_UTF8_stdenc_get_state_desc_generic(_UTF8EncodingInfo * __restrict ei, 351 _UTF8State * __restrict psenc, 352 int * __restrict rstate) 353 { 354 355 if (psenc->chlen == 0) 356 *rstate = _STDENC_SDGEN_INITIAL; 357 else 358 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR; 359 360 return 0; 361 } 362 363 static int 364 /*ARGSUSED*/ 365 _citrus_UTF8_encoding_module_init(_UTF8EncodingInfo * __restrict ei, 366 const void * __restrict var, size_t lenvar) 367 { 368 _UTF8_init_count(); 369 370 return 0; 371 } 372 373 static void 374 /*ARGSUSED*/ 375 _citrus_UTF8_encoding_module_uninit(_UTF8EncodingInfo *ei) 376 { 377 } 378 379 380 /* ---------------------------------------------------------------------- 381 * public interface for ctype 382 */ 383 384 _CITRUS_CTYPE_DECLS(UTF8); 385 _CITRUS_CTYPE_DEF_OPS(UTF8); 386 387 #include "citrus_ctype_template.h" 388 389 /* ---------------------------------------------------------------------- 390 * public interface for stdenc 391 */ 392 393 _CITRUS_STDENC_DECLS(UTF8); 394 _CITRUS_STDENC_DEF_OPS(UTF8); 395 396 #include "citrus_stdenc_template.h" 397