1 /* $FreeBSD$ */ 2 /* $NetBSD: citrus_big5.c,v 1.12 2008/06/14 16:01:07 tnozaki Exp $ */ 3 4 /*- 5 * Copyright (c)2002, 2006 Citrus Project, 6 * All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 */ 29 30 /*- 31 * Copyright (c) 1993 32 * The Regents of the University of California. All rights reserved. 33 * 34 * This code is derived from software contributed to Berkeley by 35 * Paul Borman at Krystal Technologies. 36 * 37 * Redistribution and use in source and binary forms, with or without 38 * modification, are permitted provided that the following conditions 39 * are met: 40 * 1. Redistributions of source code must retain the above copyright 41 * notice, this list of conditions and the following disclaimer. 42 * 2. Redistributions in binary form must reproduce the above copyright 43 * notice, this list of conditions and the following disclaimer in the 44 * documentation and/or other materials provided with the distribution. 45 * 3. Neither the name of the University nor the names of its contributors 46 * may be used to endorse or promote products derived from this software 47 * without specific prior written permission. 48 * 49 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 50 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 51 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 52 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 53 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 54 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 55 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 56 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 57 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 58 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 59 * SUCH DAMAGE. 60 */ 61 62 #include <sys/cdefs.h> 63 #include <sys/queue.h> 64 #include <sys/types.h> 65 66 #include <assert.h> 67 #include <errno.h> 68 #include <limits.h> 69 #include <stddef.h> 70 #include <stdint.h> 71 #include <stdio.h> 72 #include <stdlib.h> 73 #include <string.h> 74 #include <wchar.h> 75 76 #include "citrus_namespace.h" 77 #include "citrus_prop.h" 78 #include "citrus_types.h" 79 #include "citrus_bcs.h" 80 #include "citrus_module.h" 81 #include "citrus_stdenc.h" 82 #include "citrus_big5.h" 83 84 /* ---------------------------------------------------------------------- 85 * private stuffs used by templates 86 */ 87 88 typedef struct { 89 int chlen; 90 char ch[2]; 91 } _BIG5State; 92 93 typedef struct _BIG5Exclude { 94 TAILQ_ENTRY(_BIG5Exclude) entry; 95 wint_t end; 96 wint_t start; 97 } _BIG5Exclude; 98 99 typedef TAILQ_HEAD(_BIG5ExcludeList, _BIG5Exclude) _BIG5ExcludeList; 100 101 typedef struct { 102 _BIG5ExcludeList excludes; 103 int cell[0x100]; 104 } _BIG5EncodingInfo; 105 106 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei) 107 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_ 108 109 #define _FUNCNAME(m) _citrus_BIG5_##m 110 #define _ENCODING_INFO _BIG5EncodingInfo 111 #define _ENCODING_STATE _BIG5State 112 #define _ENCODING_MB_CUR_MAX(_ei_) 2 113 #define _ENCODING_IS_STATE_DEPENDENT 0 114 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0 115 116 117 static __inline void 118 /*ARGSUSED*/ 119 _citrus_BIG5_init_state(_BIG5EncodingInfo * __restrict ei __unused, 120 _BIG5State * __restrict s) 121 { 122 123 memset(s, 0, sizeof(*s)); 124 } 125 126 #if 0 127 static __inline void 128 /*ARGSUSED*/ 129 _citrus_BIG5_pack_state(_BIG5EncodingInfo * __restrict ei __unused, 130 void * __restrict pspriv, 131 const _BIG5State * __restrict s) 132 { 133 134 memcpy(pspriv, (const void *)s, sizeof(*s)); 135 } 136 137 static __inline void 138 /*ARGSUSED*/ 139 _citrus_BIG5_unpack_state(_BIG5EncodingInfo * __restrict ei __unused, 140 _BIG5State * __restrict s, 141 const void * __restrict pspriv) 142 { 143 144 memcpy((void *)s, pspriv, sizeof(*s)); 145 } 146 #endif 147 148 static __inline int 149 _citrus_BIG5_check(_BIG5EncodingInfo *ei, unsigned int c) 150 { 151 152 return ((ei->cell[c & 0xFF] & 0x1) ? 2 : 1); 153 } 154 155 static __inline int 156 _citrus_BIG5_check2(_BIG5EncodingInfo *ei, unsigned int c) 157 { 158 159 return ((ei->cell[c & 0xFF] & 0x2) ? 1 : 0); 160 } 161 162 static __inline int 163 _citrus_BIG5_check_excludes(_BIG5EncodingInfo *ei, wint_t c) 164 { 165 _BIG5Exclude *exclude; 166 167 TAILQ_FOREACH(exclude, &ei->excludes, entry) { 168 if (c >= exclude->start && c <= exclude->end) 169 return (EILSEQ); 170 } 171 return (0); 172 } 173 174 static int 175 _citrus_BIG5_fill_rowcol(void ** __restrict ctx, const char * __restrict s, 176 uint64_t start, uint64_t end) 177 { 178 _BIG5EncodingInfo *ei; 179 uint64_t n; 180 int i; 181 182 if (start > 0xFF || end > 0xFF) 183 return (EINVAL); 184 ei = (_BIG5EncodingInfo *)ctx; 185 i = strcmp("row", s) ? 1 : 0; 186 i = 1 << i; 187 for (n = start; n <= end; ++n) 188 ei->cell[n & 0xFF] |= i; 189 return (0); 190 } 191 192 static int 193 /*ARGSUSED*/ 194 _citrus_BIG5_fill_excludes(void ** __restrict ctx, 195 const char * __restrict s __unused, uint64_t start, uint64_t end) 196 { 197 _BIG5EncodingInfo *ei; 198 _BIG5Exclude *exclude; 199 200 if (start > 0xFFFF || end > 0xFFFF) 201 return (EINVAL); 202 ei = (_BIG5EncodingInfo *)ctx; 203 exclude = TAILQ_LAST(&ei->excludes, _BIG5ExcludeList); 204 if (exclude != NULL && (wint_t)start <= exclude->end) 205 return (EINVAL); 206 exclude = (void *)malloc(sizeof(*exclude)); 207 if (exclude == NULL) 208 return (ENOMEM); 209 exclude->start = (wint_t)start; 210 exclude->end = (wint_t)end; 211 TAILQ_INSERT_TAIL(&ei->excludes, exclude, entry); 212 213 return (0); 214 } 215 216 static const _citrus_prop_hint_t root_hints[] = { 217 _CITRUS_PROP_HINT_NUM("row", &_citrus_BIG5_fill_rowcol), 218 _CITRUS_PROP_HINT_NUM("col", &_citrus_BIG5_fill_rowcol), 219 _CITRUS_PROP_HINT_NUM("excludes", &_citrus_BIG5_fill_excludes), 220 _CITRUS_PROP_HINT_END 221 }; 222 223 static void 224 /*ARGSUSED*/ 225 _citrus_BIG5_encoding_module_uninit(_BIG5EncodingInfo *ei) 226 { 227 _BIG5Exclude *exclude; 228 229 while ((exclude = TAILQ_FIRST(&ei->excludes)) != NULL) { 230 TAILQ_REMOVE(&ei->excludes, exclude, entry); 231 free(exclude); 232 } 233 } 234 235 static int 236 /*ARGSUSED*/ 237 _citrus_BIG5_encoding_module_init(_BIG5EncodingInfo * __restrict ei, 238 const void * __restrict var, size_t lenvar) 239 { 240 void *ctx = (void *)ei; 241 const char *s; 242 int err; 243 244 memset((void *)ei, 0, sizeof(*ei)); 245 TAILQ_INIT(&ei->excludes); 246 247 if (lenvar > 0 && var != NULL) { 248 s = _bcs_skip_ws_len((const char *)var, &lenvar); 249 if (lenvar > 0 && *s != '\0') { 250 err = _citrus_prop_parse_variable( 251 root_hints, (void *)ei, s, lenvar); 252 if (err == 0) 253 return (0); 254 255 _citrus_BIG5_encoding_module_uninit(ei); 256 memset((void *)ei, 0, sizeof(*ei)); 257 TAILQ_INIT(&ei->excludes); 258 } 259 } 260 261 /* fallback Big5-1984, for backward compatibility. */ 262 _citrus_BIG5_fill_rowcol((void **)&ctx, "row", 0xA1, 0xFE); 263 _citrus_BIG5_fill_rowcol((void **)&ctx, "col", 0x40, 0x7E); 264 _citrus_BIG5_fill_rowcol((void **)&ctx, "col", 0xA1, 0xFE); 265 266 return (0); 267 } 268 269 static int 270 /*ARGSUSED*/ 271 _citrus_BIG5_mbrtowc_priv(_BIG5EncodingInfo * __restrict ei, 272 wchar_t * __restrict pwc, 273 const char ** __restrict s, size_t n, 274 _BIG5State * __restrict psenc, 275 size_t * __restrict nresult) 276 { 277 wchar_t wchar; 278 const char *s0; 279 int c, chlenbak; 280 281 s0 = *s; 282 283 if (s0 == NULL) { 284 _citrus_BIG5_init_state(ei, psenc); 285 *nresult = 0; 286 return (0); 287 } 288 289 chlenbak = psenc->chlen; 290 291 /* make sure we have the first byte in the buffer */ 292 switch (psenc->chlen) { 293 case 0: 294 if (n < 1) 295 goto restart; 296 psenc->ch[0] = *s0++; 297 psenc->chlen = 1; 298 n--; 299 break; 300 case 1: 301 break; 302 default: 303 /* illegal state */ 304 goto ilseq; 305 } 306 307 c = _citrus_BIG5_check(ei, psenc->ch[0] & 0xff); 308 if (c == 0) 309 goto ilseq; 310 while (psenc->chlen < c) { 311 if (n < 1) { 312 goto restart; 313 } 314 psenc->ch[psenc->chlen] = *s0++; 315 psenc->chlen++; 316 n--; 317 } 318 319 switch (c) { 320 case 1: 321 wchar = psenc->ch[0] & 0xff; 322 break; 323 case 2: 324 if (!_citrus_BIG5_check2(ei, psenc->ch[1] & 0xff)) 325 goto ilseq; 326 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff); 327 break; 328 default: 329 /* illegal state */ 330 goto ilseq; 331 } 332 333 if (_citrus_BIG5_check_excludes(ei, (wint_t)wchar) != 0) 334 goto ilseq; 335 336 *s = s0; 337 psenc->chlen = 0; 338 if (pwc) 339 *pwc = wchar; 340 *nresult = wchar ? c - chlenbak : 0; 341 342 return (0); 343 344 ilseq: 345 psenc->chlen = 0; 346 *nresult = (size_t)-1; 347 return (EILSEQ); 348 349 restart: 350 *s = s0; 351 *nresult = (size_t)-2; 352 return (0); 353 } 354 355 static int 356 /*ARGSUSED*/ 357 _citrus_BIG5_wcrtomb_priv(_BIG5EncodingInfo * __restrict ei, 358 char * __restrict s, 359 size_t n, wchar_t wc, _BIG5State * __restrict psenc __unused, 360 size_t * __restrict nresult) 361 { 362 unsigned char l; 363 int ret; 364 365 /* check invalid sequence */ 366 if (wc & ~0xffff || 367 _citrus_BIG5_check_excludes(ei, (wint_t)wc) != 0) { 368 ret = EILSEQ; 369 goto err; 370 } 371 372 if (wc & 0x8000) { 373 if (_citrus_BIG5_check(ei, (wc >> 8) & 0xff) != 2 || 374 !_citrus_BIG5_check2(ei, wc & 0xff)) { 375 ret = EILSEQ; 376 goto err; 377 } 378 l = 2; 379 } else { 380 if (wc & ~0xff || !_citrus_BIG5_check(ei, wc & 0xff)) { 381 ret = EILSEQ; 382 goto err; 383 } 384 l = 1; 385 } 386 387 if (n < l) { 388 /* bound check failure */ 389 ret = E2BIG; 390 goto err; 391 } 392 393 if (l == 2) { 394 s[0] = (wc >> 8) & 0xff; 395 s[1] = wc & 0xff; 396 } else 397 s[0] = wc & 0xff; 398 399 *nresult = l; 400 401 return (0); 402 403 err: 404 *nresult = (size_t)-1; 405 return (ret); 406 } 407 408 static __inline int 409 /*ARGSUSED*/ 410 _citrus_BIG5_stdenc_wctocs(_BIG5EncodingInfo * __restrict ei __unused, 411 _csid_t * __restrict csid, 412 _index_t * __restrict idx, wchar_t wc) 413 { 414 415 *csid = (wc < 0x100) ? 0 : 1; 416 *idx = (_index_t)wc; 417 418 return (0); 419 } 420 421 static __inline int 422 /*ARGSUSED*/ 423 _citrus_BIG5_stdenc_cstowc(_BIG5EncodingInfo * __restrict ei __unused, 424 wchar_t * __restrict wc, 425 _csid_t csid, _index_t idx) 426 { 427 428 switch (csid) { 429 case 0: 430 case 1: 431 *wc = (wchar_t)idx; 432 break; 433 default: 434 return (EILSEQ); 435 } 436 437 return (0); 438 } 439 440 static __inline int 441 /*ARGSUSED*/ 442 _citrus_BIG5_stdenc_get_state_desc_generic(_BIG5EncodingInfo * __restrict ei __unused, 443 _BIG5State * __restrict psenc, 444 int * __restrict rstate) 445 { 446 447 *rstate = (psenc->chlen == 0) ? _STDENC_SDGEN_INITIAL : 448 _STDENC_SDGEN_INCOMPLETE_CHAR; 449 return (0); 450 } 451 452 /* ---------------------------------------------------------------------- 453 * public interface for stdenc 454 */ 455 456 _CITRUS_STDENC_DECLS(BIG5); 457 _CITRUS_STDENC_DEF_OPS(BIG5); 458 459 #include "citrus_stdenc_template.h" 460