lib/libc/citrus/modules/citrus_mskanji.c

   1 /*      $NetBSD: citrus_mskanji.c,v 1.14 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  *    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 1.14 2013/05/28 16:57:56 joerg 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_mbsnrtowcs;
 111                 _MSKanjiState   s_wcrtomb;
 112                 _MSKanjiState   s_wcsrtombs;
 113                 _MSKanjiState   s_wcsnrtombs;
 114                 _MSKanjiState   s_wctomb;
 115         } states;
 116 } _MSKanjiCTypeInfo;
 117
 118 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 119 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
 120
 121 #define _FUNCNAME(m)                    _citrus_MSKanji_##m
 122 #define _ENCODING_INFO                  _MSKanjiEncodingInfo
 123 #define _CTYPE_INFO                     _MSKanjiCTypeInfo
 124 #define _ENCODING_STATE                 _MSKanjiState
 125 #define _ENCODING_MB_CUR_MAX(_ei_)      2
 126 #define _ENCODING_IS_STATE_DEPENDENT    0
 127 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 128
 129
 130 static int
 131 _mskanji1(int c)
 132 {
 133
 134         if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xfc))
 135                 return 1;
 136         else
 137                 return 0;
 138 }
 139
 140 static int
 141 _mskanji2(int c)
 142 {
 143
 144         if ((c >= 0x40 && c <= 0x7e) || (c >= 0x80 && c <= 0xfc))
 145                 return 1;
 146         else
 147                 return 0;
 148 }
 149
 150 static __inline void
 151 /*ARGSUSED*/
 152 _citrus_MSKanji_init_state(_MSKanjiEncodingInfo * __restrict ei,
 153                            _MSKanjiState * __restrict s)
 154 {
 155         s->chlen = 0;
 156 }
 157
 158 static __inline void
 159 /*ARGSUSED*/
 160 _citrus_MSKanji_pack_state(_MSKanjiEncodingInfo * __restrict ei,
 161                            void * __restrict pspriv,
 162                            const _MSKanjiState * __restrict s)
 163 {
 164         memcpy(pspriv, (const void *)s, sizeof(*s));
 165 }
 166
 167 static __inline void
 168 /*ARGSUSED*/
 169 _citrus_MSKanji_unpack_state(_MSKanjiEncodingInfo * __restrict ei,
 170                              _MSKanjiState * __restrict s,
 171                              const void * __restrict pspriv)
 172 {
 173         memcpy((void *)s, pspriv, sizeof(*s));
 174 }
 175
 176 static int
 177 /*ARGSUSED*/
 178 _citrus_MSKanji_mbrtowc_priv(_MSKanjiEncodingInfo * __restrict ei,
 179                              wchar_t * __restrict pwc,
 180                              const char ** __restrict s, size_t n,
 181                              _MSKanjiState * __restrict psenc,
 182                              size_t * __restrict nresult)
 183 {
 184         wchar_t wchar;
 185         int len;
 186         int chlenbak;
 187         const char *s0;
 188
 189         _DIAGASSERT(nresult != 0);
 190         _DIAGASSERT(ei != NULL);
 191         _DIAGASSERT(s != NULL);
 192         _DIAGASSERT(psenc != NULL);
 193
 194         s0 = *s;
 195
 196         if (s0 == NULL) {
 197                 _citrus_MSKanji_init_state(ei, psenc);
 198                 *nresult = 0; /* state independent */
 199                 return (0);
 200         }
 201
 202         chlenbak = psenc->chlen;
 203
 204         /* make sure we have the first byte in the buffer */
 205         switch (psenc->chlen) {
 206         case 0:
 207                 if (n < 1)
 208                         goto restart;
 209                 psenc->ch[0] = *s0++;
 210                 psenc->chlen = 1;
 211                 n--;
 212                 break;
 213         case 1:
 214                 break;
 215         default:
 216                 /* illegal state */
 217                 goto encoding_error;
 218         }
 219
 220         len = _mskanji1(psenc->ch[0] & 0xff) ? 2 : 1;
 221         while (psenc->chlen < len) {
 222                 if (n < 1)
 223                         goto restart;
 224                 psenc->ch[psenc->chlen] = *s0++;
 225                 psenc->chlen++;
 226                 n--;
 227         }
 228
 229         *s = s0;
 230
 231         switch (len) {
 232         case 1:
 233                 wchar = psenc->ch[0] & 0xff;
 234                 break;
 235         case 2:
 236                 if (!_mskanji2(psenc->ch[1] & 0xff))
 237                         goto encoding_error;
 238                 wchar = ((psenc->ch[0] & 0xff) << 8) | (psenc->ch[1] & 0xff);
 239                 break;
 240         default:
 241                 /* illegal state */
 242                 goto encoding_error;
 243         }
 244
 245         psenc->chlen = 0;
 246
 247         if (pwc)
 248                 *pwc = wchar;
 249
 250         if (!wchar)
 251                 *nresult = 0;
 252         else
 253                 *nresult = len - chlenbak;
 254
 255         return (0);
 256
 257 encoding_error:
 258         psenc->chlen = 0;
 259         *nresult = (size_t)-1;
 260         return (EILSEQ);
 261
 262 restart:
 263         *nresult = (size_t)-2;
 264         *s = s0;
 265         return (0);
 266 }
 267
 268
 269 static int
 270 _citrus_MSKanji_wcrtomb_priv(_MSKanjiEncodingInfo * __restrict ei,
 271                              char * __restrict s, size_t n, wchar_t wc,
 272                              _MSKanjiState * __restrict psenc,
 273                              size_t * __restrict nresult)
 274 {
 275         int ret;
 276
 277         _DIAGASSERT(ei != NULL);
 278         _DIAGASSERT(psenc != NULL);
 279         _DIAGASSERT(s != NULL);
 280
 281         /* check invalid sequence */
 282         if (wc & ~0xffff) {
 283                 ret = EILSEQ;
 284                 goto err;
 285         }
 286
 287         if (wc & 0xff00) {
 288                 if (n < 2) {
 289                         ret = E2BIG;
 290                         goto err;
 291                 }
 292
 293                 s[0] = (wc >> 8) & 0xff;
 294                 s[1] = wc & 0xff;
 295                 if (!_mskanji1(s[0] & 0xff) || !_mskanji2(s[1] & 0xff)) {
 296                         ret = EILSEQ;
 297                         goto err;
 298                 }
 299
 300                 *nresult = 2;
 301                 return 0;
 302         } else {
 303                 if (n < 1) {
 304                         ret = E2BIG;
 305                         goto err;
 306                 }
 307
 308                 s[0] = wc & 0xff;
 309                 if (_mskanji1(s[0] & 0xff)) {
 310                         ret = EILSEQ;
 311                         goto err;
 312                 }
 313
 314                 *nresult = 1;
 315                 return 0;
 316         }
 317
 318 err:
 319         *nresult = (size_t)-1;
 320         return ret;
 321 }
 322
 323
 324 static __inline int
 325 /*ARGSUSED*/
 326 _citrus_MSKanji_stdenc_wctocs(_MSKanjiEncodingInfo * __restrict ei,
 327                               _csid_t * __restrict csid,
 328                               _index_t * __restrict idx, wchar_t wc)
 329 {
 330         _index_t row, col;
 331         int offset;
 332
 333         _DIAGASSERT(csid != NULL && idx != NULL);
 334
 335         if ((_wc_t)wc < 0x80) {
 336                 /* ISO-646 */
 337                 *csid = 0;
 338                 *idx = (_index_t)wc;
 339         } else if ((_wc_t)wc < 0x100) {
 340                 /* KANA */
 341                 *csid = 1;
 342                 *idx = (_index_t)wc & 0x7F;
 343         } else {
 344                 /* Kanji (containing Gaiji zone) */
 345                 /*
 346                  * 94^2 zone (contains a part of Gaiji (0xED40 - 0xEEFC)):
 347                  * 0x8140 - 0x817E -> 0x2121 - 0x215F
 348                  * 0x8180 - 0x819E -> 0x2160 - 0x217E
 349                  * 0x819F - 0x81FC -> 0x2221 - 0x227E
 350                  *
 351                  * 0x8240 - 0x827E -> 0x2321 - 0x235F
 352                  *  ...
 353                  * 0x9F9F - 0x9FFc -> 0x5E21 - 0x5E7E
 354                  *
 355                  * 0xE040 - 0xE07E -> 0x5F21 - 0x5F5F
 356                  *  ...
 357                  * 0xEF9F - 0xEFFC -> 0x7E21 - 0x7E7E
 358                  *
 359                  * extended Gaiji zone:
 360                  * 0xF040 - 0xFCFC
 361                  *
 362                  * JIS X0213-plane2:
 363                  * 0xF040 - 0xF09E -> 0x2121 - 0x217E
 364                  * 0xF140 - 0xF19E -> 0x2321 - 0x237E
 365                  * ...
 366                  * 0xF240 - 0xF29E -> 0x2521 - 0x257E
 367                  *
 368                  * 0xF09F - 0xF0FC -> 0x2821 - 0x287E
 369                  * 0xF29F - 0xF2FC -> 0x2C21 - 0x2C7E
 370                  * ...
 371                  * 0xF44F - 0xF49E -> 0x2F21 - 0x2F7E
 372                  *
 373                  * 0xF49F - 0xF4FC -> 0x6E21 - 0x6E7E
 374                  * ...
 375                  * 0xFC9F - 0xFCFC -> 0x7E21 - 0x7E7E
 376                  */
 377                 row = ((_wc_t)wc >> 8) & 0xFF;
 378                 col = (_wc_t)wc & 0xFF;
 379                 if (!_mskanji1(row) || !_mskanji2(col))
 380                         return EILSEQ;
 381                 if ((ei->mode & MODE_JIS2004) == 0 || row < 0xF0) {
 382                         *csid = 2;
 383                         offset = 0x81;
 384                 } else {
 385                         *csid = 3;
 386                         if ((_wc_t)wc <= 0xF49E) {
 387                                 offset = (_wc_t)wc >= 0xF29F ||
 388                                   ((_wc_t)wc >= 0xF09F && (_wc_t)wc <= 0xF0FC)
 389                                     ? 0xED : 0xF0;
 390                         } else
 391                                 offset = 0xCE;
 392                 }
 393                 row -= offset;
 394                 if (row >= 0x5F)
 395                         row -= 0x40;
 396                 row = row * 2 + 0x21;
 397                 col -= 0x1F;
 398                 if (col >= 0x61)
 399                         col -= 1;
 400                 if (col > 0x7E) {
 401                         row += 1;
 402                         col -= 0x5E;
 403                 }
 404                 *idx = ((_index_t)row << 8) | col;
 405         }
 406
 407         return 0;
 408 }
 409
 410 static __inline int
 411 /*ARGSUSED*/
 412 _citrus_MSKanji_stdenc_cstowc(_MSKanjiEncodingInfo * __restrict ei,
 413                               wchar_t * __restrict wc,
 414                               _csid_t csid, _index_t idx)
 415 {
 416         u_int32_t row, col;
 417         int offset;
 418
 419         _DIAGASSERT(wc != NULL);
 420
 421         switch (csid) {
 422         case 0:
 423                 /* ISO-646 */
 424                 if (idx >= 0x80)
 425                         return EILSEQ;
 426                 *wc = (wchar_t)idx;
 427                 break;
 428         case 1:
 429                 /* kana */
 430                 if (idx >= 0x80)
 431                         return EILSEQ;
 432                 *wc = (wchar_t)idx + 0x80;
 433                 break;
 434         case 3:
 435                 if ((ei->mode & MODE_JIS2004) == 0)
 436                         return EILSEQ;
 437         /*FALLTHROUGH*/
 438         case 2:
 439                 /* kanji */
 440                 row = (idx >> 8);
 441                 if (row < 0x21)
 442                         return EILSEQ;
 443                 if (csid == 3) {
 444                         if (row <= 0x2F)
 445                                 offset = (row == 0x22 || row >= 0x26)
 446                                     ? 0xED : 0xF0;
 447                         else if (row >= 0x4D && row <= 0x7E)
 448                                 offset = 0xCE;
 449                         else
 450                                 return EILSEQ;
 451                 } else {
 452                         if (row > 0x97)
 453                                 return EILSEQ;
 454                         offset = (row < 0x5F) ? 0x81 : 0xC1;
 455                 }
 456                 col = idx & 0xFF;
 457                 if (col < 0x21 || col > 0x7E)
 458                         return EILSEQ;
 459                 row -= 0x21; col -= 0x21;
 460                 if ((row & 1) == 0) {
 461                         col += 0x40;
 462                         if (col >= 0x7F)
 463                                 col += 1;
 464                 } else
 465                         col += 0x9F;
 466                 row = row / 2 + offset;
 467                 *wc = ((wchar_t)row << 8) | col;
 468                 break;
 469         default:
 470                 return EILSEQ;
 471         }
 472
 473         return 0;
 474 }
 475
 476 static __inline int
 477 /*ARGSUSED*/
 478 _citrus_MSKanji_stdenc_get_state_desc_generic(_MSKanjiEncodingInfo * __restrict ei,
 479                                               _MSKanjiState * __restrict psenc,
 480                                               int * __restrict rstate)
 481 {
 482
 483         if (psenc->chlen == 0)
 484                 *rstate = _STDENC_SDGEN_INITIAL;
 485         else
 486                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 487
 488         return 0;
 489 }
 490
 491 static int
 492 /*ARGSUSED*/
 493 _citrus_MSKanji_encoding_module_init(_MSKanjiEncodingInfo *  __restrict ei,
 494                                      const void * __restrict var,
 495                                      size_t lenvar)
 496 {
 497         const char *p;
 498
 499         _DIAGASSERT(ei != NULL);
 500
 501         p = var;
 502 #define MATCH(x, act)                                           \
 503 do {                                                            \
 504         if (lenvar >= (sizeof(#x)-1) &&                         \
 505             _bcs_strncasecmp(p, #x, sizeof(#x)-1) == 0) {       \
 506                 act;                                            \
 507                 lenvar -= sizeof(#x)-1;                         \
 508                 p += sizeof(#x)-1;                              \
 509         }                                                       \
 510 } while (/*CONSTCOND*/0)
 511         memset((void *)ei, 0, sizeof(*ei));
 512         while (lenvar > 0) {
 513                 switch (_bcs_toupper(*p)) {
 514                 case 'J':
 515                         MATCH(JIS2004, ei->mode |= MODE_JIS2004);
 516                         break;
 517                 }
 518                 ++p;
 519                 --lenvar;
 520         }
 521
 522         return 0;
 523 }
 524
 525 static void
 526 _citrus_MSKanji_encoding_module_uninit(_MSKanjiEncodingInfo *ei)
 527 {
 528 }
 529
 530 /* ----------------------------------------------------------------------
 531  * public interface for ctype
 532  */
 533
 534 _CITRUS_CTYPE_DECLS(MSKanji);
 535 _CITRUS_CTYPE_DEF_OPS(MSKanji);
 536
 537 #include "citrus_ctype_template.h"
 538
 539 /* ----------------------------------------------------------------------
 540  * public interface for stdenc
 541  */
 542
 543 _CITRUS_STDENC_DECLS(MSKanji);
 544 _CITRUS_STDENC_DEF_OPS(MSKanji);
 545
 546 #include "citrus_stdenc_template.h"