1 /* $NetBSD: citrus_euctw.c,v 1.10 2005/10/29 18:02:04 tshiozak 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)1999 Citrus Project, 31 * All rights reserved. 32 * 33 * Redistribution and use in source and binary forms, with or without 34 * modification, are permitted provided that the following conditions 35 * are met: 36 * 1. Redistributions of source code must retain the above copyright 37 * notice, this list of conditions and the following disclaimer. 38 * 2. Redistributions in binary form must reproduce the above copyright 39 * notice, this list of conditions and the following disclaimer in the 40 * documentation and/or other materials provided with the distribution. 41 * 42 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 43 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 44 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 45 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 46 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 47 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 48 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 49 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 50 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 51 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 52 * SUCH DAMAGE. 53 * 54 * $Citrus: xpg4dl/FreeBSD/lib/libc/locale/euctw.c,v 1.13 2001/06/21 01:51:44 yamt Exp $ 55 */ 56 57 #include <sys/cdefs.h> 58 #if defined(LIBC_SCCS) && !defined(lint) 59 __RCSID("$NetBSD: citrus_euctw.c,v 1.10 2005/10/29 18:02:04 tshiozak Exp $"); 60 #endif /* LIBC_SCCS and not lint */ 61 62 #include <assert.h> 63 #include <errno.h> 64 #include <string.h> 65 #include <stdio.h> 66 #include <stdlib.h> 67 #include <stddef.h> 68 #include <locale.h> 69 #include <wchar.h> 70 #include <sys/types.h> 71 #include <limits.h> 72 73 #include "citrus_namespace.h" 74 #include "citrus_types.h" 75 #include "citrus_module.h" 76 #include "citrus_ctype.h" 77 #include "citrus_stdenc.h" 78 #include "citrus_euctw.h" 79 80 81 /* ---------------------------------------------------------------------- 82 * private stuffs used by templates 83 */ 84 85 typedef struct { 86 char ch[4]; 87 int chlen; 88 } _EUCTWState; 89 90 typedef struct { 91 int dummy; 92 } _EUCTWEncodingInfo; 93 typedef struct { 94 _EUCTWEncodingInfo ei; 95 struct { 96 /* for future multi-locale facility */ 97 _EUCTWState s_mblen; 98 _EUCTWState s_mbrlen; 99 _EUCTWState s_mbrtowc; 100 _EUCTWState s_mbtowc; 101 _EUCTWState s_mbsrtowcs; 102 _EUCTWState s_wcrtomb; 103 _EUCTWState s_wcsrtombs; 104 _EUCTWState s_wctomb; 105 } states; 106 } _EUCTWCTypeInfo; 107 108 #define _SS2 0x008e 109 #define _SS3 0x008f 110 111 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei) 112 #define _CEI_TO_STATE(_cei_, _func_) (_cei_)->states.s_##_func_ 113 114 #define _FUNCNAME(m) _citrus_EUCTW_##m 115 #define _ENCODING_INFO _EUCTWEncodingInfo 116 #define _CTYPE_INFO _EUCTWCTypeInfo 117 #define _ENCODING_STATE _EUCTWState 118 #define _ENCODING_MB_CUR_MAX(_ei_) 4 119 #define _ENCODING_IS_STATE_DEPENDENT 0 120 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0 121 122 static __inline int 123 _citrus_EUCTW_cs(u_int c) 124 { 125 c &= 0xff; 126 127 return ((c & 0x80) ? (c == _SS2 ? 2 : 1) : 0); 128 } 129 130 static __inline int 131 _citrus_EUCTW_count(int cs) 132 { 133 switch (cs) { 134 case 0: 135 return 1; 136 case 1: 137 return 2; 138 case 2: 139 return 4; 140 case 3: 141 abort(); 142 /*NOTREACHED*/ 143 } 144 return 0; 145 } 146 147 static __inline void 148 /*ARGSUSED*/ 149 _citrus_EUCTW_init_state(_EUCTWEncodingInfo * __restrict ei, 150 _EUCTWState * __restrict s) 151 { 152 memset(s, 0, sizeof(*s)); 153 } 154 155 static __inline void 156 /*ARGSUSED*/ 157 _citrus_EUCTW_pack_state(_EUCTWEncodingInfo * __restrict ei, 158 void * __restrict pspriv, 159 const _EUCTWState * __restrict s) 160 { 161 memcpy(pspriv, (const void *)s, sizeof(*s)); 162 } 163 164 static __inline void 165 /*ARGSUSED*/ 166 _citrus_EUCTW_unpack_state(_EUCTWEncodingInfo * __restrict ei, 167 _EUCTWState * __restrict s, 168 const void * __restrict pspriv) 169 { 170 memcpy((void *)s, pspriv, sizeof(*s)); 171 } 172 173 static int 174 /*ARGSUSED*/ 175 _citrus_EUCTW_encoding_module_init(_EUCTWEncodingInfo * __restrict ei, 176 const void * __restrict var, size_t lenvar) 177 { 178 179 _DIAGASSERT(ei != NULL); 180 181 memset((void *)ei, 0, sizeof(*ei)); 182 183 return 0; 184 } 185 186 static void 187 /*ARGSUSED*/ 188 _citrus_EUCTW_encoding_module_uninit(_EUCTWEncodingInfo *ei) 189 { 190 } 191 192 static int 193 _citrus_EUCTW_mbrtowc_priv(_EUCTWEncodingInfo * __restrict ei, 194 wchar_t * __restrict pwc, 195 const char ** __restrict s, 196 size_t n, _EUCTWState * __restrict psenc, 197 size_t * __restrict nresult) 198 { 199 wchar_t wchar; 200 int c, cs; 201 int chlenbak; 202 const char *s0; 203 204 _DIAGASSERT(nresult != 0); 205 _DIAGASSERT(ei != NULL); 206 _DIAGASSERT(psenc != NULL); 207 _DIAGASSERT(s != NULL); 208 209 s0 = *s; 210 211 if (s0 == NULL) { 212 _citrus_EUCTW_init_state(ei, psenc); 213 *nresult = 0; /* state independent */ 214 return (0); 215 } 216 217 chlenbak = psenc->chlen; 218 219 /* make sure we have the first byte in the buffer */ 220 switch (psenc->chlen) { 221 case 0: 222 if (n < 1) 223 goto restart; 224 psenc->ch[0] = *s0++; 225 psenc->chlen = 1; 226 n--; 227 break; 228 case 1: 229 case 2: 230 break; 231 default: 232 /* illgeal state */ 233 goto ilseq; 234 } 235 236 c = _citrus_EUCTW_count(cs = _citrus_EUCTW_cs(psenc->ch[0] & 0xff)); 237 if (c == 0) 238 goto ilseq; 239 while (psenc->chlen < c) { 240 if (n < 1) 241 goto ilseq; 242 psenc->ch[psenc->chlen] = *s0++; 243 psenc->chlen++; 244 n--; 245 } 246 247 wchar = 0; 248 switch (cs) { 249 case 0: 250 if (psenc->ch[0] & 0x80) 251 goto ilseq; 252 wchar = psenc->ch[0] & 0xff; 253 break; 254 case 1: 255 if (!(psenc->ch[0] & 0x80) || !(psenc->ch[1] & 0x80)) 256 goto ilseq; 257 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff); 258 wchar |= 'G' << 24; 259 break; 260 case 2: 261 if ((u_char)psenc->ch[1] < 0xa1 || 0xa7 < (u_char)psenc->ch[1]) 262 goto ilseq; 263 if (!(psenc->ch[2] & 0x80) || !(psenc->ch[3] & 0x80)) 264 goto ilseq; 265 wchar = ((psenc->ch[2] & 0xff) << 8) | (psenc->ch[3] & 0xff); 266 wchar |= ('G' + psenc->ch[1] - 0xa1) << 24; 267 break; 268 default: 269 goto ilseq; 270 } 271 272 *s = s0; 273 psenc->chlen = 0; 274 275 if (pwc) 276 *pwc = wchar; 277 278 if (!wchar) 279 *nresult = 0; 280 else 281 *nresult = c - chlenbak; 282 283 return (0); 284 285 ilseq: 286 psenc->chlen = 0; 287 *nresult = (size_t)-1; 288 return (EILSEQ); 289 290 restart: 291 *s = s0; 292 *nresult = (size_t)-1; 293 return (0); 294 } 295 296 static int 297 _citrus_EUCTW_wcrtomb_priv(_EUCTWEncodingInfo * __restrict ei, 298 char * __restrict s, size_t n, wchar_t wc, 299 _EUCTWState * __restrict psenc, 300 size_t * __restrict nresult) 301 { 302 wchar_t cs = wc & 0x7f000080; 303 wchar_t v; 304 int i, len, clen, ret; 305 306 _DIAGASSERT(ei != NULL); 307 _DIAGASSERT(nresult != 0); 308 _DIAGASSERT(s != NULL); 309 310 clen = 1; 311 if (wc & 0x00007f00) 312 clen = 2; 313 if ((wc & 0x007f0000) && !(wc & 0x00800000)) 314 clen = 3; 315 316 if (clen == 1 && cs == 0x00000000) { 317 /* ASCII */ 318 len = 1; 319 if (n < len) { 320 ret = E2BIG; 321 goto err; 322 } 323 v = wc & 0x0000007f; 324 } else if (clen == 2 && cs == ('G' << 24)) { 325 /* CNS-11643-1 */ 326 len = 2; 327 if (n < len) { 328 ret = E2BIG; 329 goto err; 330 } 331 v = wc & 0x00007f7f; 332 v |= 0x00008080; 333 } else if (clen == 2 && 'H' <= (cs >> 24) && (cs >> 24) <= 'M') { 334 /* CNS-11643-[2-7] */ 335 len = 4; 336 if (n < len) { 337 ret = E2BIG; 338 goto err; 339 } 340 *s++ = _SS2; 341 *s++ = (cs >> 24) - 'H' + 0xa2; 342 v = wc & 0x00007f7f; 343 v |= 0x00008080; 344 } else { 345 ret = EILSEQ; 346 goto err; 347 } 348 349 i = clen; 350 while (i-- > 0) 351 *s++ = (v >> (i << 3)) & 0xff; 352 353 *nresult = len; 354 return 0; 355 356 err: 357 *nresult = (size_t)-1; 358 return ret; 359 } 360 361 static __inline int 362 /*ARGSUSED*/ 363 _citrus_EUCTW_stdenc_wctocs(_EUCTWEncodingInfo * __restrict ei, 364 _csid_t * __restrict csid, 365 _index_t * __restrict idx, wchar_t wc) 366 { 367 368 _DIAGASSERT(ei != NULL && csid != NULL && idx != NULL); 369 370 *csid = (_csid_t)(wc >> 24) & 0xFF; 371 *idx = (_index_t)(wc & 0x7F7F); 372 373 return (0); 374 } 375 376 static __inline int 377 /*ARGSUSED*/ 378 _citrus_EUCTW_stdenc_cstowc(_EUCTWEncodingInfo * __restrict ei, 379 wchar_t * __restrict wc, 380 _csid_t csid, _index_t idx) 381 { 382 383 _DIAGASSERT(ei != NULL && wc != NULL); 384 385 if (csid==0) { 386 if ((idx & ~0x7F) != 0) 387 return (EINVAL); 388 *wc = (wchar_t)idx; 389 } else { 390 if (csid < 'G' || csid > 'M' || (idx & ~0x7F7F) != 0) 391 return (EINVAL); 392 *wc = (wchar_t)idx | ((wchar_t)csid<<24); 393 } 394 395 return (0); 396 } 397 398 static __inline int 399 /*ARGSUSED*/ 400 _citrus_EUCTW_stdenc_get_state_desc_generic(_EUCTWEncodingInfo * __restrict ei, 401 _EUCTWState * __restrict psenc, 402 int * __restrict rstate) 403 { 404 405 if (psenc->chlen == 0) 406 *rstate = _STDENC_SDGEN_INITIAL; 407 else 408 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR; 409 410 return 0; 411 } 412 413 /* ---------------------------------------------------------------------- 414 * public interface for ctype 415 */ 416 417 _CITRUS_CTYPE_DECLS(EUCTW); 418 _CITRUS_CTYPE_DEF_OPS(EUCTW); 419 420 #include "citrus_ctype_template.h" 421 422 /* ---------------------------------------------------------------------- 423 * public interface for stdenc 424 */ 425 426 _CITRUS_STDENC_DECLS(EUCTW); 427 _CITRUS_STDENC_DEF_OPS(EUCTW); 428 429 #include "citrus_stdenc_template.h" 430