lib/libc/citrus/modules/citrus_euctw.c

   1 /* $NetBSD: citrus_euctw.c,v 1.10 2005/10/29 18:02:04 tshiozak Exp $ */
   2 /* $DragonFly: src/lib/libc/citrus/modules/citrus_euctw.c,v 1.2 2008/04/10 10:21:01 hasso Exp $ */
   3
   4 /*-
   5  * Copyright (c)2002 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)1999 Citrus Project,
  32  * All rights reserved.
  33  *
  34  * Redistribution and use in source and binary forms, with or without
  35  * modification, are permitted provided that the following conditions
  36  * are met:
  37  * 1. Redistributions of source code must retain the above copyright
  38  *    notice, this list of conditions and the following disclaimer.
  39  * 2. Redistributions in binary form must reproduce the above copyright
  40  *    notice, this list of conditions and the following disclaimer in the
  41  *    documentation and/or other materials provided with the distribution.
  42  *
  43  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  44  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  45  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  46  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  47  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  48  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  49  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  50  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  51  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  52  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  53  * SUCH DAMAGE.
  54  *
  55  *      $Citrus: xpg4dl/FreeBSD/lib/libc/locale/euctw.c,v 1.13 2001/06/21 01:51:44 yamt Exp $
  56  */
  57
  58 #include <sys/types.h>
  59 #include <assert.h>
  60 #include <errno.h>
  61 #include <limits.h>
  62 #include <locale.h>
  63 #include <stddef.h>
  64 #include <stdio.h>
  65 #include <stdlib.h>
  66 #include <string.h>
  67 #include <wchar.h>
  68
  69 #include "citrus_namespace.h"
  70 #include "citrus_types.h"
  71 #include "citrus_module.h"
  72 #include "citrus_ctype.h"
  73 #include "citrus_stdenc.h"
  74 #include "citrus_euctw.h"
  75
  76
  77 /* ----------------------------------------------------------------------
  78  * private stuffs used by templates
  79  */
  80
  81 typedef struct {
  82         char ch[4];
  83         int chlen;
  84 } _EUCTWState;
  85
  86 typedef struct {
  87         int dummy;
  88 } _EUCTWEncodingInfo;
  89 typedef struct {
  90         _EUCTWEncodingInfo      ei;
  91         struct {
  92                 /* for future multi-locale facility */
  93                 _EUCTWState     s_mblen;
  94                 _EUCTWState     s_mbrlen;
  95                 _EUCTWState     s_mbrtowc;
  96                 _EUCTWState     s_mbtowc;
  97                 _EUCTWState     s_mbsrtowcs;
  98                 _EUCTWState     s_wcrtomb;
  99                 _EUCTWState     s_wcsrtombs;
 100                 _EUCTWState     s_wctomb;
 101         } states;
 102 } _EUCTWCTypeInfo;
 103
 104 #define _SS2    0x008e
 105 #define _SS3    0x008f
 106
 107 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 108 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
 109
 110 #define _FUNCNAME(m)                    _citrus_EUCTW_##m
 111 #define _ENCODING_INFO                  _EUCTWEncodingInfo
 112 #define _CTYPE_INFO                     _EUCTWCTypeInfo
 113 #define _ENCODING_STATE                 _EUCTWState
 114 #define _ENCODING_MB_CUR_MAX(_ei_)      4
 115 #define _ENCODING_IS_STATE_DEPENDENT    0
 116 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 117
 118 static __inline int
 119 _citrus_EUCTW_cs(u_int c)
 120 {
 121         c &= 0xff;
 122
 123         return ((c & 0x80) ? (c == _SS2 ? 2 : 1) : 0);
 124 }
 125
 126 static __inline int
 127 _citrus_EUCTW_count(int cs)
 128 {
 129         switch (cs) {
 130         case 0:
 131                 return 1;
 132         case 1:
 133                 return 2;
 134         case 2:
 135                 return 4;
 136         case 3:
 137                 abort();
 138                 /*NOTREACHED*/
 139         }
 140         return 0;
 141 }
 142
 143 static __inline void
 144 /*ARGSUSED*/
 145 _citrus_EUCTW_init_state(_EUCTWEncodingInfo * __restrict ei,
 146                          _EUCTWState * __restrict s)
 147 {
 148         memset(s, 0, sizeof(*s));
 149 }
 150
 151 static __inline void
 152 /*ARGSUSED*/
 153 _citrus_EUCTW_pack_state(_EUCTWEncodingInfo * __restrict ei,
 154                          void * __restrict pspriv,
 155                          const _EUCTWState * __restrict s)
 156 {
 157         memcpy(pspriv, (const void *)s, sizeof(*s));
 158 }
 159
 160 static __inline void
 161 /*ARGSUSED*/
 162 _citrus_EUCTW_unpack_state(_EUCTWEncodingInfo * __restrict ei,
 163                            _EUCTWState * __restrict s,
 164                            const void * __restrict pspriv)
 165 {
 166         memcpy((void *)s, pspriv, sizeof(*s));
 167 }
 168
 169 static int
 170 /*ARGSUSED*/
 171 _citrus_EUCTW_encoding_module_init(_EUCTWEncodingInfo * __restrict ei,
 172                                    const void * __restrict var, size_t lenvar)
 173 {
 174
 175         _DIAGASSERT(ei != NULL);
 176
 177         memset((void *)ei, 0, sizeof(*ei));
 178
 179         return 0;
 180 }
 181
 182 static void
 183 /*ARGSUSED*/
 184 _citrus_EUCTW_encoding_module_uninit(_EUCTWEncodingInfo *ei)
 185 {
 186 }
 187
 188 static int
 189 _citrus_EUCTW_mbrtowc_priv(_EUCTWEncodingInfo * __restrict ei,
 190                            wchar_t * __restrict pwc,
 191                            const char ** __restrict s,
 192                            size_t n, _EUCTWState * __restrict psenc,
 193                            size_t * __restrict nresult)
 194 {
 195         wchar_t wchar;
 196         int c, cs;
 197         int chlenbak;
 198         const char *s0;
 199
 200         _DIAGASSERT(nresult != 0);
 201         _DIAGASSERT(ei != NULL);
 202         _DIAGASSERT(psenc != NULL);
 203         _DIAGASSERT(s != NULL);
 204
 205         s0 = *s;
 206
 207         if (s0 == NULL) {
 208                 _citrus_EUCTW_init_state(ei, psenc);
 209                 *nresult = 0; /* state independent */
 210                 return (0);
 211         }
 212
 213         chlenbak = psenc->chlen;
 214
 215         /* make sure we have the first byte in the buffer */
 216         switch (psenc->chlen) {
 217         case 0:
 218                 if (n < 1)
 219                         goto restart;
 220                 psenc->ch[0] = *s0++;
 221                 psenc->chlen = 1;
 222                 n--;
 223                 break;
 224         case 1:
 225         case 2:
 226                 break;
 227         default:
 228                 /* illgeal state */
 229                 goto ilseq;
 230         }
 231
 232         c = _citrus_EUCTW_count(cs = _citrus_EUCTW_cs(psenc->ch[0] & 0xff));
 233         if (c == 0)
 234                 goto ilseq;
 235         while (psenc->chlen < c) {
 236                 if (n < 1)
 237                         goto ilseq;
 238                 psenc->ch[psenc->chlen] = *s0++;
 239                 psenc->chlen++;
 240                 n--;
 241         }
 242
 243         wchar = 0;
 244         switch (cs) {
 245         case 0:
 246                 if (psenc->ch[0] & 0x80)
 247                         goto ilseq;
 248                 wchar = psenc->ch[0] & 0xff;
 249                 break;
 250         case 1:
 251                 if (!(psenc->ch[0] & 0x80) || !(psenc->ch[1] & 0x80))
 252                         goto ilseq;
 253                 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
 254                 wchar |= 'G' << 24;
 255                 break;
 256         case 2:
 257                 if ((u_char)psenc->ch[1] < 0xa1 || 0xa7 < (u_char)psenc->ch[1])
 258                         goto ilseq;
 259                 if (!(psenc->ch[2] & 0x80) || !(psenc->ch[3] & 0x80))
 260                         goto ilseq;
 261                 wchar = ((psenc->ch[2] & 0xff) << 8) | (psenc->ch[3] & 0xff);
 262                 wchar |= ('G' + psenc->ch[1] - 0xa1) << 24;
 263                 break;
 264         default:
 265                 goto ilseq;
 266         }
 267
 268         *s = s0;
 269         psenc->chlen = 0;
 270
 271         if (pwc)
 272                 *pwc = wchar;
 273
 274         if (!wchar)
 275                 *nresult = 0;
 276         else
 277                 *nresult = c - chlenbak;
 278
 279         return (0);
 280
 281 ilseq:
 282         psenc->chlen = 0;
 283         *nresult = (size_t)-1;
 284         return (EILSEQ);
 285
 286 restart:
 287         *s = s0;
 288         *nresult = (size_t)-1;
 289         return (0);
 290 }
 291
 292 static int
 293 _citrus_EUCTW_wcrtomb_priv(_EUCTWEncodingInfo * __restrict ei,
 294                            char * __restrict s, size_t n, wchar_t wc,
 295                            _EUCTWState * __restrict psenc,
 296                            size_t * __restrict nresult)
 297 {
 298         wchar_t cs = wc & 0x7f000080;
 299         wchar_t v;
 300         int i, len, clen, ret;
 301
 302         _DIAGASSERT(ei != NULL);
 303         _DIAGASSERT(nresult != 0);
 304         _DIAGASSERT(s != NULL);
 305
 306         clen = 1;
 307         if (wc & 0x00007f00)
 308                 clen = 2;
 309         if ((wc & 0x007f0000) && !(wc & 0x00800000))
 310                 clen = 3;
 311
 312         if (clen == 1 && cs == 0x00000000) {
 313                 /* ASCII */
 314                 len = 1;
 315                 if (n < len) {
 316                         ret = E2BIG;
 317                         goto err;
 318                 }
 319                 v = wc & 0x0000007f;
 320         } else if (clen == 2 && cs == ('G' << 24)) {
 321                 /* CNS-11643-1 */
 322                 len = 2;
 323                 if (n < len) {
 324                         ret = E2BIG;
 325                         goto err;
 326                 }
 327                 v = wc & 0x00007f7f;
 328                 v |= 0x00008080;
 329         } else if (clen == 2 && 'H' <= (cs >> 24) && (cs >> 24) <= 'M') {
 330                 /* CNS-11643-[2-7] */
 331                 len = 4;
 332                 if (n < len) {
 333                         ret = E2BIG;
 334                         goto err;
 335                 }
 336                 *s++ = _SS2;
 337                 *s++ = (cs >> 24) - 'H' + 0xa2;
 338                 v = wc & 0x00007f7f;
 339                 v |= 0x00008080;
 340         } else {
 341                 ret = EILSEQ;
 342                 goto err;
 343         }
 344
 345         i = clen;
 346         while (i-- > 0)
 347                 *s++ = (v >> (i << 3)) & 0xff;
 348
 349         *nresult = len;
 350         return 0;
 351
 352 err:
 353         *nresult = (size_t)-1;
 354         return ret;
 355 }
 356
 357 static __inline int
 358 /*ARGSUSED*/
 359 _citrus_EUCTW_stdenc_wctocs(_EUCTWEncodingInfo * __restrict ei,
 360                             _csid_t * __restrict csid,
 361                             _index_t * __restrict idx, wchar_t wc)
 362 {
 363
 364         _DIAGASSERT(ei != NULL && csid != NULL && idx != NULL);
 365
 366         *csid = (_csid_t)(wc >> 24) & 0xFF;
 367         *idx  = (_index_t)(wc & 0x7F7F);
 368
 369         return (0);
 370 }
 371
 372 static __inline int
 373 /*ARGSUSED*/
 374 _citrus_EUCTW_stdenc_cstowc(_EUCTWEncodingInfo * __restrict ei,
 375                             wchar_t * __restrict wc,
 376                             _csid_t csid, _index_t idx)
 377 {
 378
 379         _DIAGASSERT(ei != NULL && wc != NULL);
 380
 381         if (csid==0) {
 382                 if ((idx & ~0x7F) != 0)
 383                         return (EINVAL);
 384                 *wc = (wchar_t)idx;
 385         } else {
 386                 if (csid < 'G' || csid > 'M' || (idx & ~0x7F7F) != 0)
 387                         return (EINVAL);
 388                 *wc = (wchar_t)idx | ((wchar_t)csid<<24);
 389         }
 390
 391         return (0);
 392 }
 393
 394 static __inline int
 395 /*ARGSUSED*/
 396 _citrus_EUCTW_stdenc_get_state_desc_generic(_EUCTWEncodingInfo * __restrict ei,
 397                                             _EUCTWState * __restrict psenc,
 398                                             int * __restrict rstate)
 399 {
 400
 401         if (psenc->chlen == 0)
 402                 *rstate = _STDENC_SDGEN_INITIAL;
 403         else
 404                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 405
 406         return 0;
 407 }
 408
 409 /* ----------------------------------------------------------------------
 410  * public interface for ctype
 411  */
 412
 413 _CITRUS_CTYPE_DECLS(EUCTW);
 414 _CITRUS_CTYPE_DEF_OPS(EUCTW);
 415
 416 #include "citrus_ctype_template.h"
 417
 418 /* ----------------------------------------------------------------------
 419  * public interface for stdenc
 420  */
 421
 422 _CITRUS_STDENC_DECLS(EUCTW);
 423 _CITRUS_STDENC_DEF_OPS(EUCTW);
 424
 425 #include "citrus_stdenc_template.h"