lib/libc/citrus/modules/citrus_johab.c

   1 /* $NetBSD: citrus_johab.c,v 1.5 2013/05/28 16:57:56 joerg 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 1.5 2013/05/28 16:57:56 joerg 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_mbsnrtowcs;
  75                 _JOHABState     s_wcrtomb;
  76                 _JOHABState     s_wcsrtombs;
  77                 _JOHABState     s_wcsnrtombs;
  78                 _JOHABState     s_wctomb;
  79         } states;
  80 } _JOHABCTypeInfo;
  81
  82 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
  83 #define _CEI_TO_STATE(_cei_, _func_)    (_cei_)->states.s_##_func_
  84
  85 #define _FUNCNAME(m)                    _citrus_JOHAB_##m
  86 #define _ENCODING_INFO                  _JOHABEncodingInfo
  87 #define _CTYPE_INFO                     _JOHABCTypeInfo
  88 #define _ENCODING_STATE                 _JOHABState
  89 #define _ENCODING_MB_CUR_MAX(_ei_)              2
  90 #define _ENCODING_IS_STATE_DEPENDENT            0
  91 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
  92
  93
  94 static __inline void
  95 /*ARGSUSED*/
  96 _citrus_JOHAB_init_state(_JOHABEncodingInfo * __restrict ei,
  97         _JOHABState * __restrict psenc)
  98 {
  99         /* ei may be null */
 100         _DIAGASSERT(psenc != NULL);
 101
 102         psenc->chlen = 0;
 103 }
 104
 105 static __inline void
 106 /*ARGSUSED*/
 107 _citrus_JOHAB_pack_state(_JOHABEncodingInfo * __restrict ei,
 108         void * __restrict pspriv,
 109         const _JOHABState * __restrict psenc)
 110 {
 111         /* ei may be null */
 112         _DIAGASSERT(pspriv != NULL);
 113         _DIAGASSERT(psenc != NULL);
 114
 115         memcpy(pspriv, (const void *)psenc, sizeof(*psenc));
 116 }
 117
 118 static __inline void
 119 /*ARGSUSED*/
 120 _citrus_JOHAB_unpack_state(_JOHABEncodingInfo * __restrict ei,
 121         _JOHABState * __restrict psenc,
 122         const void * __restrict pspriv)
 123 {
 124         /* ei may be null */
 125         _DIAGASSERT(psenc != NULL);
 126         _DIAGASSERT(pspriv != NULL);
 127
 128         memcpy((void *)psenc, pspriv, sizeof(*psenc));
 129 }
 130
 131 static void
 132 /*ARGSUSED*/
 133 _citrus_JOHAB_encoding_module_uninit(_JOHABEncodingInfo *ei)
 134 {
 135         /* ei may be null */
 136 }
 137
 138 static int
 139 /*ARGSUSED*/
 140 _citrus_JOHAB_encoding_module_init(_JOHABEncodingInfo * __restrict ei,
 141         const void * __restrict var, size_t lenvar)
 142 {
 143         /* ei may be null */
 144         return 0;
 145 }
 146
 147 static __inline int
 148 ishangul(int l, int t)
 149 {
 150
 151         return (l >= 0x84 && l <= 0xD3) &&
 152               ((t >= 0x41 && t <= 0x7E) || (t >= 0x81 && t <= 0xFE));
 153 }
 154
 155 static __inline int
 156 isuda(int l, int t)
 157 {
 158         return (l == 0xD8) &&
 159                ((t >= 0x31 && t <= 0x7E) || (t >= 0x91 && t <= 0xFE));
 160 }
 161
 162 static __inline int
 163 ishanja(int l, int t)
 164 {
 165         return ((l >= 0xD9 && l <= 0xDE) || (l >= 0xE0 && l <= 0xF9)) &&
 166                ((t >= 0x31 && t <= 0x7E) || (t >= 0x91 && t <= 0xFE));
 167 }
 168
 169 static int
 170 /*ARGSUSED*/
 171 _citrus_JOHAB_mbrtowc_priv(_JOHABEncodingInfo * __restrict ei,
 172         wchar_t * __restrict pwc, const char ** __restrict s, size_t n,
 173         _JOHABState * __restrict psenc, size_t * __restrict nresult)
 174 {
 175         const char *s0;
 176         int l, t;
 177
 178         /* ei may be unused */
 179         _DIAGASSERT(s != NULL);
 180         _DIAGASSERT(psenc != NULL);
 181         _DIAGASSERT(nresult != 0);
 182
 183         if (*s == NULL) {
 184                 _citrus_JOHAB_init_state(ei, psenc);
 185                 *nresult = _ENCODING_IS_STATE_DEPENDENT;
 186                 return 0;
 187         }
 188         s0 = *s;
 189
 190         switch (psenc->chlen) {
 191         case 0:
 192                 if (n-- < 1)
 193                         goto restart;
 194                 l = *s0++ & 0xFF;
 195                 if (l <= 0x7F) {
 196                         if (pwc != NULL)
 197                                 *pwc = (wchar_t)l;
 198                         *nresult = (l == 0) ? 0 : 1;
 199                         *s = s0;
 200                         return 0;
 201                 }
 202                 psenc->ch[psenc->chlen++] = l;
 203                 break;
 204         case 1:
 205                 l = psenc->ch[0] & 0xFF;
 206                 break;
 207         default:
 208                 return EINVAL;
 209         }
 210         if (n-- < 1) {
 211 restart:
 212                 *nresult = (size_t)-2;
 213                 *s = s0;
 214                 return 0;
 215         }
 216         t = *s0++ & 0xFF;
 217         if (!ishangul(l, t) && !isuda(l, t) && !ishanja(l, t)) {
 218                 *nresult = (size_t)-1;
 219                 return EILSEQ;
 220         }
 221         if (pwc != NULL)
 222                 *pwc = (wchar_t)(l << 8 | t);
 223         *nresult = s0 - *s;
 224         *s = s0;
 225         psenc->chlen = 0;
 226
 227         return 0;
 228 }
 229
 230 static int
 231 /*ARGSUSED*/
 232 _citrus_JOHAB_wcrtomb_priv(_JOHABEncodingInfo * __restrict ei,
 233         char * __restrict s, size_t n, wchar_t wc,
 234         _JOHABState * __restrict psenc, size_t * __restrict nresult)
 235 {
 236         int l, t;
 237
 238         /* ei may be unused */
 239         _DIAGASSERT(s != NULL);
 240         _DIAGASSERT(psenc != NULL);
 241         _DIAGASSERT(nresult != NULL);
 242
 243         if (psenc->chlen != 0)
 244                 return EINVAL;
 245
 246         /* XXX assume wchar_t as int */
 247         if ((uint32_t)wc <= 0x7F) {
 248                 if (n < 1)
 249                         goto e2big;
 250                 *s = wc & 0xFF;
 251                 *nresult = 1;
 252         } else if ((uint32_t)wc <= 0xFFFF) {
 253                 if (n < 2) {
 254 e2big:
 255                         *nresult = (size_t)-1;
 256                         return E2BIG;
 257                 }
 258                 l = (wc >> 8) & 0xFF;
 259                 t = wc & 0xFF;
 260                 if (!ishangul(l, t) && !isuda(l, t) && !ishanja(l, t))
 261                         goto ilseq;
 262                 *s++ = l;
 263                 *s = t;
 264                 *nresult = 2;
 265         } else {
 266 ilseq:
 267                 *nresult = (size_t)-1;
 268                 return EILSEQ;
 269         }
 270         return 0;
 271
 272 }
 273
 274 static __inline int
 275 /*ARGSUSED*/
 276 _citrus_JOHAB_stdenc_wctocs(_JOHABEncodingInfo * __restrict ei,
 277         _csid_t * __restrict csid, _index_t * __restrict idx, wchar_t wc)
 278 {
 279         int m, l, t, linear;
 280
 281         /* ei may be unused */
 282         _DIAGASSERT(csid != NULL);
 283         _DIAGASSERT(idx != NULL);
 284
 285         /* XXX assume wchar_t as int */
 286         if ((uint32_t)wc <= 0x7F) {
 287                 *idx = (_index_t)wc;
 288                 *csid = 0;
 289         } else if ((uint32_t)wc <= 0xFFFF) {
 290                 l = (wc >> 8) & 0xFF;
 291                 t = wc & 0xFF;
 292                 if (ishangul(l, t) || isuda(l, t)) {
 293                         *idx = (_index_t)wc;
 294                         *csid = 1;
 295                 } else {
 296                         if (l >= 0xD9 && l <= 0xDE) {
 297                                 linear = l - 0xD9;
 298                                 m = 0x21;
 299                         } else if (l >= 0xE0 && l <= 0xF9) {
 300                                 linear = l - 0xE0;
 301                                 m = 0x4A;
 302                         } else {
 303                                 return EILSEQ;
 304                         }
 305                         linear *= 188;
 306                         if (t >= 0x31 && t <= 0x7E) {
 307                                 linear += t - 0x31;
 308                         } else if (t >= 0x91 && t <= 0xFE) {
 309                                 linear += t - 0x43;
 310                         } else {
 311                                 return EILSEQ;
 312                         }
 313                         l = (linear / 94) + m;
 314                         t = (linear % 94) + 0x21;
 315                         *idx = (_index_t)((l << 8) | t);
 316                         *csid = 2;
 317                 }
 318         } else {
 319                 return EILSEQ;
 320         }
 321         return 0;
 322 }
 323
 324 static __inline int
 325 /*ARGSUSED*/
 326 _citrus_JOHAB_stdenc_cstowc(_JOHABEncodingInfo * __restrict ei,
 327         wchar_t * __restrict wc, _csid_t csid, _index_t idx)
 328 {
 329         int m, n, l, t, linear;
 330
 331         /* ei may be unused */
 332         _DIAGASSERT(wc != NULL);
 333
 334         switch (csid) {
 335         case 0:
 336         case 1:
 337                 *wc = (wchar_t)idx;
 338                 break;
 339         case 2:
 340                 if (idx >= 0x2121 && idx <= 0x2C71) {
 341                         m = 0xD9;
 342                         n = 0x21;
 343                 } else if (idx >= 0x4A21 && idx <= 0x7D7E) {
 344                         m = 0xE0;
 345                         n = 0x4A;
 346                 } else {
 347                         return EILSEQ;
 348                 }
 349                 l = ((idx >> 8) & 0xFF) - n;
 350                 t = (idx & 0xFF) - 0x21;
 351                 linear = (l * 94) + t;
 352                 l = (linear / 188) + m;
 353                 t = linear % 188;
 354                 t += (t <= 0x4D) ? 0x31 : 0x43;
 355                 break;
 356         default:
 357                 return EILSEQ;
 358         }
 359         return 0;
 360 }
 361
 362 static __inline int
 363 /*ARGSUSED*/
 364 _citrus_JOHAB_stdenc_get_state_desc_generic(_JOHABEncodingInfo * __restrict ei,
 365         _JOHABState * __restrict psenc, int * __restrict rstate)
 366 {
 367         /* ei may be unused */
 368         _DIAGASSERT(psenc != NULL);
 369         _DIAGASSERT(rstate != NULL);
 370
 371         *rstate = (psenc->chlen == 0)
 372             ? _STDENC_SDGEN_INITIAL
 373             : _STDENC_SDGEN_INCOMPLETE_CHAR;
 374         return 0;
 375 }
 376
 377 /* ----------------------------------------------------------------------
 378  * public interface for ctype
 379  */
 380
 381 _CITRUS_CTYPE_DECLS(JOHAB);
 382 _CITRUS_CTYPE_DEF_OPS(JOHAB);
 383
 384 #include "citrus_ctype_template.h"
 385
 386
 387 /* ----------------------------------------------------------------------
 388  * public interface for stdenc
 389  */
 390
 391 _CITRUS_STDENC_DECLS(JOHAB);
 392 _CITRUS_STDENC_DEF_OPS(JOHAB);
 393
 394 #include "citrus_stdenc_template.h"