lib/libc/citrus/modules/citrus_euctw.c

   1 /*      $NetBSD: citrus_euctw.c,v 1.12 2013/05/28 16:57:56 joerg 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 1.12 2013/05/28 16:57:56 joerg 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 <wchar.h>
  69 #include <sys/types.h>
  70 #include <limits.h>
  71
  72 #include "citrus_namespace.h"
  73 #include "citrus_types.h"
  74 #include "citrus_module.h"
  75 #include "citrus_ctype.h"
  76 #include "citrus_stdenc.h"
  77 #include "citrus_euctw.h"
  78
  79
  80 /* ----------------------------------------------------------------------
  81  * private stuffs used by templates
  82  */
  83
  84 typedef struct {
  85         char ch[4];
  86         int chlen;
  87 } _EUCTWState;
  88
  89 typedef struct {
  90         int dummy;
  91 } _EUCTWEncodingInfo;
  92 typedef struct {
  93         _EUCTWEncodingInfo      ei;
  94         struct {
  95                 /* for future multi-locale facility */
  96                 _EUCTWState     s_mblen;
  97                 _EUCTWState     s_mbrlen;
  98                 _EUCTWState     s_mbrtowc;
  99                 _EUCTWState     s_mbtowc;
 100                 _EUCTWState     s_mbsrtowcs;
 101                 _EUCTWState     s_mbsnrtowcs;
 102                 _EUCTWState     s_wcrtomb;
 103                 _EUCTWState     s_wcsrtombs;
 104                 _EUCTWState     s_wcsnrtombs;
 105                 _EUCTWState     s_wctomb;
 106         } states;
 107 } _EUCTWCTypeInfo;
 108
 109 #define _SS2    0x008e
 110 #define _SS3    0x008f
 111
 112 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 113 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
 114
 115 #define _FUNCNAME(m)                    _citrus_EUCTW_##m
 116 #define _ENCODING_INFO                  _EUCTWEncodingInfo
 117 #define _CTYPE_INFO                     _EUCTWCTypeInfo
 118 #define _ENCODING_STATE                 _EUCTWState
 119 #define _ENCODING_MB_CUR_MAX(_ei_)      4
 120 #define _ENCODING_IS_STATE_DEPENDENT    0
 121 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 122
 123 static __inline int
 124 _citrus_EUCTW_cs(u_int c)
 125 {
 126         c &= 0xff;
 127
 128         return ((c & 0x80) ? (c == _SS2 ? 2 : 1) : 0);
 129 }
 130
 131 static __inline int
 132 _citrus_EUCTW_count(int cs)
 133 {
 134         switch (cs) {
 135         case 0:
 136                 return 1;
 137         case 1:
 138                 return 2;
 139         case 2:
 140                 return 4;
 141         case 3:
 142                 abort();
 143                 /*NOTREACHED*/
 144         }
 145         return 0;
 146 }
 147
 148 static __inline void
 149 /*ARGSUSED*/
 150 _citrus_EUCTW_init_state(_EUCTWEncodingInfo * __restrict ei,
 151                          _EUCTWState * __restrict s)
 152 {
 153         memset(s, 0, sizeof(*s));
 154 }
 155
 156 static __inline void
 157 /*ARGSUSED*/
 158 _citrus_EUCTW_pack_state(_EUCTWEncodingInfo * __restrict ei,
 159                          void * __restrict pspriv,
 160                          const _EUCTWState * __restrict s)
 161 {
 162         memcpy(pspriv, (const void *)s, sizeof(*s));
 163 }
 164
 165 static __inline void
 166 /*ARGSUSED*/
 167 _citrus_EUCTW_unpack_state(_EUCTWEncodingInfo * __restrict ei,
 168                            _EUCTWState * __restrict s,
 169                            const void * __restrict pspriv)
 170 {
 171         memcpy((void *)s, pspriv, sizeof(*s));
 172 }
 173
 174 static int
 175 /*ARGSUSED*/
 176 _citrus_EUCTW_encoding_module_init(_EUCTWEncodingInfo * __restrict ei,
 177                                    const void * __restrict var, size_t lenvar)
 178 {
 179
 180         _DIAGASSERT(ei != NULL);
 181
 182         memset((void *)ei, 0, sizeof(*ei));
 183
 184         return 0;
 185 }
 186
 187 static void
 188 /*ARGSUSED*/
 189 _citrus_EUCTW_encoding_module_uninit(_EUCTWEncodingInfo *ei)
 190 {
 191 }
 192
 193 static int
 194 _citrus_EUCTW_mbrtowc_priv(_EUCTWEncodingInfo * __restrict ei,
 195                            wchar_t * __restrict pwc,
 196                            const char ** __restrict s,
 197                            size_t n, _EUCTWState * __restrict psenc,
 198                            size_t * __restrict nresult)
 199 {
 200         wchar_t wchar;
 201         int c, cs;
 202         int chlenbak;
 203         const char *s0;
 204
 205         _DIAGASSERT(nresult != 0);
 206         _DIAGASSERT(ei != NULL);
 207         _DIAGASSERT(psenc != NULL);
 208         _DIAGASSERT(s != NULL);
 209
 210         s0 = *s;
 211
 212         if (s0 == NULL) {
 213                 _citrus_EUCTW_init_state(ei, psenc);
 214                 *nresult = 0; /* state independent */
 215                 return (0);
 216         }
 217
 218         chlenbak = psenc->chlen;
 219
 220         /* make sure we have the first byte in the buffer */
 221         switch (psenc->chlen) {
 222         case 0:
 223                 if (n < 1)
 224                         goto restart;
 225                 psenc->ch[0] = *s0++;
 226                 psenc->chlen = 1;
 227                 n--;
 228                 break;
 229         case 1:
 230         case 2:
 231                 break;
 232         default:
 233                 /* illgeal state */
 234                 goto ilseq;
 235         }
 236
 237         c = _citrus_EUCTW_count(cs = _citrus_EUCTW_cs(psenc->ch[0] & 0xff));
 238         if (c == 0)
 239                 goto ilseq;
 240         while (psenc->chlen < c) {
 241                 if (n < 1)
 242                         goto ilseq;
 243                 psenc->ch[psenc->chlen] = *s0++;
 244                 psenc->chlen++;
 245                 n--;
 246         }
 247
 248         wchar = 0;
 249         switch (cs) {
 250         case 0:
 251                 if (psenc->ch[0] & 0x80)
 252                         goto ilseq;
 253                 wchar = psenc->ch[0] & 0xff;
 254                 break;
 255         case 1:
 256                 if (!(psenc->ch[0] & 0x80) || !(psenc->ch[1] & 0x80))
 257                         goto ilseq;
 258                 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
 259                 wchar |= 'G' << 24;
 260                 break;
 261         case 2:
 262                 if ((u_char)psenc->ch[1] < 0xa1 || 0xa7 < (u_char)psenc->ch[1])
 263                         goto ilseq;
 264                 if (!(psenc->ch[2] & 0x80) || !(psenc->ch[3] & 0x80))
 265                         goto ilseq;
 266                 wchar = ((psenc->ch[2] & 0xff) << 8) | (psenc->ch[3] & 0xff);
 267                 wchar |= ('G' + psenc->ch[1] - 0xa1) << 24;
 268                 break;
 269         default:
 270                 goto ilseq;
 271         }
 272
 273         *s = s0;
 274         psenc->chlen = 0;
 275
 276         if (pwc)
 277                 *pwc = wchar;
 278
 279         if (!wchar)
 280                 *nresult = 0;
 281         else
 282                 *nresult = c - chlenbak;
 283
 284         return (0);
 285
 286 ilseq:
 287         psenc->chlen = 0;
 288         *nresult = (size_t)-1;
 289         return (EILSEQ);
 290
 291 restart:
 292         *s = s0;
 293         *nresult = (size_t)-1;
 294         return (0);
 295 }
 296
 297 static int
 298 _citrus_EUCTW_wcrtomb_priv(_EUCTWEncodingInfo * __restrict ei,
 299                            char * __restrict s, size_t n, wchar_t wc,
 300                            _EUCTWState * __restrict psenc,
 301                            size_t * __restrict nresult)
 302 {
 303         wchar_t cs = wc & 0x7f000080;
 304         wchar_t v;
 305         int i, len, clen, ret;
 306
 307         _DIAGASSERT(ei != NULL);
 308         _DIAGASSERT(nresult != 0);
 309         _DIAGASSERT(s != NULL);
 310
 311         clen = 1;
 312         if (wc & 0x00007f00)
 313                 clen = 2;
 314         if ((wc & 0x007f0000) && !(wc & 0x00800000))
 315                 clen = 3;
 316
 317         if (clen == 1 && cs == 0x00000000) {
 318                 /* ASCII */
 319                 len = 1;
 320                 if (n < len) {
 321                         ret = E2BIG;
 322                         goto err;
 323                 }
 324                 v = wc & 0x0000007f;
 325         } else if (clen == 2 && cs == ('G' << 24)) {
 326                 /* CNS-11643-1 */
 327                 len = 2;
 328                 if (n < len) {
 329                         ret = E2BIG;
 330                         goto err;
 331                 }
 332                 v = wc & 0x00007f7f;
 333                 v |= 0x00008080;
 334         } else if (clen == 2 && 'H' <= (cs >> 24) && (cs >> 24) <= 'M') {
 335                 /* CNS-11643-[2-7] */
 336                 len = 4;
 337                 if (n < len) {
 338                         ret = E2BIG;
 339                         goto err;
 340                 }
 341                 *s++ = _SS2;
 342                 *s++ = (cs >> 24) - 'H' + 0xa2;
 343                 v = wc & 0x00007f7f;
 344                 v |= 0x00008080;
 345         } else {
 346                 ret = EILSEQ;
 347                 goto err;
 348         }
 349
 350         i = clen;
 351         while (i-- > 0)
 352                 *s++ = (v >> (i << 3)) & 0xff;
 353
 354         *nresult = len;
 355         return 0;
 356
 357 err:
 358         *nresult = (size_t)-1;
 359         return ret;
 360 }
 361
 362 static __inline int
 363 /*ARGSUSED*/
 364 _citrus_EUCTW_stdenc_wctocs(_EUCTWEncodingInfo * __restrict ei,
 365                             _csid_t * __restrict csid,
 366                             _index_t * __restrict idx, wchar_t wc)
 367 {
 368
 369         _DIAGASSERT(ei != NULL && csid != NULL && idx != NULL);
 370
 371         *csid = (_csid_t)(wc >> 24) & 0xFF;
 372         *idx  = (_index_t)(wc & 0x7F7F);
 373
 374         return (0);
 375 }
 376
 377 static __inline int
 378 /*ARGSUSED*/
 379 _citrus_EUCTW_stdenc_cstowc(_EUCTWEncodingInfo * __restrict ei,
 380                             wchar_t * __restrict wc,
 381                             _csid_t csid, _index_t idx)
 382 {
 383
 384         _DIAGASSERT(ei != NULL && wc != NULL);
 385
 386         if (csid==0) {
 387                 if ((idx & ~0x7F) != 0)
 388                         return (EINVAL);
 389                 *wc = (wchar_t)idx;
 390         } else {
 391                 if (csid < 'G' || csid > 'M' || (idx & ~0x7F7F) != 0)
 392                         return (EINVAL);
 393                 *wc = (wchar_t)idx | ((wchar_t)csid<<24);
 394         }
 395
 396         return (0);
 397 }
 398
 399 static __inline int
 400 /*ARGSUSED*/
 401 _citrus_EUCTW_stdenc_get_state_desc_generic(_EUCTWEncodingInfo * __restrict ei,
 402                                             _EUCTWState * __restrict psenc,
 403                                             int * __restrict rstate)
 404 {
 405
 406         if (psenc->chlen == 0)
 407                 *rstate = _STDENC_SDGEN_INITIAL;
 408         else
 409                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 410
 411         return 0;
 412 }
 413
 414 /* ----------------------------------------------------------------------
 415  * public interface for ctype
 416  */
 417
 418 _CITRUS_CTYPE_DECLS(EUCTW);
 419 _CITRUS_CTYPE_DEF_OPS(EUCTW);
 420
 421 #include "citrus_ctype_template.h"
 422
 423 /* ----------------------------------------------------------------------
 424  * public interface for stdenc
 425  */
 426
 427 _CITRUS_STDENC_DECLS(EUCTW);
 428 _CITRUS_STDENC_DEF_OPS(EUCTW);
 429
 430 #include "citrus_stdenc_template.h"