lib/libc/citrus/modules/citrus_johab.c

   1 /* $NetBSD: citrus_johab.c,v 1.4 2008/06/14 16:01:07 tnozaki Exp $ */
   2
   3 /*-
   4  * Copyright (c)2006 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 #include <sys/cdefs.h>
  29 #if defined(LIBC_SCCS) && !defined(lint)
  30 __RCSID("$NetBSD: citrus_johab.c,v 1.4 2008/06/14 16:01:07 tnozaki Exp $");
  31 #endif /* LIBC_SCCS and not lint */
  32
  33 #include <sys/types.h>
  34 #include <assert.h>
  35 #include <errno.h>
  36 #include <string.h>
  37 #include <stdint.h>
  38 #include <stdio.h>
  39 #include <stdlib.h>
  40 #include <stddef.h>
  41 #include <wchar.h>
  42 #include <limits.h>
  43
  44 #include "citrus_namespace.h"
  45 #include "citrus_types.h"
  46 #include "citrus_bcs.h"
  47 #include "citrus_module.h"
  48 #include "citrus_ctype.h"
  49 #include "citrus_stdenc.h"
  50 #include "citrus_johab.h"
  51
  52 /* ----------------------------------------------------------------------
  53  * private stuffs used by templates
  54  */
  55
  56 typedef struct {
  57         int chlen;
  58         char ch[2];
  59 } _JOHABState;
  60
  61 typedef struct {
  62         int dummy;
  63 } _JOHABEncodingInfo;
  64
  65 typedef struct {
  66         _JOHABEncodingInfo      ei;
  67         struct {
  68                 /* for future multi-locale facility */
  69                 _JOHABState     s_mblen;
  70                 _JOHABState     s_mbrlen;
  71                 _JOHABState     s_mbrtowc;
  72                 _JOHABState     s_mbtowc;
  73                 _JOHABState     s_mbsrtowcs;
  74                 _JOHABState     s_wcrtomb;
  75                 _JOHABState     s_wcsrtombs;
  76                 _JOHABState     s_wctomb;
  77         } states;
  78 } _JOHABCTypeInfo;
  79
  80 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
  81 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
  82
  83 #define _FUNCNAME(m)                    _citrus_JOHAB_##m
  84 #define _ENCODING_INFO                  _JOHABEncodingInfo
  85 #define _CTYPE_INFO                     _JOHABCTypeInfo
  86 #define _ENCODING_STATE                 _JOHABState
  87 #define _ENCODING_MB_CUR_MAX(_ei_)              2
  88 #define _ENCODING_IS_STATE_DEPENDENT            0
  89 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
  90
  91
  92 static __inline void
  93 /*ARGSUSED*/
  94 _citrus_JOHAB_init_state(_JOHABEncodingInfo * __restrict ei,
  95         _JOHABState * __restrict psenc)
  96 {
  97         /* ei may be null */
  98         _DIAGASSERT(psenc != NULL);
  99
 100         psenc->chlen = 0;
 101 }
 102
 103 static __inline void
 104 /*ARGSUSED*/
 105 _citrus_JOHAB_pack_state(_JOHABEncodingInfo * __restrict ei,
 106         void * __restrict pspriv,
 107         const _JOHABState * __restrict psenc)
 108 {
 109         /* ei may be null */
 110         _DIAGASSERT(pspriv != NULL);
 111         _DIAGASSERT(psenc != NULL);
 112
 113         memcpy(pspriv, (const void *)psenc, sizeof(*psenc));
 114 }
 115
 116 static __inline void
 117 /*ARGSUSED*/
 118 _citrus_JOHAB_unpack_state(_JOHABEncodingInfo * __restrict ei,
 119         _JOHABState * __restrict psenc,
 120         const void * __restrict pspriv)
 121 {
 122         /* ei may be null */
 123         _DIAGASSERT(psenc != NULL);
 124         _DIAGASSERT(pspriv != NULL);
 125
 126         memcpy((void *)psenc, pspriv, sizeof(*psenc));
 127 }
 128
 129 static void
 130 /*ARGSUSED*/
 131 _citrus_JOHAB_encoding_module_uninit(_JOHABEncodingInfo *ei)
 132 {
 133         /* ei may be null */
 134 }
 135
 136 static int
 137 /*ARGSUSED*/
 138 _citrus_JOHAB_encoding_module_init(_JOHABEncodingInfo * __restrict ei,
 139         const void * __restrict var, size_t lenvar)
 140 {
 141         /* ei may be null */
 142         return 0;
 143 }
 144
 145 static __inline int
 146 ishangul(int l, int t)
 147 {
 148
 149         return (l >= 0x84 && l <= 0xD3) &&
 150               ((t >= 0x41 && t <= 0x7E) || (t >= 0x81 && t <= 0xFE));
 151 }
 152
 153 static __inline int
 154 isuda(int l, int t)
 155 {
 156         return (l == 0xD8) &&
 157                ((t >= 0x31 && t <= 0x7E) || (t >= 0x91 && t <= 0xFE));
 158 }
 159
 160 static __inline int
 161 ishanja(int l, int t)
 162 {
 163         return ((l >= 0xD9 && l <= 0xDE) || (l >= 0xE0 && l <= 0xF9)) &&
 164                ((t >= 0x31 && t <= 0x7E) || (t >= 0x91 && t <= 0xFE));
 165 }
 166
 167 static int
 168 /*ARGSUSED*/
 169 _citrus_JOHAB_mbrtowc_priv(_JOHABEncodingInfo * __restrict ei,
 170         wchar_t * __restrict pwc, const char ** __restrict s, size_t n,
 171         _JOHABState * __restrict psenc, size_t * __restrict nresult)
 172 {
 173         const char *s0;
 174         int l, t;
 175
 176         /* ei may be unused */
 177         _DIAGASSERT(s != NULL);
 178         _DIAGASSERT(psenc != NULL);
 179         _DIAGASSERT(nresult != 0);
 180
 181         if (*s == NULL) {
 182                 _citrus_JOHAB_init_state(ei, psenc);
 183                 *nresult = _ENCODING_IS_STATE_DEPENDENT;
 184                 return 0;
 185         }
 186         s0 = *s;
 187
 188         switch (psenc->chlen) {
 189         case 0:
 190                 if (n-- < 1)
 191                         goto restart;
 192                 l = *s0++ & 0xFF;
 193                 if (l <= 0x7F) {
 194                         if (pwc != NULL)
 195                                 *pwc = (wchar_t)l;
 196                         *nresult = (l == 0) ? 0 : 1;
 197                         *s = s0;
 198                         return 0;
 199                 }
 200                 psenc->ch[psenc->chlen++] = l;
 201                 break;
 202         case 1:
 203                 l = psenc->ch[0] & 0xFF;
 204                 break;
 205         default:
 206                 return EINVAL;
 207         }
 208         if (n-- < 1) {
 209 restart:
 210                 *nresult = (size_t)-2;
 211                 *s = s0;
 212                 return 0;
 213         }
 214         t = *s0++ & 0xFF;
 215         if (!ishangul(l, t) && !isuda(l, t) && !ishanja(l, t)) {
 216                 *nresult = (size_t)-1;
 217                 return EILSEQ;
 218         }
 219         if (pwc != NULL)
 220                 *pwc = (wchar_t)(l << 8 | t);
 221         *nresult = s0 - *s;
 222         *s = s0;
 223         psenc->chlen = 0;
 224
 225         return 0;
 226 }
 227
 228 static int
 229 /*ARGSUSED*/
 230 _citrus_JOHAB_wcrtomb_priv(_JOHABEncodingInfo * __restrict ei,
 231         char * __restrict s, size_t n, wchar_t wc,
 232         _JOHABState * __restrict psenc, size_t * __restrict nresult)
 233 {
 234         int l, t;
 235
 236         /* ei may be unused */
 237         _DIAGASSERT(s != NULL);
 238         _DIAGASSERT(psenc != NULL);
 239         _DIAGASSERT(nresult != NULL);
 240
 241         if (psenc->chlen != 0)
 242                 return EINVAL;
 243
 244         /* XXX assume wchar_t as int */
 245         if ((uint32_t)wc <= 0x7F) {
 246                 if (n < 1)
 247                         goto e2big;
 248                 *s = wc & 0xFF;
 249                 *nresult = 1;
 250         } else if ((uint32_t)wc <= 0xFFFF) {
 251                 if (n < 2) {
 252 e2big:
 253                         *nresult = (size_t)-1;
 254                         return E2BIG;
 255                 }
 256                 l = (wc >> 8) & 0xFF;
 257                 t = wc & 0xFF;
 258                 if (!ishangul(l, t) && !isuda(l, t) && !ishanja(l, t))
 259                         goto ilseq;
 260                 *s++ = l;
 261                 *s = t;
 262                 *nresult = 2;
 263         } else {
 264 ilseq:
 265                 *nresult = (size_t)-1;
 266                 return EILSEQ;
 267         }
 268         return 0;
 269
 270 }
 271
 272 static __inline int
 273 /*ARGSUSED*/
 274 _citrus_JOHAB_stdenc_wctocs(_JOHABEncodingInfo * __restrict ei,
 275         _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
 276 {
 277         int m, l, t, linear;
 278
 279         /* ei may be unused */
 280         _DIAGASSERT(csid != NULL);
 281         _DIAGASSERT(idx != NULL);
 282
 283         /* XXX assume wchar_t as int */
 284         if ((uint32_t)wc <= 0x7F) {
 285                 *idx = (_index_t)wc;
 286                 *csid = 0;
 287         } else if ((uint32_t)wc <= 0xFFFF) {
 288                 l = (wc >> 8) & 0xFF;
 289                 t = wc & 0xFF;
 290                 if (ishangul(l, t) || isuda(l, t)) {
 291                         *idx = (_index_t)wc;
 292                         *csid = 1;
 293                 } else {
 294                         if (l >= 0xD9 && l <= 0xDE) {
 295                                 linear = l - 0xD9;
 296                                 m = 0x21;
 297                         } else if (l >= 0xE0 && l <= 0xF9) {
 298                                 linear = l - 0xE0;
 299                                 m = 0x4A;
 300                         } else {
 301                                 return EILSEQ;
 302                         }
 303                         linear *= 188;
 304                         if (t >= 0x31 && t <= 0x7E) {
 305                                 linear += t - 0x31;
 306                         } else if (t >= 0x91 && t <= 0xFE) {
 307                                 linear += t - 0x43;
 308                         } else {
 309                                 return EILSEQ;
 310                         }
 311                         l = (linear / 94) + m;
 312                         t = (linear % 94) + 0x21;
 313                         *idx = (_index_t)((l << 8) | t);
 314                         *csid = 2;
 315                 }
 316         } else {
 317                 return EILSEQ;
 318         }
 319         return 0;
 320 }
 321
 322 static __inline int
 323 /*ARGSUSED*/
 324 _citrus_JOHAB_stdenc_cstowc(_JOHABEncodingInfo * __restrict ei,
 325         wchar_t * __restrict wc, _csid_t csid, _index_t idx)
 326 {
 327         int m, n, l, t, linear;
 328
 329         /* ei may be unused */
 330         _DIAGASSERT(wc != NULL);
 331
 332         switch (csid) {
 333         case 0:
 334         case 1:
 335                 *wc = (wchar_t)idx;
 336                 break;
 337         case 2:
 338                 if (idx >= 0x2121 && idx <= 0x2C71) {
 339                         m = 0xD9;
 340                         n = 0x21;
 341                 } else if (idx >= 0x4A21 && idx <= 0x7D7E) {
 342                         m = 0xE0;
 343                         n = 0x4A;
 344                 } else {
 345                         return EILSEQ;
 346                 }
 347                 l = ((idx >> 8) & 0xFF) - n;
 348                 t = (idx & 0xFF) - 0x21;
 349                 linear = (l * 94) + t;
 350                 l = (linear / 188) + m;
 351                 t = linear % 188;
 352                 t += (t <= 0x4D) ? 0x31 : 0x43;
 353                 break;
 354         default:
 355                 return EILSEQ;
 356         }
 357         return 0;
 358 }
 359
 360 static __inline int
 361 /*ARGSUSED*/
 362 _citrus_JOHAB_stdenc_get_state_desc_generic(_JOHABEncodingInfo * __restrict ei,
 363         _JOHABState * __restrict psenc, int * __restrict rstate)
 364 {
 365         /* ei may be unused */
 366         _DIAGASSERT(psenc != NULL);
 367         _DIAGASSERT(rstate != NULL);
 368
 369         *rstate = (psenc->chlen == 0)
 370             ? _STDENC_SDGEN_INITIAL
 371             : _STDENC_SDGEN_INCOMPLETE_CHAR;
 372         return 0;
 373 }
 374
 375 /* ----------------------------------------------------------------------
 376  * public interface for ctype
 377  */
 378
 379 _CITRUS_CTYPE_DECLS(JOHAB);
 380 _CITRUS_CTYPE_DEF_OPS(JOHAB);
 381
 382 #include "citrus_ctype_template.h"
 383
 384
 385 /* ----------------------------------------------------------------------
 386  * public interface for stdenc
 387  */
 388
 389 _CITRUS_STDENC_DECLS(JOHAB);
 390 _CITRUS_STDENC_DEF_OPS(JOHAB);
 391
 392 #include "citrus_stdenc_template.h"