locale/FreeBSD/gb18030.c

   1 /*-
   2  * Copyright (c) 2002-2004 Tim J. Robbins
   3  * All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer.
  10  * 2. Redistributions in binary form must reproduce the above copyright
  11  *    notice, this list of conditions and the following disclaimer in the
  12  *    documentation and/or other materials provided with the distribution.
  13  *
  14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  15  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  16  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  17  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  18  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  19  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  20  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  21  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  22  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  23  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  24  * SUCH DAMAGE.
  25  */
  26 /*
  27  * PRC National Standard GB 18030-2000 encoding of Chinese text.
  28  *
  29  * See gb18030(5) for details.
  30  */
  31
  32 #include <sys/param.h>
  33 __FBSDID("$FreeBSD: src/lib/libc/locale/gb18030.c,v 1.8 2007/10/13 16:28:21 ache Exp $");
  34
  35 #include "xlocale_private.h"
  36
  37 #include <errno.h>
  38 #include <runetype.h>
  39 #include <stdlib.h>
  40 #include <string.h>
  41 #include <wchar.h>
  42 #include "mblocal.h"
  43
  44 #define GB18030_MB_CUR_MAX      4
  45
  46 static size_t   _GB18030_mbrtowc(wchar_t * __restrict, const char * __restrict,
  47                     size_t, mbstate_t * __restrict, locale_t);
  48 static int      _GB18030_mbsinit(const mbstate_t *, locale_t);
  49 static size_t   _GB18030_wcrtomb(char * __restrict, wchar_t,
  50                     mbstate_t * __restrict, locale_t);
  51
  52 typedef struct {
  53         int     count;
  54         u_char  bytes[4];
  55 } _GB18030State;
  56
  57 int
  58 _GB18030_init(struct __xlocale_st_runelocale *xrl)
  59 {
  60
  61         xrl->__mbrtowc = _GB18030_mbrtowc;
  62         xrl->__wcrtomb = _GB18030_wcrtomb;
  63         xrl->__mbsinit = _GB18030_mbsinit;
  64         xrl->__mb_cur_max = GB18030_MB_CUR_MAX;
  65         xrl->__mb_sb_limit = 128;
  66
  67         return (0);
  68 }
  69
  70 static int
  71 _GB18030_mbsinit(const mbstate_t *ps, locale_t loc __unused)
  72 {
  73
  74         return (ps == NULL || ((const _GB18030State *)ps)->count == 0);
  75 }
  76
  77 static size_t
  78 _GB18030_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s,
  79     size_t n, mbstate_t * __restrict ps, locale_t loc __unused)
  80 {
  81         _GB18030State *gs;
  82         wchar_t wch;
  83         int ch, len, ocount;
  84         size_t ncopy;
  85
  86         gs = (_GB18030State *)ps;
  87
  88         if (gs->count < 0 || gs->count > sizeof(gs->bytes)) {
  89                 errno = EINVAL;
  90                 return ((size_t)-1);
  91         }
  92
  93         if (s == NULL) {
  94                 s = "";
  95                 n = 1;
  96                 pwc = NULL;
  97         }
  98
  99         ncopy = MIN(MIN(n, GB18030_MB_CUR_MAX), sizeof(gs->bytes) - gs->count);
 100         memcpy(gs->bytes + gs->count, s, ncopy);
 101         ocount = gs->count;
 102         gs->count += ncopy;
 103         s = (char *)gs->bytes;
 104         n = gs->count;
 105
 106         if (n == 0)
 107                 /* Incomplete multibyte sequence */
 108                 return ((size_t)-2);
 109
 110         /*
 111          * Single byte:         [00-7f]
 112          * Two byte:            [81-fe][40-7e,80-fe]
 113          * Four byte:           [81-fe][30-39][81-fe][30-39]
 114          */
 115         ch = (unsigned char)*s++;
 116         if (ch <= 0x7f) {
 117                 len = 1;
 118                 wch = ch;
 119         } else if (ch >= 0x81 && ch <= 0xfe) {
 120                 wch = ch;
 121                 if (n < 2)
 122                         return ((size_t)-2);
 123                 ch = (unsigned char)*s++;
 124                 if ((ch >= 0x40 && ch <= 0x7e) || (ch >= 0x80 && ch <= 0xfe)) {
 125                         wch = (wch << 8) | ch;
 126                         len = 2;
 127                 } else if (ch >= 0x30 && ch <= 0x39) {
 128                         /*
 129                          * Strip high bit off the wide character we will
 130                          * eventually output so that it is positive when
 131                          * cast to wint_t on 32-bit twos-complement machines.
 132                          */
 133                         wch = ((wch & 0x7f) << 8) | ch;
 134                         if (n < 3)
 135                                 return ((size_t)-2);
 136                         ch = (unsigned char)*s++;
 137                         if (ch < 0x81 || ch > 0xfe)
 138                                 goto ilseq;
 139                         wch = (wch << 8) | ch;
 140                         if (n < 4)
 141                                 return ((size_t)-2);
 142                         ch = (unsigned char)*s++;
 143                         if (ch < 0x30 || ch > 0x39)
 144                                 goto ilseq;
 145                         wch = (wch << 8) | ch;
 146                         len = 4;
 147                 } else
 148                         goto ilseq;
 149         } else
 150                 goto ilseq;
 151
 152         if (pwc != NULL)
 153                 *pwc = wch;
 154         gs->count = 0;
 155         return (wch == L'\0' ? 0 : len - ocount);
 156 ilseq:
 157         errno = EILSEQ;
 158         return ((size_t)-1);
 159 }
 160
 161 static size_t
 162 _GB18030_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps, locale_t loc __unused)
 163 {
 164         _GB18030State *gs;
 165         size_t len;
 166         int c;
 167
 168         gs = (_GB18030State *)ps;
 169
 170         if (gs->count != 0) {
 171                 errno = EINVAL;
 172                 return ((size_t)-1);
 173         }
 174
 175         if (s == NULL)
 176                 /* Reset to initial shift state (no-op) */
 177                 return (1);
 178         if ((wc & ~0x7fffffff) != 0)
 179                 goto ilseq;
 180         if (wc & 0x7f000000) {
 181                 /* Replace high bit that mbrtowc() removed. */
 182                 wc |= 0x80000000;
 183                 c = (wc >> 24) & 0xff;
 184                 if (c < 0x81 || c > 0xfe)
 185                         goto ilseq;
 186                 *s++ = c;
 187                 c = (wc >> 16) & 0xff;
 188                 if (c < 0x30 || c > 0x39)
 189                         goto ilseq;
 190                 *s++ = c;
 191                 c = (wc >> 8) & 0xff;
 192                 if (c < 0x81 || c > 0xfe)
 193                         goto ilseq;
 194                 *s++ = c;
 195                 c = wc & 0xff;
 196                 if (c < 0x30 || c > 0x39)
 197                         goto ilseq;
 198                 *s++ = c;
 199                 len = 4;
 200         } else if (wc & 0x00ff0000)
 201                 goto ilseq;
 202         else if (wc & 0x0000ff00) {
 203                 c = (wc >> 8) & 0xff;
 204                 if (c < 0x81 || c > 0xfe)
 205                         goto ilseq;
 206                 *s++ = c;
 207                 c = wc & 0xff;
 208                 if (c < 0x40 || c == 0x7f || c == 0xff)
 209                         goto ilseq;
 210                 *s++ = c;
 211                 len = 2;
 212         } else if (wc <= 0x7f) {
 213                 *s++ = wc;
 214                 len = 1;
 215         } else
 216                 goto ilseq;
 217
 218         return (len);
 219 ilseq:
 220         errno = EILSEQ;
 221         return ((size_t)-1);
 222 }