lib/libc/citrus/modules/citrus_utf8.c

   1 /*      $NetBSD: citrus_utf8.c,v 1.18 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  * 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 1.18 2013/05/28 16:57:56 joerg 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_mbsnrtowcs;
 115                 _UTF8State      s_wcrtomb;
 116                 _UTF8State      s_wcsrtombs;
 117                 _UTF8State      s_wcsnrtombs;
 118                 _UTF8State      s_wctomb;
 119         } states;
 120 } _UTF8CTypeInfo;
 121
 122 #define _CEI_TO_EI(_cei_)               (&(_cei_)->ei)
 123 #define _CEI_TO_STATE(_ei_, _func_)     (_ei_)->states.s_##_func_
 124
 125 #define _FUNCNAME(m)                    _citrus_UTF8_##m
 126 #define _ENCODING_INFO                  _UTF8EncodingInfo
 127 #define _CTYPE_INFO                     _UTF8CTypeInfo
 128 #define _ENCODING_STATE                 _UTF8State
 129 #define _ENCODING_MB_CUR_MAX(_ei_)      6
 130 #define _ENCODING_IS_STATE_DEPENDENT    0
 131 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_)        0
 132
 133
 134 static __inline void
 135 _UTF8_init_count(void)
 136 {
 137         int i;
 138         if (!_UTF8_count) {
 139                 memset(_UTF8_count_array, 0, sizeof(_UTF8_count_array));
 140                 for (i = 0; i <= 0x7f; i++)
 141                         _UTF8_count_array[i] = 1;
 142                 for (i = 0xc0; i <= 0xdf; i++)
 143                         _UTF8_count_array[i] = 2;
 144                 for (i = 0xe0; i <= 0xef; i++)
 145                         _UTF8_count_array[i] = 3;
 146                 for (i = 0xf0; i <= 0xf7; i++)
 147                         _UTF8_count_array[i] = 4;
 148                 for (i = 0xf8; i <= 0xfb; i++)
 149                         _UTF8_count_array[i] = 5;
 150                 for (i = 0xfc; i <= 0xfd; i++)
 151                         _UTF8_count_array[i] = 6;
 152                 _UTF8_count = _UTF8_count_array;
 153         }
 154 }
 155
 156 static int
 157 _UTF8_findlen(wchar_t v)
 158 {
 159         int i;
 160         u_int32_t c;
 161
 162         c = (u_int32_t)v;       /*XXX*/
 163         for (i = 1; i < sizeof(_UTF8_range) / sizeof(_UTF8_range[0]) - 1; i++)
 164                 if (c >= _UTF8_range[i] && c < _UTF8_range[i + 1])
 165                         return i;
 166
 167         return -1;      /*out of range*/
 168 }
 169
 170 static __inline int
 171 _UTF8_surrogate(wchar_t wc)
 172 {
 173         return wc >= 0xd800 && wc <= 0xdfff;
 174 }
 175
 176 static __inline void
 177 /*ARGSUSED*/
 178 _citrus_UTF8_init_state(_UTF8EncodingInfo *ei, _UTF8State *s)
 179 {
 180         s->chlen = 0;
 181 }
 182
 183 static __inline void
 184 /*ARGSUSED*/
 185 _citrus_UTF8_pack_state(_UTF8EncodingInfo *ei, void *pspriv,
 186                         const _UTF8State *s)
 187 {
 188         memcpy(pspriv, (const void *)s, sizeof(*s));
 189 }
 190
 191 static __inline void
 192 /*ARGSUSED*/
 193 _citrus_UTF8_unpack_state(_UTF8EncodingInfo *ei, _UTF8State *s,
 194                           const void *pspriv)
 195 {
 196         memcpy((void *)s, pspriv, sizeof(*s));
 197 }
 198
 199 static int
 200 _citrus_UTF8_mbrtowc_priv(_UTF8EncodingInfo *ei, wchar_t *pwc, const char **s,
 201                           size_t n, _UTF8State *psenc, size_t *nresult)
 202 {
 203         wchar_t wchar;
 204         const char *s0;
 205         int c;
 206         int i;
 207
 208         _DIAGASSERT(nresult != 0);
 209         _DIAGASSERT(s != NULL);
 210         _DIAGASSERT(psenc != NULL);
 211
 212         s0 = *s;
 213
 214         if (s0 == NULL) {
 215                 _citrus_UTF8_init_state(ei, psenc);
 216                 *nresult = 0; /* state independent */
 217                 return 0;
 218         }
 219
 220         /* make sure we have the first byte in the buffer */
 221         if (psenc->chlen == 0) {
 222                 if (n-- < 1)
 223                         goto restart;
 224                 psenc->ch[psenc->chlen++] = *s0++;
 225         }
 226
 227         c = _UTF8_count[psenc->ch[0] & 0xff];
 228         if (c < 1 || c < psenc->chlen)
 229                 goto ilseq;
 230
 231         if (c == 1)
 232                 wchar = psenc->ch[0] & 0xff;
 233         else {
 234                 while (psenc->chlen < c) {
 235                         if (n-- < 1)
 236                                 goto restart;
 237                         psenc->ch[psenc->chlen++] = *s0++;
 238                 }
 239                 wchar = psenc->ch[0] & (0x7f >> c);
 240                 for (i = 1; i < c; i++) {
 241                         if ((psenc->ch[i] & 0xc0) != 0x80)
 242                                 goto ilseq;
 243                         wchar <<= 6;
 244                         wchar |= (psenc->ch[i] & 0x3f);
 245                 }
 246                 if (_UTF8_surrogate(wchar) || _UTF8_findlen(wchar) != c)
 247                         goto ilseq;
 248         }
 249         if (pwc != NULL)
 250                 *pwc = wchar;
 251         *nresult = (wchar == 0) ? 0 : s0 - *s;
 252         *s = s0;
 253         psenc->chlen = 0;
 254
 255         return 0;
 256
 257 ilseq:
 258         *nresult = (size_t)-1;
 259         return EILSEQ;
 260
 261 restart:
 262         *s = s0;
 263         *nresult = (size_t)-2;
 264         return 0;
 265 }
 266
 267 static int
 268 _citrus_UTF8_wcrtomb_priv(_UTF8EncodingInfo *ei, char *s, size_t n, wchar_t wc,
 269                           _UTF8State *psenc, size_t *nresult)
 270 {
 271         int cnt, i, ret;
 272         wchar_t c;
 273
 274         _DIAGASSERT(nresult != 0);
 275         _DIAGASSERT(s != NULL);
 276
 277         if (_UTF8_surrogate(wc)) {
 278                 ret = EILSEQ;
 279                 goto err;
 280         }
 281         cnt = _UTF8_findlen(wc);
 282         if (cnt <= 0 || cnt > 6) {
 283                 /* invalid UCS4 value */
 284                 ret = EILSEQ;
 285                 goto err;
 286         }
 287         if (n < cnt) {
 288                 /* bound check failure */
 289                 ret = E2BIG;
 290                 goto err;
 291         }
 292
 293         c = wc;
 294         if (s) {
 295                 for (i = cnt - 1; i > 0; i--) {
 296                         s[i] = 0x80 | (c & 0x3f);
 297                         c >>= 6;
 298                 }
 299                 s[0] = c;
 300                 if (cnt == 1)
 301                         s[0] &= 0x7f;
 302                 else {
 303                         s[0] &= (0x7f >> cnt);
 304                         s[0] |= ((0xff00 >> cnt) & 0xff);
 305                 }
 306         }
 307
 308         *nresult = (size_t)cnt;
 309         return 0;
 310
 311 err:
 312         *nresult = (size_t)-1;
 313         return ret;
 314 }
 315
 316 static __inline int
 317 /*ARGSUSED*/
 318 _citrus_UTF8_stdenc_wctocs(_UTF8EncodingInfo * __restrict ei,
 319                            _csid_t * __restrict csid,
 320                            _index_t * __restrict idx,
 321                            wchar_t wc)
 322 {
 323
 324         _DIAGASSERT(csid != NULL && idx != NULL);
 325
 326         *csid = 0;
 327         *idx = (_citrus_index_t)wc;
 328
 329         return (0);
 330 }
 331
 332 static __inline int
 333 /*ARGSUSED*/
 334 _citrus_UTF8_stdenc_cstowc(_UTF8EncodingInfo * __restrict ei,
 335                            wchar_t * __restrict wc,
 336                            _csid_t csid, _index_t idx)
 337 {
 338
 339         _DIAGASSERT(wc != NULL);
 340
 341         if (csid != 0)
 342                 return (EILSEQ);
 343
 344         *wc = (wchar_t)idx;
 345
 346         return (0);
 347 }
 348
 349 static __inline int
 350 /*ARGSUSED*/
 351 _citrus_UTF8_stdenc_get_state_desc_generic(_UTF8EncodingInfo * __restrict ei,
 352                                            _UTF8State * __restrict psenc,
 353                                            int * __restrict rstate)
 354 {
 355
 356         if (psenc->chlen == 0)
 357                 *rstate = _STDENC_SDGEN_INITIAL;
 358         else
 359                 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
 360
 361         return 0;
 362 }
 363
 364 static int
 365 /*ARGSUSED*/
 366 _citrus_UTF8_encoding_module_init(_UTF8EncodingInfo * __restrict ei,
 367                                   const void * __restrict var, size_t lenvar)
 368 {
 369         _UTF8_init_count();
 370
 371         return 0;
 372 }
 373
 374 static void
 375 /*ARGSUSED*/
 376 _citrus_UTF8_encoding_module_uninit(_UTF8EncodingInfo *ei)
 377 {
 378 }
 379
 380
 381 /* ----------------------------------------------------------------------
 382  * public interface for ctype
 383  */
 384
 385 _CITRUS_CTYPE_DECLS(UTF8);
 386 _CITRUS_CTYPE_DEF_OPS(UTF8);
 387
 388 #include "citrus_ctype_template.h"
 389
 390 /* ----------------------------------------------------------------------
 391  * public interface for stdenc
 392  */
 393
 394 _CITRUS_STDENC_DECLS(UTF8);
 395 _CITRUS_STDENC_DEF_OPS(UTF8);
 396
 397 #include "citrus_stdenc_template.h"