1 /*- 2 * Copyright (c) 2002-2004 Tim J. Robbins 3 * All rights reserved. 4 * 5 * Copyright (c) 2011 The FreeBSD Foundation 6 * All rights reserved. 7 * Portions of this software were developed by David Chisnall 8 * under sponsorship from the FreeBSD Foundation. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29 * SUCH DAMAGE. 30 * 31 * $FreeBSD: head/lib/libc/locale/gb18030.c 227753 2011-11-20 14:45:42Z theraven $ 32 */ 33 /* 34 * PRC National Standard GB 18030-2000 encoding of Chinese text. 35 * 36 * See gb18030(5) for details. 37 */ 38 39 #include <sys/param.h> 40 41 #include <errno.h> 42 #include <runetype.h> 43 #include <stdlib.h> 44 #include <string.h> 45 #include <wchar.h> 46 #include "mblocal.h" 47 48 static size_t _GB18030_mbrtowc(wchar_t * __restrict, const char * __restrict, 49 size_t, mbstate_t * __restrict); 50 static int _GB18030_mbsinit(const mbstate_t *); 51 static size_t _GB18030_wcrtomb(char * __restrict, wchar_t, 52 mbstate_t * __restrict); 53 54 typedef struct { 55 int count; 56 u_char bytes[4]; 57 } _GB18030State; 58 59 int 60 _GB18030_init(struct xlocale_ctype *l, _RuneLocale *rl) 61 { 62 63 l->__mbrtowc = _GB18030_mbrtowc; 64 l->__wcrtomb = _GB18030_wcrtomb; 65 l->__mbsinit = _GB18030_mbsinit; 66 l->runes = rl; 67 l->__mb_cur_max = 4; 68 l->__mb_sb_limit = 128; 69 70 return (0); 71 } 72 73 static int 74 _GB18030_mbsinit(const mbstate_t *ps) 75 { 76 77 return (ps == NULL || ((const _GB18030State *)ps)->count == 0); 78 } 79 80 static size_t 81 _GB18030_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, 82 size_t n, mbstate_t * __restrict ps) 83 { 84 _GB18030State *gs; 85 wchar_t wch; 86 int ch, len, ocount; 87 size_t ncopy; 88 89 gs = (_GB18030State *)ps; 90 91 if (gs->count < 0 || gs->count > sizeof(gs->bytes)) { 92 errno = EINVAL; 93 return ((size_t)-1); 94 } 95 96 if (s == NULL) { 97 s = ""; 98 n = 1; 99 pwc = NULL; 100 } 101 102 ncopy = MIN(MIN(n, MB_CUR_MAX), sizeof(gs->bytes) - gs->count); 103 memcpy(gs->bytes + gs->count, s, ncopy); 104 ocount = gs->count; 105 gs->count += ncopy; 106 s = (char *)gs->bytes; 107 n = gs->count; 108 109 if (n == 0) 110 /* Incomplete multibyte sequence */ 111 return ((size_t)-2); 112 113 /* 114 * Single byte: [00-7f] 115 * Two byte: [81-fe][40-7e,80-fe] 116 * Four byte: [81-fe][30-39][81-fe][30-39] 117 */ 118 ch = (unsigned char)*s++; 119 if (ch <= 0x7f) { 120 len = 1; 121 wch = ch; 122 } else if (ch >= 0x81 && ch <= 0xfe) { 123 wch = ch; 124 if (n < 2) 125 return ((size_t)-2); 126 ch = (unsigned char)*s++; 127 if ((ch >= 0x40 && ch <= 0x7e) || (ch >= 0x80 && ch <= 0xfe)) { 128 wch = (wch << 8) | ch; 129 len = 2; 130 } else if (ch >= 0x30 && ch <= 0x39) { 131 /* 132 * Strip high bit off the wide character we will 133 * eventually output so that it is positive when 134 * cast to wint_t on 32-bit twos-complement machines. 135 */ 136 wch = ((wch & 0x7f) << 8) | ch; 137 if (n < 3) 138 return ((size_t)-2); 139 ch = (unsigned char)*s++; 140 if (ch < 0x81 || ch > 0xfe) 141 goto ilseq; 142 wch = (wch << 8) | ch; 143 if (n < 4) 144 return ((size_t)-2); 145 ch = (unsigned char)*s++; 146 if (ch < 0x30 || ch > 0x39) 147 goto ilseq; 148 wch = (wch << 8) | ch; 149 len = 4; 150 } else 151 goto ilseq; 152 } else 153 goto ilseq; 154 155 if (pwc != NULL) 156 *pwc = wch; 157 gs->count = 0; 158 return (wch == L'\0' ? 0 : len - ocount); 159 ilseq: 160 errno = EILSEQ; 161 return ((size_t)-1); 162 } 163 164 static size_t 165 _GB18030_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps) 166 { 167 _GB18030State *gs; 168 size_t len; 169 int c; 170 171 gs = (_GB18030State *)ps; 172 173 if (gs->count != 0) { 174 errno = EINVAL; 175 return ((size_t)-1); 176 } 177 178 if (s == NULL) 179 /* Reset to initial shift state (no-op) */ 180 return (1); 181 if ((wc & ~0x7fffffff) != 0) 182 goto ilseq; 183 if (wc & 0x7f000000) { 184 /* Replace high bit that mbrtowc() removed. */ 185 wc |= 0x80000000; 186 c = (wc >> 24) & 0xff; 187 if (c < 0x81 || c > 0xfe) 188 goto ilseq; 189 *s++ = c; 190 c = (wc >> 16) & 0xff; 191 if (c < 0x30 || c > 0x39) 192 goto ilseq; 193 *s++ = c; 194 c = (wc >> 8) & 0xff; 195 if (c < 0x81 || c > 0xfe) 196 goto ilseq; 197 *s++ = c; 198 c = wc & 0xff; 199 if (c < 0x30 || c > 0x39) 200 goto ilseq; 201 *s++ = c; 202 len = 4; 203 } else if (wc & 0x00ff0000) 204 goto ilseq; 205 else if (wc & 0x0000ff00) { 206 c = (wc >> 8) & 0xff; 207 if (c < 0x81 || c > 0xfe) 208 goto ilseq; 209 *s++ = c; 210 c = wc & 0xff; 211 if (c < 0x40 || c == 0x7f || c == 0xff) 212 goto ilseq; 213 *s++ = c; 214 len = 2; 215 } else if (wc <= 0x7f) { 216 *s++ = wc; 217 len = 1; 218 } else 219 goto ilseq; 220 221 return (len); 222 ilseq: 223 errno = EILSEQ; 224 return ((size_t)-1); 225 } 226