usr/src/lib/libc/port/locale/gb2312.c

   1 /*
   2  * Copyright 2013 Garrett D'Amore <garrett@damore.org>
   3  * Copyright 2010 Nexenta Systems, Inc.  All rights reserved.
   4  * Copyright (c) 2004 Tim J. Robbins. All rights reserved.
   5  * Copyright (c) 2003 David Xu <davidxu@freebsd.org>
   6  * All rights reserved.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  *
  17  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  27  * SUCH DAMAGE.
  28  */
  29
  30 #include "lint.h"
  31 #include <sys/types.h>
  32 #include <errno.h>
  33 #include "runetype.h"
  34 #include <stdlib.h>
  35 #include <string.h>
  36 #include <wchar.h>
  37 #include "mblocal.h"
  38 #include "lctype.h"
  39
  40 static size_t   _GB2312_mbrtowc(wchar_t *_RESTRICT_KYWD,
  41                     const char *_RESTRICT_KYWD,
  42                     size_t, mbstate_t *_RESTRICT_KYWD);
  43 static int      _GB2312_mbsinit(const mbstate_t *);
  44 static size_t   _GB2312_wcrtomb(char *_RESTRICT_KYWD, wchar_t,
  45                     mbstate_t *_RESTRICT_KYWD);
  46 static size_t   _GB2312_mbsnrtowcs(wchar_t *_RESTRICT_KYWD,
  47                     const char **_RESTRICT_KYWD, size_t, size_t,
  48                     mbstate_t *_RESTRICT_KYWD);
  49 static size_t   _GB2312_wcsnrtombs(char *_RESTRICT_KYWD,
  50                     const wchar_t **_RESTRICT_KYWD, size_t, size_t,
  51                     mbstate_t *_RESTRICT_KYWD);
  52
  53
  54 typedef struct {
  55         int     count;
  56         uchar_t bytes[2];
  57 } _GB2312State;
  58
  59 void
  60 _GB2312_init(struct lc_ctype *lct)
  61 {
  62
  63         lct->lc_mbrtowc = _GB2312_mbrtowc;
  64         lct->lc_wcrtomb = _GB2312_wcrtomb;
  65         lct->lc_mbsinit = _GB2312_mbsinit;
  66         lct->lc_mbsnrtowcs = _GB2312_mbsnrtowcs;
  67         lct->lc_wcsnrtombs = _GB2312_wcsnrtombs;
  68         lct->lc_max_mblen = 2;
  69         lct->lc_is_ascii = 0;
  70 }
  71
  72 static int
  73 _GB2312_mbsinit(const mbstate_t *ps)
  74 {
  75
  76         return (ps == NULL || ((const _GB2312State *)ps)->count == 0);
  77 }
  78
  79 static int
  80 _GB2312_check(const char *str, size_t n)
  81 {
  82         const uchar_t *s = (const uchar_t *)str;
  83
  84         if (n == 0)
  85                 /* Incomplete multibyte sequence */
  86                 return (-2);
  87         if (s[0] >= 0xa1 && s[0] <= 0xfe) {
  88                 if (n < 2)
  89                         /* Incomplete multibyte sequence */
  90                         return (-2);
  91                 if (s[1] < 0xa1 || s[1] > 0xfe)
  92                         /* Invalid multibyte sequence */
  93                         return (-1);
  94                 return (2);
  95         } else if (s[0] & 0x80) {
  96                 /* Invalid multibyte sequence */
  97                 return (-1);
  98         }
  99         return (1);
 100 }
 101
 102 static size_t
 103 _GB2312_mbrtowc(wchar_t *_RESTRICT_KYWD pwc, const char *_RESTRICT_KYWD s,
 104     size_t n, mbstate_t *_RESTRICT_KYWD ps)
 105 {
 106         _GB2312State *gs;
 107         wchar_t wc;
 108         int i, len, ocount;
 109         size_t ncopy;
 110
 111         gs = (_GB2312State *)ps;
 112
 113         if (gs->count < 0 || gs->count > sizeof (gs->bytes)) {
 114                 errno = EINVAL;
 115                 return ((size_t)-1);
 116         }
 117
 118         if (s == NULL) {
 119                 s = "";
 120                 n = 1;
 121                 pwc = NULL;
 122         }
 123
 124         ncopy = MIN(MIN(n, MB_CUR_MAX), sizeof (gs->bytes) - gs->count);
 125         (void) memcpy(gs->bytes + gs->count, s, ncopy);
 126         ocount = gs->count;
 127         gs->count += ncopy;
 128         s = (char *)gs->bytes;
 129         n = gs->count;
 130
 131         if ((len = _GB2312_check(s, n)) < 0)
 132                 return ((size_t)len);
 133         wc = 0;
 134         i = len;
 135         while (i-- > 0)
 136                 wc = (wc << 8) | (unsigned char)*s++;
 137         if (pwc != NULL)
 138                 *pwc = wc;
 139         gs->count = 0;
 140         return (wc == L'\0' ? 0 : len - ocount);
 141 }
 142
 143 static size_t
 144 _GB2312_wcrtomb(char *_RESTRICT_KYWD s, wchar_t wc,
 145     mbstate_t *_RESTRICT_KYWD ps)
 146 {
 147         _GB2312State *gs;
 148
 149         gs = (_GB2312State *)ps;
 150
 151         if (gs->count != 0) {
 152                 errno = EINVAL;
 153                 return ((size_t)-1);
 154         }
 155
 156         if (s == NULL)
 157                 /* Reset to initial shift state (no-op) */
 158                 return (1);
 159         if (wc & 0x8000) {
 160                 *s++ = (wc >> 8) & 0xff;
 161                 *s = wc & 0xff;
 162                 return (2);
 163         }
 164         *s = wc & 0xff;
 165         return (1);
 166 }
 167
 168 static size_t
 169 _GB2312_mbsnrtowcs(wchar_t *_RESTRICT_KYWD dst,
 170     const char **_RESTRICT_KYWD src, size_t nms, size_t len,
 171     mbstate_t *_RESTRICT_KYWD ps)
 172 {
 173         return (__mbsnrtowcs_std(dst, src, nms, len, ps, _GB2312_mbrtowc));
 174 }
 175
 176 static size_t
 177 _GB2312_wcsnrtombs(char *_RESTRICT_KYWD dst,
 178     const wchar_t **_RESTRICT_KYWD src, size_t nwc, size_t len,
 179     mbstate_t *_RESTRICT_KYWD ps)
 180 {
 181         return (__wcsnrtombs_std(dst, src, nwc, len, ps, _GB2312_wcrtomb));
 182 }