lib/libc/locale/gb18030.c

   1 /*
   2  * Copyright 2013 Garrett D'Amore <garrett@damore.org>
   3  * Copyright 2010 Nexenta Systems, Inc.  All rights reserved.
   4  * Copyright (c) 2002-2004 Tim J. Robbins
   5  * All rights reserved.
   6  *
   7  * Copyright (c) 2011 The FreeBSD Foundation
   8  * All rights reserved.
   9  * Portions of this software were developed by David Chisnall
  10  * under sponsorship from the FreeBSD Foundation.
  11  *
  12  * Redistribution and use in source and binary forms, with or without
  13  * modification, are permitted provided that the following conditions
  14  * are met:
  15  * 1. Redistributions of source code must retain the above copyright
  16  *    notice, this list of conditions and the following disclaimer.
  17  * 2. Redistributions in binary form must reproduce the above copyright
  18  *    notice, this list of conditions and the following disclaimer in the
  19  *    documentation and/or other materials provided with the distribution.
  20  *
  21  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  22  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  23  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  24  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  25  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  26  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  27  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  28  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  29  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  30  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  31  * SUCH DAMAGE.
  32  */
  33
  34 /*
  35  * PRC National Standard GB 18030-2000 encoding of Chinese text.
  36  *
  37  * See gb18030(5) for details.
  38  */
  39
  40 #include <sys/param.h>
  41
  42 #include <errno.h>
  43 #include <runetype.h>
  44 #include <stdlib.h>
  45 #include <string.h>
  46 #include <wchar.h>
  47 #include "mblocal.h"
  48
  49 static size_t   _GB18030_mbrtowc(wchar_t * __restrict, const char * __restrict,
  50                     size_t, mbstate_t * __restrict);
  51 static int      _GB18030_mbsinit(const mbstate_t *);
  52 static size_t   _GB18030_wcrtomb(char * __restrict, wchar_t,
  53                     mbstate_t * __restrict);
  54 static size_t   _GB18030_mbsnrtowcs(wchar_t * __restrict,
  55                     const char ** __restrict, size_t, size_t,
  56                     mbstate_t * __restrict);
  57 static size_t   _GB18030_wcsnrtombs(char * __restrict,
  58                     const wchar_t ** __restrict, size_t, size_t,
  59                     mbstate_t * __restrict);
  60
  61
  62 typedef struct {
  63         int     count;
  64         u_char  bytes[4];
  65 } _GB18030State;
  66
  67 int
  68 _GB18030_init(struct xlocale_ctype *l, _RuneLocale *rl)
  69 {
  70
  71         l->__mbrtowc = _GB18030_mbrtowc;
  72         l->__wcrtomb = _GB18030_wcrtomb;
  73         l->__mbsinit = _GB18030_mbsinit;
  74         l->__mbsnrtowcs = _GB18030_mbsnrtowcs;
  75         l->__wcsnrtombs = _GB18030_wcsnrtombs;
  76         l->runes = rl;
  77         l->__mb_cur_max = 4;
  78         l->__mb_sb_limit = 128;
  79
  80         return (0);
  81 }
  82
  83 static int
  84 _GB18030_mbsinit(const mbstate_t *ps)
  85 {
  86
  87         return (ps == NULL || ((const _GB18030State *)ps)->count == 0);
  88 }
  89
  90 static size_t
  91 _GB18030_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s,
  92     size_t n, mbstate_t * __restrict ps)
  93 {
  94         _GB18030State *gs;
  95         wchar_t wch;
  96         int ch, len, ocount;
  97         size_t ncopy;
  98
  99         gs = (_GB18030State *)ps;
 100
 101         if (gs->count < 0 || gs->count > sizeof(gs->bytes)) {
 102                 errno = EINVAL;
 103                 return ((size_t)-1);
 104         }
 105
 106         if (s == NULL) {
 107                 s = "";
 108                 n = 1;
 109                 pwc = NULL;
 110         }
 111
 112         ncopy = MIN(MIN(n, MB_CUR_MAX), sizeof(gs->bytes) - gs->count);
 113         (void) memcpy(gs->bytes + gs->count, s, ncopy);
 114         ocount = gs->count;
 115         gs->count += ncopy;
 116         s = (char *)gs->bytes;
 117         n = gs->count;
 118
 119         if (n == 0)
 120                 /* Incomplete multibyte sequence */
 121                 return ((size_t)-2);
 122
 123         /*
 124          * Single byte:         [00-7f]
 125          * Two byte:            [81-fe][40-7e,80-fe]
 126          * Four byte:           [81-fe][30-39][81-fe][30-39]
 127          */
 128         ch = (unsigned char)*s++;
 129         if (ch <= 0x7f) {
 130                 len = 1;
 131                 wch = ch;
 132         } else if (ch >= 0x81 && ch <= 0xfe) {
 133                 wch = ch;
 134                 if (n < 2)
 135                         return ((size_t)-2);
 136                 ch = (unsigned char)*s++;
 137                 if ((ch >= 0x40 && ch <= 0x7e) || (ch >= 0x80 && ch <= 0xfe)) {
 138                         wch = (wch << 8) | ch;
 139                         len = 2;
 140                 } else if (ch >= 0x30 && ch <= 0x39) {
 141                         /*
 142                          * Strip high bit off the wide character we will
 143                          * eventually output so that it is positive when
 144                          * cast to wint_t on 32-bit twos-complement machines.
 145                          */
 146                         wch = ((wch & 0x7f) << 8) | ch;
 147                         if (n < 3)
 148                                 return ((size_t)-2);
 149                         ch = (unsigned char)*s++;
 150                         if (ch < 0x81 || ch > 0xfe)
 151                                 goto ilseq;
 152                         wch = (wch << 8) | ch;
 153                         if (n < 4)
 154                                 return ((size_t)-2);
 155                         ch = (unsigned char)*s++;
 156                         if (ch < 0x30 || ch > 0x39)
 157                                 goto ilseq;
 158                         wch = (wch << 8) | ch;
 159                         len = 4;
 160                 } else
 161                         goto ilseq;
 162         } else
 163                 goto ilseq;
 164
 165         if (pwc != NULL)
 166                 *pwc = wch;
 167         gs->count = 0;
 168         return (wch == L'\0' ? 0 : len - ocount);
 169 ilseq:
 170         errno = EILSEQ;
 171         return ((size_t)-1);
 172 }
 173
 174 static size_t
 175 _GB18030_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps)
 176 {
 177         _GB18030State *gs;
 178         size_t len;
 179         int c;
 180
 181         gs = (_GB18030State *)ps;
 182
 183         if (gs->count != 0) {
 184                 errno = EINVAL;
 185                 return ((size_t)-1);
 186         }
 187
 188         if (s == NULL)
 189                 /* Reset to initial shift state (no-op) */
 190                 return (1);
 191         if ((wc & ~0x7fffffff) != 0)
 192                 goto ilseq;
 193         if (wc & 0x7f000000) {
 194                 /* Replace high bit that mbrtowc() removed. */
 195                 wc |= 0x80000000;
 196                 c = (wc >> 24) & 0xff;
 197                 if (c < 0x81 || c > 0xfe)
 198                         goto ilseq;
 199                 *s++ = c;
 200                 c = (wc >> 16) & 0xff;
 201                 if (c < 0x30 || c > 0x39)
 202                         goto ilseq;
 203                 *s++ = c;
 204                 c = (wc >> 8) & 0xff;
 205                 if (c < 0x81 || c > 0xfe)
 206                         goto ilseq;
 207                 *s++ = c;
 208                 c = wc & 0xff;
 209                 if (c < 0x30 || c > 0x39)
 210                         goto ilseq;
 211                 *s++ = c;
 212                 len = 4;
 213         } else if (wc & 0x00ff0000)
 214                 goto ilseq;
 215         else if (wc & 0x0000ff00) {
 216                 c = (wc >> 8) & 0xff;
 217                 if (c < 0x81 || c > 0xfe)
 218                         goto ilseq;
 219                 *s++ = c;
 220                 c = wc & 0xff;
 221                 if (c < 0x40 || c == 0x7f || c == 0xff)
 222                         goto ilseq;
 223                 *s++ = c;
 224                 len = 2;
 225         } else if (wc <= 0x7f) {
 226                 *s++ = wc;
 227                 len = 1;
 228         } else
 229                 goto ilseq;
 230
 231         return (len);
 232 ilseq:
 233         errno = EILSEQ;
 234         return ((size_t)-1);
 235 }
 236
 237 static size_t
 238 _GB18030_mbsnrtowcs(wchar_t * __restrict dst,
 239     const char ** __restrict src, size_t nms, size_t len,
 240     mbstate_t * __restrict ps)
 241 {
 242         return (__mbsnrtowcs_std(dst, src, nms, len, ps, _GB18030_mbrtowc));
 243 }
 244
 245 static size_t
 246 _GB18030_wcsnrtombs(char * __restrict dst,
 247     const wchar_t ** __restrict src, size_t nwc, size_t len,
 248     mbstate_t * __restrict ps)
 249 {
 250         return (__wcsnrtombs_std(dst, src, nwc, len, ps, _GB18030_wcrtomb));
 251 }