lib/libc/citrus/modules/citrus_mskanji.c

   1 /*      $NetBSD: citrus_mskanji.c,v 1.13 2008/06/14 16:01:08 tnozaki 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  *    ja_JP.SJIS locale table for BSD4.4/rune
  31  *    version 1.0
  32  *    (C) Sin'ichiro MIYATANI / Phase One, Inc
  33  *    May 12, 1995
  34  *
  35  * Redistribution and use in source and binary forms, with or without
  36  * modification, are permitted provided that the following conditions
  37  * are met:
  38  * 1. Redistributions of source code must retain the above copyright
  39  *    notice, this list of conditions and the following disclaimer.
  40  * 2. Redistributions in binary form must reproduce the above copyright
  41  *    notice, this list of conditions and the following disclaimer in the
  42  *    documentation and/or other materials provided with the distribution.
  43  * 3. All advertising materials mentioning features or use of this software
  44  *    must display the following acknowledgement:
  45  *      This product includes software developed by Phase One, Inc.
  46  * 4. The name of Phase One, Inc. may be used to endorse or promote products
  47  *    derived from this software 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
  63 #include <sys/cdefs.h>
  64 #if defined(LIBC_SCCS) && !defined(lint)
  65 __RCSID("$NetBSD: citrus_mskanji.c,v 1.13 2008/06/14 16:01:08 tnozaki Exp $");
  66 #endif /* LIBC_SCCS and not lint */
  67
  68 #include <assert.h>
  69 #include <errno.h>
  70 #include <string.h>
  71 #include <stdio.h>
  72 #include <stdlib.h>
  73 #include <stddef.h>
  74 #include <wchar.h>
  75 #include <sys/types.h>
  76 #include <limits.h>
  77
  78 #include "citrus_namespace.h"
  79 #include "citrus_types.h"
  80 #include "citrus_bcs.h"
  81 #include "citrus_module.h"
  82 #include "citrus_ctype.h"
  83 #include "citrus_stdenc.h"
  84 #include "citrus_mskanji.h"
  85
  86
  87 /* ----------------------------------------------------------------------
  88  * private stuffs used by templates
  89  */
  90
  91 typedef struct _MSKanjiState {
  92         char ch[2];
  93         int chlen;
  94 } _MSKanjiState;
  95
  96 typedef struct {
  97         int mode;
  98 #define MODE_JIS2004    1
  99 } _MSKanjiEncodingInfo;
 100
 101 typedef struct {
 102         _MSKanjiEncodingInfo    ei;
 103         struct {
 104                 /* for future multi-locale facility */
 105                 _MSKanjiState   s_mblen;
 106                 _MSKanjiState   s_mbrlen;
 107                 _MSKanjiState   s_mbrtowc;
 108                 _MSKanjiState   s_mbtowc;
 109                 _MSKanjiState   s_mbsrtowcs;
 110                 _MSKanjiState   s_wcrtomb;
 111                 _MSKanjiState   s_wcsrtombs;
 112                 _MSKanjiState   s_wctomb;
 113         } states;
 114 } _MSKanjiCTypeInfo;
 115
 116 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 117 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
 118
 119 #define _FUNCNAME(m)                    _citrus_MSKanji_##m
 120 #define _ENCODING_INFO                  _MSKanjiEncodingInfo
 121 #define _CTYPE_INFO                     _MSKanjiCTypeInfo
 122 #define _ENCODING_STATE                 _MSKanjiState
 123 #define _ENCODING_MB_CUR_MAX(_ei_)      2
 124 #define _ENCODING_IS_STATE_DEPENDENT    0
 125 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 126
 127
 128 static int
 129 _mskanji1(int c)
 130 {
 131
 132         if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xfc))
 133                 return 1;
 134         else
 135                 return 0;
 136 }
 137
 138 static int
 139 _mskanji2(int c)
 140 {
 141
 142         if ((c >= 0x40 && c <= 0x7e) || (c >= 0x80 && c <= 0xfc))
 143                 return 1;
 144         else
 145                 return 0;
 146 }
 147
 148 static __inline void
 149 /*ARGSUSED*/
 150 _citrus_MSKanji_init_state(_MSKanjiEncodingInfo * __restrict ei,
 151                            _MSKanjiState * __restrict s)
 152 {
 153         s->chlen = 0;
 154 }
 155
 156 static __inline void
 157 /*ARGSUSED*/
 158 _citrus_MSKanji_pack_state(_MSKanjiEncodingInfo * __restrict ei,
 159                            void * __restrict pspriv,
 160                            const _MSKanjiState * __restrict s)
 161 {
 162         memcpy(pspriv, (const void *)s, sizeof(*s));
 163 }
 164
 165 static __inline void
 166 /*ARGSUSED*/
 167 _citrus_MSKanji_unpack_state(_MSKanjiEncodingInfo * __restrict ei,
 168                              _MSKanjiState * __restrict s,
 169                              const void * __restrict pspriv)
 170 {
 171         memcpy((void *)s, pspriv, sizeof(*s));
 172 }
 173
 174 static int
 175 /*ARGSUSED*/
 176 _citrus_MSKanji_mbrtowc_priv(_MSKanjiEncodingInfo * __restrict ei,
 177                              wchar_t * __restrict pwc,
 178                              const char ** __restrict s, size_t n,
 179                              _MSKanjiState * __restrict psenc,
 180                              size_t * __restrict nresult)
 181 {
 182         wchar_t wchar;
 183         int len;
 184         int chlenbak;
 185         const char *s0;
 186
 187         _DIAGASSERT(nresult != 0);
 188         _DIAGASSERT(ei != NULL);
 189         _DIAGASSERT(s != NULL);
 190         _DIAGASSERT(psenc != NULL);
 191
 192         s0 = *s;
 193
 194         if (s0 == NULL) {
 195                 _citrus_MSKanji_init_state(ei, psenc);
 196                 *nresult = 0; /* state independent */
 197                 return (0);
 198         }
 199
 200         chlenbak = psenc->chlen;
 201
 202         /* make sure we have the first byte in the buffer */
 203         switch (psenc->chlen) {
 204         case 0:
 205                 if (n < 1)
 206                         goto restart;
 207                 psenc->ch[0] = *s0++;
 208                 psenc->chlen = 1;
 209                 n--;
 210                 break;
 211         case 1:
 212                 break;
 213         default:
 214                 /* illegal state */
 215                 goto encoding_error;
 216         }
 217
 218         len = _mskanji1(psenc->ch[0] & 0xff) ? 2 : 1;
 219         while (psenc->chlen < len) {
 220                 if (n < 1)
 221                         goto restart;
 222                 psenc->ch[psenc->chlen] = *s0++;
 223                 psenc->chlen++;
 224                 n--;
 225         }
 226
 227         *s = s0;
 228
 229         switch (len) {
 230         case 1:
 231                 wchar = psenc->ch[0] & 0xff;
 232                 break;
 233         case 2:
 234                 if (!_mskanji2(psenc->ch[1] & 0xff))
 235                         goto encoding_error;
 236                 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
 237                 break;
 238         default:
 239                 /* illegal state */
 240                 goto encoding_error;
 241         }
 242
 243         psenc->chlen = 0;
 244
 245         if (pwc)
 246                 *pwc = wchar;
 247
 248         if (!wchar)
 249                 *nresult = 0;
 250         else
 251                 *nresult = len - chlenbak;
 252
 253         return (0);
 254
 255 encoding_error:
 256         psenc->chlen = 0;
 257         *nresult = (size_t)-1;
 258         return (EILSEQ);
 259
 260 restart:
 261         *nresult = (size_t)-2;
 262         *s = s0;
 263         return (0);
 264 }
 265
 266
 267 static int
 268 _citrus_MSKanji_wcrtomb_priv(_MSKanjiEncodingInfo * __restrict ei,
 269                              char * __restrict s, size_t n, wchar_t wc,
 270                              _MSKanjiState * __restrict psenc,
 271                              size_t * __restrict nresult)
 272 {
 273         int ret;
 274
 275         _DIAGASSERT(ei != NULL);
 276         _DIAGASSERT(psenc != NULL);
 277         _DIAGASSERT(s != NULL);
 278
 279         /* check invalid sequence */
 280         if (wc & ~0xffff) {
 281                 ret = EILSEQ;
 282                 goto err;
 283         }
 284
 285         if (wc & 0xff00) {
 286                 if (n < 2) {
 287                         ret = E2BIG;
 288                         goto err;
 289                 }
 290
 291                 s[0] = (wc >> 8) & 0xff;
 292                 s[1] = wc & 0xff;
 293                 if (!_mskanji1(s[0] & 0xff) || !_mskanji2(s[1] & 0xff)) {
 294                         ret = EILSEQ;
 295                         goto err;
 296                 }
 297
 298                 *nresult = 2;
 299                 return 0;
 300         } else {
 301                 if (n < 1) {
 302                         ret = E2BIG;
 303                         goto err;
 304                 }
 305
 306                 s[0] = wc & 0xff;
 307                 if (_mskanji1(s[0] & 0xff)) {
 308                         ret = EILSEQ;
 309                         goto err;
 310                 }
 311
 312                 *nresult = 1;
 313                 return 0;
 314         }
 315
 316 err:
 317         *nresult = (size_t)-1;
 318         return ret;
 319 }
 320
 321
 322 static __inline int
 323 /*ARGSUSED*/
 324 _citrus_MSKanji_stdenc_wctocs(_MSKanjiEncodingInfo * __restrict ei,
 325                               _csid_t * __restrict csid,
 326                               _index_t * __restrict idx, wchar_t wc)
 327 {
 328         _index_t row, col;
 329         int offset;
 330
 331         _DIAGASSERT(csid != NULL && idx != NULL);
 332
 333         if ((_wc_t)wc < 0x80) {
 334                 /* ISO-646 */
 335                 *csid = 0;
 336                 *idx = (_index_t)wc;
 337         } else if ((_wc_t)wc < 0x100) {
 338                 /* KANA */
 339                 *csid = 1;
 340                 *idx = (_index_t)wc & 0x7F;
 341         } else {
 342                 /* Kanji (containing Gaiji zone) */
 343                 /*
 344                  * 94^2 zone (contains a part of Gaiji (0xED40 - 0xEEFC)):
 345                  * 0x8140 - 0x817E -> 0x2121 - 0x215F
 346                  * 0x8180 - 0x819E -> 0x2160 - 0x217E
 347                  * 0x819F - 0x81FC -> 0x2221 - 0x227E
 348                  *
 349                  * 0x8240 - 0x827E -> 0x2321 - 0x235F
 350                  *  ...
 351                  * 0x9F9F - 0x9FFc -> 0x5E21 - 0x5E7E
 352                  *
 353                  * 0xE040 - 0xE07E -> 0x5F21 - 0x5F5F
 354                  *  ...
 355                  * 0xEF9F - 0xEFFC -> 0x7E21 - 0x7E7E
 356                  *
 357                  * extended Gaiji zone:
 358                  * 0xF040 - 0xFCFC
 359                  *
 360                  * JIS X0213-plane2:
 361                  * 0xF040 - 0xF09E -> 0x2121 - 0x217E
 362                  * 0xF140 - 0xF19E -> 0x2321 - 0x237E
 363                  * ...
 364                  * 0xF240 - 0xF29E -> 0x2521 - 0x257E
 365                  *
 366                  * 0xF09F - 0xF0FC -> 0x2821 - 0x287E
 367                  * 0xF29F - 0xF2FC -> 0x2C21 - 0x2C7E
 368                  * ...
 369                  * 0xF44F - 0xF49E -> 0x2F21 - 0x2F7E
 370                  *
 371                  * 0xF49F - 0xF4FC -> 0x6E21 - 0x6E7E
 372                  * ...
 373                  * 0xFC9F - 0xFCFC -> 0x7E21 - 0x7E7E
 374                  */
 375                 row = ((_wc_t)wc >> 8) & 0xFF;
 376                 col = (_wc_t)wc & 0xFF;
 377                 if (!_mskanji1(row) || !_mskanji2(col))
 378                         return EILSEQ;
 379                 if ((ei->mode & MODE_JIS2004) == 0 || row < 0xF0) {
 380                         *csid = 2;
 381                         offset = 0x81;
 382                 } else {
 383                         *csid = 3;
 384                         if ((_wc_t)wc <= 0xF49E) {
 385                                 offset = (_wc_t)wc >= 0xF29F ||
 386                                   ((_wc_t)wc >= 0xF09F && (_wc_t)wc <= 0xF0FC)
 387                                     ? 0xED : 0xF0;
 388                         } else
 389                                 offset = 0xCE;
 390                 }
 391                 row -= offset;
 392                 if (row >= 0x5F)
 393                         row -= 0x40;
 394                 row = row * 2 + 0x21;
 395                 col -= 0x1F;
 396                 if (col >= 0x61)
 397                         col -= 1;
 398                 if (col > 0x7E) {
 399                         row += 1;
 400                         col -= 0x5E;
 401                 }
 402                 *idx = ((_index_t)row << 8) | col;
 403         }
 404
 405         return 0;
 406 }
 407
 408 static __inline int
 409 /*ARGSUSED*/
 410 _citrus_MSKanji_stdenc_cstowc(_MSKanjiEncodingInfo * __restrict ei,
 411                               wchar_t * __restrict wc,
 412                               _csid_t csid, _index_t idx)
 413 {
 414         u_int32_t row, col;
 415         int offset;
 416
 417         _DIAGASSERT(wc != NULL);
 418
 419         switch (csid) {
 420         case 0:
 421                 /* ISO-646 */
 422                 if (idx >= 0x80)
 423                         return EILSEQ;
 424                 *wc = (wchar_t)idx;
 425                 break;
 426         case 1:
 427                 /* kana */
 428                 if (idx >= 0x80)
 429                         return EILSEQ;
 430                 *wc = (wchar_t)idx + 0x80;
 431                 break;
 432         case 3:
 433                 if ((ei->mode & MODE_JIS2004) == 0)
 434                         return EILSEQ;
 435         /*FALLTHROUGH*/
 436         case 2:
 437                 /* kanji */
 438                 row = (idx >> 8);
 439                 if (row < 0x21)
 440                         return EILSEQ;
 441                 if (csid == 3) {
 442                         if (row <= 0x2F)
 443                                 offset = (row == 0x22 || row >= 0x26)
 444                                     ? 0xED : 0xF0;
 445                         else if (row >= 0x4D && row <= 0x7E)
 446                                 offset = 0xCE;
 447                         else
 448                                 return EILSEQ;
 449                 } else {
 450                         if (row > 0x97)
 451                                 return EILSEQ;
 452                         offset = (row < 0x5F) ? 0x81 : 0xC1;
 453                 }
 454                 col = idx & 0xFF;
 455                 if (col < 0x21 || col > 0x7E)
 456                         return EILSEQ;
 457                 row -= 0x21; col -= 0x21;
 458                 if ((row & 1) == 0) {
 459                         col += 0x40;
 460                         if (col >= 0x7F)
 461                                 col += 1;
 462                 } else
 463                         col += 0x9F;
 464                 row = row / 2 + offset;
 465                 *wc = ((wchar_t)row << 8) | col;
 466                 break;
 467         default:
 468                 return EILSEQ;
 469         }
 470
 471         return 0;
 472 }
 473
 474 static __inline int
 475 /*ARGSUSED*/
 476 _citrus_MSKanji_stdenc_get_state_desc_generic(_MSKanjiEncodingInfo * __restrict ei,
 477                                               _MSKanjiState * __restrict psenc,
 478                                               int * __restrict rstate)
 479 {
 480
 481         if (psenc->chlen == 0)
 482                 *rstate = _STDENC_SDGEN_INITIAL;
 483         else
 484                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 485
 486         return 0;
 487 }
 488
 489 static int
 490 /*ARGSUSED*/
 491 _citrus_MSKanji_encoding_module_init(_MSKanjiEncodingInfo *  __restrict ei,
 492                                      const void * __restrict var,
 493                                      size_t lenvar)
 494 {
 495         const char *p;
 496
 497         _DIAGASSERT(ei != NULL);
 498
 499         p = var;
 500 #define MATCH(x, act)                                           \
 501 do {                                                            \
 502         if (lenvar >= (sizeof(#x)-1) &&                         \
 503             _bcs_strncasecmp(p, #x, sizeof(#x)-1) == 0) {       \
 504                 act;                                            \
 505                 lenvar -= sizeof(#x)-1;                         \
 506                 p += sizeof(#x)-1;                              \
 507         }                                                       \
 508 } while (/*CONSTCOND*/0)
 509         memset((void *)ei, 0, sizeof(*ei));
 510         while (lenvar > 0) {
 511                 switch (_bcs_toupper(*p)) {
 512                 case 'J':
 513                         MATCH(JIS2004, ei->mode |= MODE_JIS2004);
 514                         break;
 515                 }
 516                 ++p;
 517                 --lenvar;
 518         }
 519
 520         return 0;
 521 }
 522
 523 static void
 524 _citrus_MSKanji_encoding_module_uninit(_MSKanjiEncodingInfo *ei)
 525 {
 526 }
 527
 528 /* ----------------------------------------------------------------------
 529  * public interface for ctype
 530  */
 531
 532 _CITRUS_CTYPE_DECLS(MSKanji);
 533 _CITRUS_CTYPE_DEF_OPS(MSKanji);
 534
 535 #include "citrus_ctype_template.h"
 536
 537 /* ----------------------------------------------------------------------
 538  * public interface for stdenc
 539  */
 540
 541 _CITRUS_STDENC_DECLS(MSKanji);
 542 _CITRUS_STDENC_DEF_OPS(MSKanji);
 543
 544 #include "citrus_stdenc_template.h"