1 /* $NetBSD: citrus_utf8.c,v 1.15 2006/03/22 00:08:09 christos 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 1.15 2006/03/22 00:08:09 christos 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 void 170 /*ARGSUSED*/ 171 _citrus_UTF8_init_state(_UTF8EncodingInfo *ei, _UTF8State *s) 172 { 173 memset(s, 0, sizeof(*s)); 174 } 175 176 static __inline void 177 /*ARGSUSED*/ 178 _citrus_UTF8_pack_state(_UTF8EncodingInfo *ei, void *pspriv, 179 const _UTF8State *s) 180 { 181 memcpy(pspriv, (const void *)s, sizeof(*s)); 182 } 183 184 static __inline void 185 /*ARGSUSED*/ 186 _citrus_UTF8_unpack_state(_UTF8EncodingInfo *ei, _UTF8State *s, 187 const void *pspriv) 188 { 189 memcpy((void *)s, pspriv, sizeof(*s)); 190 } 191 192 static int 193 _citrus_UTF8_mbrtowc_priv(_UTF8EncodingInfo *ei, wchar_t *pwc, const char **s, 194 size_t n, _UTF8State *psenc, size_t *nresult) 195 { 196 wchar_t wchar; 197 const char *s0; 198 int c; 199 int i; 200 int chlenbak; 201 202 _DIAGASSERT(nresult != 0); 203 _DIAGASSERT(s != NULL); 204 _DIAGASSERT(psenc != NULL); 205 206 s0 = *s; 207 208 if (s0 == NULL) { 209 _citrus_UTF8_init_state(ei, psenc); 210 *nresult = 0; /* state independent */ 211 return (0); 212 } 213 214 chlenbak = psenc->chlen; 215 216 /* make sure we have the first byte in the buffer */ 217 switch (psenc->chlen) { 218 case 0: 219 if (n < 1) { 220 goto restart; 221 } 222 psenc->ch[0] = *s0++; 223 psenc->chlen = 1; 224 n--; 225 break; 226 case 1: case 2: case 3: case 4: case 5: 227 break; 228 default: 229 /* illegal state */ 230 goto ilseq; 231 } 232 233 c = _UTF8_count[psenc->ch[0] & 0xff]; 234 if (c == 0) 235 goto ilseq; 236 while (psenc->chlen < c) { 237 if (n < 1) { 238 goto restart; 239 } 240 psenc->ch[psenc->chlen] = *s0++; 241 psenc->chlen++; 242 n--; 243 } 244 245 switch (c) { 246 case 1: 247 wchar = psenc->ch[0] & 0xff; 248 break; 249 case 2: case 3: case 4: case 5: case 6: 250 wchar = psenc->ch[0] & (0x7f >> c); 251 for (i = 1; i < c; i++) { 252 if ((psenc->ch[i] & 0xc0) != 0x80) 253 goto ilseq; 254 wchar <<= 6; 255 wchar |= (psenc->ch[i] & 0x3f); 256 } 257 258 _DIAGASSERT(findlen(wchar) == c); 259 260 break; 261 default: 262 /* illegal state */ 263 goto ilseq; 264 } 265 266 *s = s0; 267 268 psenc->chlen = 0; 269 270 if (pwc) 271 *pwc = wchar; 272 273 if (!wchar) 274 *nresult = 0; 275 else 276 *nresult = c - chlenbak; 277 278 return (0); 279 280 ilseq: 281 psenc->chlen = 0; 282 *nresult = (size_t)-1; 283 return (EILSEQ); 284 285 restart: 286 *s = s0; 287 *nresult = (size_t)-2; 288 return (0); 289 } 290 291 static int 292 _citrus_UTF8_wcrtomb_priv(_UTF8EncodingInfo *ei, char *s, size_t n, wchar_t wc, 293 _UTF8State *psenc, size_t *nresult) 294 { 295 int cnt, i, ret; 296 wchar_t c; 297 298 _DIAGASSERT(nresult != 0); 299 _DIAGASSERT(s != NULL); 300 301 cnt = _UTF8_findlen(wc); 302 if (cnt <= 0 || cnt > 6) { 303 /* invalid UCS4 value */ 304 ret = EILSEQ; 305 goto err; 306 } 307 if (n < cnt) { 308 /* bound check failure */ 309 ret = E2BIG; 310 goto err; 311 } 312 313 c = wc; 314 if (s) { 315 for (i = cnt - 1; i > 0; i--) { 316 s[i] = 0x80 | (c & 0x3f); 317 c >>= 6; 318 } 319 s[0] = c; 320 if (cnt == 1) 321 s[0] &= 0x7f; 322 else { 323 s[0] &= (0x7f >> cnt); 324 s[0] |= ((0xff00 >> cnt) & 0xff); 325 } 326 } 327 328 *nresult = (size_t)cnt; 329 return 0; 330 331 err: 332 *nresult = (size_t)-1; 333 return ret; 334 } 335 336 static __inline int 337 /*ARGSUSED*/ 338 _citrus_UTF8_stdenc_wctocs(_UTF8EncodingInfo * __restrict ei, 339 _csid_t * __restrict csid, 340 _index_t * __restrict idx, 341 wchar_t wc) 342 { 343 344 _DIAGASSERT(csid != NULL && idx != NULL); 345 346 *csid = 0; 347 *idx = (_citrus_index_t)wc; 348 349 return (0); 350 } 351 352 static __inline int 353 /*ARGSUSED*/ 354 _citrus_UTF8_stdenc_cstowc(_UTF8EncodingInfo * __restrict ei, 355 wchar_t * __restrict wc, 356 _csid_t csid, _index_t idx) 357 { 358 359 _DIAGASSERT(wc != NULL); 360 361 if (csid != 0) 362 return (EILSEQ); 363 364 *wc = (wchar_t)idx; 365 366 return (0); 367 } 368 369 static __inline int 370 /*ARGSUSED*/ 371 _citrus_UTF8_stdenc_get_state_desc_generic(_UTF8EncodingInfo * __restrict ei, 372 _UTF8State * __restrict psenc, 373 int * __restrict rstate) 374 { 375 376 if (psenc->chlen == 0) 377 *rstate = _STDENC_SDGEN_INITIAL; 378 else 379 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR; 380 381 return 0; 382 } 383 384 static int 385 /*ARGSUSED*/ 386 _citrus_UTF8_encoding_module_init(_UTF8EncodingInfo * __restrict ei, 387 const void * __restrict var, size_t lenvar) 388 { 389 _UTF8_init_count(); 390 391 return 0; 392 } 393 394 static void 395 /*ARGSUSED*/ 396 _citrus_UTF8_encoding_module_uninit(_UTF8EncodingInfo *ei) 397 { 398 } 399 400 401 /* ---------------------------------------------------------------------- 402 * public interface for ctype 403 */ 404 405 _CITRUS_CTYPE_DECLS(UTF8); 406 _CITRUS_CTYPE_DEF_OPS(UTF8); 407 408 #include "citrus_ctype_template.h" 409 410 /* ---------------------------------------------------------------------- 411 * public interface for stdenc 412 */ 413 414 _CITRUS_STDENC_DECLS(UTF8); 415 _CITRUS_STDENC_DEF_OPS(UTF8); 416 417 #include "citrus_stdenc_template.h" 418