lib/libc/citrus/modules/citrus_utf8.c

   1 /*      $NetBSD: citrus_utf8.c,v 1.16 2007/03/06 16:13:58 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  * Copyright (c) 1993
  31  *      The Regents of the University of California.  All rights reserved.
  32  *
  33  * This code is derived from software contributed to Berkeley by
  34  * Paul Borman at Krystal Technologies.
  35  *
  36  * Redistribution and use in source and binary forms, with or without
  37  * modification, are permitted provided that the following conditions
  38  * are met:
  39  * 1. Redistributions of source code must retain the above copyright
  40  *    notice, this list of conditions and the following disclaimer.
  41  * 2. Redistributions in binary form must reproduce the above copyright
  42  *    notice, this list of conditions and the following disclaimer in the
  43  *    documentation and/or other materials provided with the distribution.
  44  * 3. Neither the name of the University nor the names of its contributors
  45  *    may be used to endorse or promote products derived from this software
  46  *    without specific prior written permission.
  47  *
  48  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  49  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  50  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  51  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  52  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  53  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  54  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  55  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  56  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  57  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  58  * SUCH DAMAGE.
  59  */
  60
  61 #include <sys/cdefs.h>
  62 #if defined(LIBC_SCCS) && !defined(lint)
  63 __RCSID("$NetBSD: citrus_utf8.c,v 1.16 2007/03/06 16:13:58 tnozaki Exp $");
  64 #endif /* LIBC_SCCS and not lint */
  65
  66 #include <assert.h>
  67 #include <errno.h>
  68 #include <string.h>
  69 #include <stdio.h>
  70 #include <stdlib.h>
  71 #include <stddef.h>
  72 #include <wchar.h>
  73 #include <sys/types.h>
  74 #include <limits.h>
  75
  76 #include "citrus_namespace.h"
  77 #include "citrus_types.h"
  78 #include "citrus_module.h"
  79 #include "citrus_ctype.h"
  80 #include "citrus_stdenc.h"
  81 #include "citrus_utf8.h"
  82
  83
  84 /* ----------------------------------------------------------------------
  85  * private stuffs used by templates
  86  */
  87
  88 static int _UTF8_count_array[256];
  89 static int const *_UTF8_count = NULL;
  90
  91 static const u_int32_t _UTF8_range[] = {
  92         0,      /*dummy*/
  93         0x00000000, 0x00000080, 0x00000800, 0x00010000,
  94         0x00200000, 0x04000000, 0x80000000,
  95 };
  96
  97 typedef struct {
  98         char ch[6];
  99         int chlen;
 100 } _UTF8State;
 101
 102 typedef struct {
 103 } _UTF8EncodingInfo;
 104
 105 typedef struct {
 106         _UTF8EncodingInfo       ei;
 107         struct {
 108                 /* for future multi-locale facility */
 109                 _UTF8State      s_mblen;
 110                 _UTF8State      s_mbrlen;
 111                 _UTF8State      s_mbrtowc;
 112                 _UTF8State      s_mbtowc;
 113                 _UTF8State      s_mbsrtowcs;
 114                 _UTF8State      s_wcrtomb;
 115                 _UTF8State      s_wcsrtombs;
 116                 _UTF8State      s_wctomb;
 117         } states;
 118 } _UTF8CTypeInfo;
 119
 120 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 121 #define _CEI_TO_STATE(_ei_, _func_)     (_ei_)->states.s_##_func_
 122
 123 #define _FUNCNAME(m)                    _citrus_UTF8_##m
 124 #define _ENCODING_INFO                  _UTF8EncodingInfo
 125 #define _CTYPE_INFO                     _UTF8CTypeInfo
 126 #define _ENCODING_STATE                 _UTF8State
 127 #define _ENCODING_MB_CUR_MAX(_ei_)      6
 128 #define _ENCODING_IS_STATE_DEPENDENT    0
 129 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 130
 131
 132 static __inline void
 133 _UTF8_init_count(void)
 134 {
 135         int i;
 136         if (!_UTF8_count) {
 137                 memset(_UTF8_count_array, 0, sizeof(_UTF8_count_array));
 138                 for (i = 0; i <= 0x7f; i++)
 139                         _UTF8_count_array[i] = 1;
 140                 for (i = 0xc0; i <= 0xdf; i++)
 141                         _UTF8_count_array[i] = 2;
 142                 for (i = 0xe0; i <= 0xef; i++)
 143                         _UTF8_count_array[i] = 3;
 144                 for (i = 0xf0; i <= 0xf7; i++)
 145                         _UTF8_count_array[i] = 4;
 146                 for (i = 0xf8; i <= 0xfb; i++)
 147                         _UTF8_count_array[i] = 5;
 148                 for (i = 0xfc; i <= 0xfd; i++)
 149                         _UTF8_count_array[i] = 6;
 150                 _UTF8_count = _UTF8_count_array;
 151         }
 152 }
 153
 154 static int
 155 _UTF8_findlen(wchar_t v)
 156 {
 157         int i;
 158         u_int32_t c;
 159
 160         c = (u_int32_t)v;       /*XXX*/
 161         for (i = 1; i < sizeof(_UTF8_range) / sizeof(_UTF8_range[0]) - 1; i++)
 162                 if (c >= _UTF8_range[i] && c < _UTF8_range[i + 1])
 163                         return i;
 164
 165         return -1;      /*out of range*/
 166 }
 167
 168 static __inline int
 169 _UTF8_surrogate(wchar_t wc)
 170 {
 171         return wc >= 0xd800 && wc <= 0xdfff;
 172 }
 173
 174 static __inline void
 175 /*ARGSUSED*/
 176 _citrus_UTF8_init_state(_UTF8EncodingInfo *ei, _UTF8State *s)
 177 {
 178         s->chlen = 0;
 179 }
 180
 181 static __inline void
 182 /*ARGSUSED*/
 183 _citrus_UTF8_pack_state(_UTF8EncodingInfo *ei, void *pspriv,
 184                         const _UTF8State *s)
 185 {
 186         memcpy(pspriv, (const void *)s, sizeof(*s));
 187 }
 188
 189 static __inline void
 190 /*ARGSUSED*/
 191 _citrus_UTF8_unpack_state(_UTF8EncodingInfo *ei, _UTF8State *s,
 192                           const void *pspriv)
 193 {
 194         memcpy((void *)s, pspriv, sizeof(*s));
 195 }
 196
 197 static int
 198 _citrus_UTF8_mbrtowc_priv(_UTF8EncodingInfo *ei, wchar_t *pwc, const char **s,
 199                           size_t n, _UTF8State *psenc, size_t *nresult)
 200 {
 201         wchar_t wchar;
 202         const char *s0;
 203         int c;
 204         int i;
 205
 206         _DIAGASSERT(nresult != 0);
 207         _DIAGASSERT(s != NULL);
 208         _DIAGASSERT(psenc != NULL);
 209
 210         s0 = *s;
 211
 212         if (s0 == NULL) {
 213                 _citrus_UTF8_init_state(ei, psenc);
 214                 *nresult = 0; /* state independent */
 215                 return 0;
 216         }
 217
 218         /* make sure we have the first byte in the buffer */
 219         if (psenc->chlen == 0) {
 220                 if (n-- < 1)
 221                         goto restart;
 222                 psenc->ch[psenc->chlen++] = *s0++;
 223         }
 224
 225         c = _UTF8_count[psenc->ch[0] & 0xff];
 226         if (c < 1 || c < psenc->chlen)
 227                 goto ilseq;
 228
 229         if (c == 1)
 230                 wchar = psenc->ch[0] & 0xff;
 231         else {
 232                 while (psenc->chlen < c) {
 233                         if (n-- < 1)
 234                                 goto restart;
 235                         psenc->ch[psenc->chlen++] = *s0++;
 236                 }
 237                 wchar = psenc->ch[0] & (0x7f >> c);
 238                 for (i = 1; i < c; i++) {
 239                         if ((psenc->ch[i] & 0xc0) != 0x80)
 240                                 goto ilseq;
 241                         wchar <<= 6;
 242                         wchar |= (psenc->ch[i] & 0x3f);
 243                 }
 244                 if (_UTF8_surrogate(wchar) || _UTF8_findlen(wchar) != c)
 245                         goto ilseq;
 246         }
 247         if (pwc != NULL)
 248                 *pwc = wchar;
 249         *nresult = (wchar == 0) ? 0 : s0 - *s;
 250         *s = s0;
 251         psenc->chlen = 0;
 252
 253         return 0;
 254
 255 ilseq:
 256         *nresult = (size_t)-1;
 257         return EILSEQ;
 258
 259 restart:
 260         *s = s0;
 261         *nresult = (size_t)-2;
 262         return 0;
 263 }
 264
 265 static int
 266 _citrus_UTF8_wcrtomb_priv(_UTF8EncodingInfo *ei, char *s, size_t n, wchar_t wc,
 267                           _UTF8State *psenc, size_t *nresult)
 268 {
 269         int cnt, i, ret;
 270         wchar_t c;
 271
 272         _DIAGASSERT(nresult != 0);
 273         _DIAGASSERT(s != NULL);
 274
 275         if (_UTF8_surrogate(wc)) {
 276                 ret = EILSEQ;
 277                 goto err;
 278         }
 279         cnt = _UTF8_findlen(wc);
 280         if (cnt <= 0 || cnt > 6) {
 281                 /* invalid UCS4 value */
 282                 ret = EILSEQ;
 283                 goto err;
 284         }
 285         if (n < cnt) {
 286                 /* bound check failure */
 287                 ret = E2BIG;
 288                 goto err;
 289         }
 290
 291         c = wc;
 292         if (s) {
 293                 for (i = cnt - 1; i > 0; i--) {
 294                         s[i] = 0x80 | (c & 0x3f);
 295                         c >>= 6;
 296                 }
 297                 s[0] = c;
 298                 if (cnt == 1)
 299                         s[0] &= 0x7f;
 300                 else {
 301                         s[0] &= (0x7f >> cnt);
 302                         s[0] |= ((0xff00 >> cnt) & 0xff);
 303                 }
 304         }
 305
 306         *nresult = (size_t)cnt;
 307         return 0;
 308
 309 err:
 310         *nresult = (size_t)-1;
 311         return ret;
 312 }
 313
 314 static __inline int
 315 /*ARGSUSED*/
 316 _citrus_UTF8_stdenc_wctocs(_UTF8EncodingInfo * __restrict ei,
 317                            _csid_t * __restrict csid,
 318                            _index_t * __restrict idx,
 319                            wchar_t wc)
 320 {
 321
 322         _DIAGASSERT(csid != NULL && idx != NULL);
 323
 324         *csid = 0;
 325         *idx = (_citrus_index_t)wc;
 326
 327         return (0);
 328 }
 329
 330 static __inline int
 331 /*ARGSUSED*/
 332 _citrus_UTF8_stdenc_cstowc(_UTF8EncodingInfo * __restrict ei,
 333                            wchar_t * __restrict wc,
 334                            _csid_t csid, _index_t idx)
 335 {
 336
 337         _DIAGASSERT(wc != NULL);
 338
 339         if (csid != 0)
 340                 return (EILSEQ);
 341
 342         *wc = (wchar_t)idx;
 343
 344         return (0);
 345 }
 346
 347 static __inline int
 348 /*ARGSUSED*/
 349 _citrus_UTF8_stdenc_get_state_desc_generic(_UTF8EncodingInfo * __restrict ei,
 350                                            _UTF8State * __restrict psenc,
 351                                            int * __restrict rstate)
 352 {
 353
 354         if (psenc->chlen == 0)
 355                 *rstate = _STDENC_SDGEN_INITIAL;
 356         else
 357                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 358
 359         return 0;
 360 }
 361
 362 static int
 363 /*ARGSUSED*/
 364 _citrus_UTF8_encoding_module_init(_UTF8EncodingInfo * __restrict ei,
 365                                   const void * __restrict var, size_t lenvar)
 366 {
 367         _UTF8_init_count();
 368
 369         return 0;
 370 }
 371
 372 static void
 373 /*ARGSUSED*/
 374 _citrus_UTF8_encoding_module_uninit(_UTF8EncodingInfo *ei)
 375 {
 376 }
 377
 378
 379 /* ----------------------------------------------------------------------
 380  * public interface for ctype
 381  */
 382
 383 _CITRUS_CTYPE_DECLS(UTF8);
 384 _CITRUS_CTYPE_DEF_OPS(UTF8);
 385
 386 #include "citrus_ctype_template.h"
 387
 388 /* ----------------------------------------------------------------------
 389  * public interface for stdenc
 390  */
 391
 392 _CITRUS_STDENC_DECLS(UTF8);
 393 _CITRUS_STDENC_DEF_OPS(UTF8);
 394
 395 #include "citrus_stdenc_template.h"