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 const char *s; 241 int err; 242 243 memset((void *)ei, 0, sizeof(*ei)); 244 TAILQ_INIT(&ei->excludes); 245 246 if (lenvar > 0 && var != NULL) { 247 s = _bcs_skip_ws_len((const char *)var, &lenvar); 248 if (lenvar > 0 && *s != '\0') { 249 err = _citrus_prop_parse_variable( 250 root_hints, (void *)ei, s, lenvar); 251 if (err == 0) 252 return (0); 253 254 _citrus_BIG5_encoding_module_uninit(ei); 255 memset((void *)ei, 0, sizeof(*ei)); 256 TAILQ_INIT(&ei->excludes); 257 } 258 } 259 260 /* fallback Big5-1984, for backward compatibility. */ 261 _citrus_BIG5_fill_rowcol(ei, "row", 0xA1, 0xFE); 262 _citrus_BIG5_fill_rowcol(ei, "col", 0x40, 0x7E); 263 _citrus_BIG5_fill_rowcol(ei, "col", 0xA1, 0xFE); 264 265 return (0); 266 } 267 268 static int 269 /*ARGSUSED*/ 270 _citrus_BIG5_mbrtowc_priv(_BIG5EncodingInfo * __restrict ei, 271 wchar_t * __restrict pwc, 272 const char ** __restrict s, size_t n, 273 _BIG5State * __restrict psenc, 274 size_t * __restrict nresult) 275 { 276 wchar_t wchar; 277 const char *s0; 278 int c, chlenbak; 279 280 s0 = *s; 281 282 if (s0 == NULL) { 283 _citrus_BIG5_init_state(ei, psenc); 284 *nresult = 0; 285 return (0); 286 } 287 288 chlenbak = psenc->chlen; 289 290 /* make sure we have the first byte in the buffer */ 291 switch (psenc->chlen) { 292 case 0: 293 if (n < 1) 294 goto restart; 295 psenc->ch[0] = *s0++; 296 psenc->chlen = 1; 297 n--; 298 break; 299 case 1: 300 break; 301 default: 302 /* illegal state */ 303 goto ilseq; 304 } 305 306 c = _citrus_BIG5_check(ei, psenc->ch[0] & 0xff); 307 if (c == 0) 308 goto ilseq; 309 while (psenc->chlen < c) { 310 if (n < 1) { 311 goto restart; 312 } 313 psenc->ch[psenc->chlen] = *s0++; 314 psenc->chlen++; 315 n--; 316 } 317 318 switch (c) { 319 case 1: 320 wchar = psenc->ch[0] & 0xff; 321 break; 322 case 2: 323 if (!_citrus_BIG5_check2(ei, psenc->ch[1] & 0xff)) 324 goto ilseq; 325 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff); 326 break; 327 default: 328 /* illegal state */ 329 goto ilseq; 330 } 331 332 if (_citrus_BIG5_check_excludes(ei, (wint_t)wchar) != 0) 333 goto ilseq; 334 335 *s = s0; 336 psenc->chlen = 0; 337 if (pwc) 338 *pwc = wchar; 339 *nresult = wchar ? c - chlenbak : 0; 340 341 return (0); 342 343 ilseq: 344 psenc->chlen = 0; 345 *nresult = (size_t)-1; 346 return (EILSEQ); 347 348 restart: 349 *s = s0; 350 *nresult = (size_t)-2; 351 return (0); 352 } 353 354 static int 355 /*ARGSUSED*/ 356 _citrus_BIG5_wcrtomb_priv(_BIG5EncodingInfo * __restrict ei, 357 char * __restrict s, 358 size_t n, wchar_t wc, _BIG5State * __restrict psenc __unused, 359 size_t * __restrict nresult) 360 { 361 unsigned char l; 362 int ret; 363 364 /* check invalid sequence */ 365 if (wc & ~0xffff || 366 _citrus_BIG5_check_excludes(ei, (wint_t)wc) != 0) { 367 ret = EILSEQ; 368 goto err; 369 } 370 371 if (wc & 0x8000) { 372 if (_citrus_BIG5_check(ei, (wc >> 8) & 0xff) != 2 || 373 !_citrus_BIG5_check2(ei, wc & 0xff)) { 374 ret = EILSEQ; 375 goto err; 376 } 377 l = 2; 378 } else { 379 if (wc & ~0xff || !_citrus_BIG5_check(ei, wc & 0xff)) { 380 ret = EILSEQ; 381 goto err; 382 } 383 l = 1; 384 } 385 386 if (n < l) { 387 /* bound check failure */ 388 ret = E2BIG; 389 goto err; 390 } 391 392 if (l == 2) { 393 s[0] = (wc >> 8) & 0xff; 394 s[1] = wc & 0xff; 395 } else 396 s[0] = wc & 0xff; 397 398 *nresult = l; 399 400 return (0); 401 402 err: 403 *nresult = (size_t)-1; 404 return (ret); 405 } 406 407 static __inline int 408 /*ARGSUSED*/ 409 _citrus_BIG5_stdenc_wctocs(_BIG5EncodingInfo * __restrict ei __unused, 410 _csid_t * __restrict csid, 411 _index_t * __restrict idx, wchar_t wc) 412 { 413 414 *csid = (wc < 0x100) ? 0 : 1; 415 *idx = (_index_t)wc; 416 417 return (0); 418 } 419 420 static __inline int 421 /*ARGSUSED*/ 422 _citrus_BIG5_stdenc_cstowc(_BIG5EncodingInfo * __restrict ei __unused, 423 wchar_t * __restrict wc, 424 _csid_t csid, _index_t idx) 425 { 426 427 switch (csid) { 428 case 0: 429 case 1: 430 *wc = (wchar_t)idx; 431 break; 432 default: 433 return (EILSEQ); 434 } 435 436 return (0); 437 } 438 439 static __inline int 440 /*ARGSUSED*/ 441 _citrus_BIG5_stdenc_get_state_desc_generic(_BIG5EncodingInfo * __restrict ei __unused, 442 _BIG5State * __restrict psenc, 443 int * __restrict rstate) 444 { 445 446 *rstate = (psenc->chlen == 0) ? _STDENC_SDGEN_INITIAL : 447 _STDENC_SDGEN_INCOMPLETE_CHAR; 448 return (0); 449 } 450 451 /* ---------------------------------------------------------------------- 452 * public interface for stdenc 453 */ 454 455 _CITRUS_STDENC_DECLS(BIG5); 456 _CITRUS_STDENC_DEF_OPS(BIG5); 457 458 #include "citrus_stdenc_template.h" 459