lib/libc/citrus/modules/citrus_euctw.c

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