1 /*- 2 * Copyright (c) 2004 Tim J. Robbins. All rights reserved. 3 * Copyright (c) 2003 David Xu <davidxu@freebsd.org> 4 * All rights reserved. 5 * 6 * Copyright (c) 2011 The FreeBSD Foundation 7 * All rights reserved. 8 * Portions of this software were developed by David Chisnall 9 * under sponsorship from the FreeBSD Foundation. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30 * SUCH DAMAGE. 31 * 32 * $FreeBSD: head/lib/libc/locale/gb2312.c 227753 2011-11-20 14:45:42Z theraven $ 33 */ 34 35 #include <sys/param.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 static size_t _GB2312_mbrtowc(wchar_t * __restrict, const char * __restrict, 45 size_t, mbstate_t * __restrict); 46 static int _GB2312_mbsinit(const mbstate_t *); 47 static size_t _GB2312_wcrtomb(char * __restrict, wchar_t, 48 mbstate_t * __restrict); 49 50 typedef struct { 51 int count; 52 u_char bytes[2]; 53 } _GB2312State; 54 55 int 56 _GB2312_init(struct xlocale_ctype *l, _RuneLocale *rl) 57 { 58 59 l->runes = rl; 60 l->__mbrtowc = _GB2312_mbrtowc; 61 l->__wcrtomb = _GB2312_wcrtomb; 62 l->__mbsinit = _GB2312_mbsinit; 63 l->__mb_cur_max = 2; 64 l->__mb_sb_limit = 128; 65 return (0); 66 } 67 68 static int 69 _GB2312_mbsinit(const mbstate_t *ps) 70 { 71 72 return (ps == NULL || ((const _GB2312State *)ps)->count == 0); 73 } 74 75 static __inline int 76 _GB2312_check(const char *str, size_t n) 77 { 78 const u_char *s = (const u_char *)str; 79 80 if (n == 0) 81 /* Incomplete multibyte sequence */ 82 return (-2); 83 if (s[0] >= 0xa1 && s[0] <= 0xfe) { 84 if (n < 2) 85 /* Incomplete multibyte sequence */ 86 return (-2); 87 if (s[1] < 0xa1 || s[1] > 0xfe) 88 /* Invalid multibyte sequence */ 89 return (-1); 90 return (2); 91 } else if (s[0] & 0x80) { 92 /* Invalid multibyte sequence */ 93 return (-1); 94 } 95 return (1); 96 } 97 98 static size_t 99 _GB2312_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, size_t n, 100 mbstate_t * __restrict ps) 101 { 102 _GB2312State *gs; 103 wchar_t wc; 104 int i, len, ocount; 105 size_t ncopy; 106 107 gs = (_GB2312State *)ps; 108 109 if (gs->count < 0 || gs->count > sizeof(gs->bytes)) { 110 errno = EINVAL; 111 return ((size_t)-1); 112 } 113 114 if (s == NULL) { 115 s = ""; 116 n = 1; 117 pwc = NULL; 118 } 119 120 ncopy = MIN(MIN(n, MB_CUR_MAX), sizeof(gs->bytes) - gs->count); 121 memcpy(gs->bytes + gs->count, s, ncopy); 122 ocount = gs->count; 123 gs->count += ncopy; 124 s = (char *)gs->bytes; 125 n = gs->count; 126 127 if ((len = _GB2312_check(s, n)) < 0) 128 return ((size_t)len); 129 wc = 0; 130 i = len; 131 while (i-- > 0) 132 wc = (wc << 8) | (unsigned char)*s++; 133 if (pwc != NULL) 134 *pwc = wc; 135 gs->count = 0; 136 return (wc == L'\0' ? 0 : len - ocount); 137 } 138 139 static size_t 140 _GB2312_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps) 141 { 142 _GB2312State *gs; 143 144 gs = (_GB2312State *)ps; 145 146 if (gs->count != 0) { 147 errno = EINVAL; 148 return ((size_t)-1); 149 } 150 151 if (s == NULL) 152 /* Reset to initial shift state (no-op) */ 153 return (1); 154 if (wc & 0x8000) { 155 *s++ = (wc >> 8) & 0xff; 156 *s = wc & 0xff; 157 return (2); 158 } 159 *s = wc & 0xff; 160 return (1); 161 } 162