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. All rights reserved. 5 * Copyright (c) 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * Paul Borman at Krystal Technologies. 10 * 11 * Copyright (c) 2011 The FreeBSD Foundation 12 * All rights reserved. 13 * Portions of this software were developed by David Chisnall 14 * under sponsorship from the FreeBSD Foundation. 15 * 16 * Redistribution and use in source and binary forms, with or without 17 * modification, are permitted provided that the following conditions 18 * are met: 19 * 1. Redistributions of source code must retain the above copyright 20 * notice, this list of conditions and the following disclaimer. 21 * 2. Redistributions in binary form must reproduce the above copyright 22 * notice, this list of conditions and the following disclaimer in the 23 * documentation and/or other materials provided with the distribution. 24 * 4. Neither the name of the University nor the names of its contributors 25 * may be used to endorse or promote products derived from this software 26 * without specific prior written permission. 27 * 28 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 29 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 30 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 31 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 32 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 33 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 34 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 35 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 37 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 38 * SUCH DAMAGE. 39 * 40 * @(#)none.c 8.1 (Berkeley) 6/4/93 41 */ 42 43 #include <errno.h> 44 #include <limits.h> 45 #include <runetype.h> 46 #include <stddef.h> 47 #include <stdio.h> 48 #include <stdlib.h> 49 #include <string.h> 50 #include <wchar.h> 51 #include "mblocal.h" 52 53 static size_t _none_mbrtowc(wchar_t * __restrict, const char * __restrict, 54 size_t, mbstate_t * __restrict); 55 static int _none_mbsinit(const mbstate_t *); 56 static size_t _none_mbsnrtowcs(wchar_t * __restrict dst, 57 const char ** __restrict src, size_t nms, size_t len, 58 mbstate_t * __restrict ps __unused); 59 static size_t _none_wcrtomb(char * __restrict, wchar_t, 60 mbstate_t * __restrict); 61 static size_t _none_wcsnrtombs(char * __restrict, const wchar_t ** __restrict, 62 size_t, size_t, mbstate_t * __restrict); 63 64 /* setup defaults */ 65 66 int __mb_cur_max = 1; 67 int __mb_sb_limit = 256; /* Expected to be <= _CACHED_RUNES */ 68 69 int 70 _none_init(struct xlocale_ctype *l, _RuneLocale *rl) 71 { 72 73 l->__mbrtowc = _none_mbrtowc; 74 l->__mbsinit = _none_mbsinit; 75 l->__mbsnrtowcs = _none_mbsnrtowcs; 76 l->__wcrtomb = _none_wcrtomb; 77 l->__wcsnrtombs = _none_wcsnrtombs; 78 l->runes = rl; 79 l->__mb_cur_max = 1; 80 l->__mb_sb_limit = 256; 81 return(0); 82 } 83 84 static int 85 _none_mbsinit(const mbstate_t *ps __unused) 86 { 87 88 /* 89 * Encoding is not state dependent - we are always in the 90 * initial state. 91 */ 92 return (1); 93 } 94 95 static size_t 96 _none_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, size_t n, 97 mbstate_t * __restrict ps __unused) 98 { 99 100 if (s == NULL) 101 /* Reset to initial shift state (no-op) */ 102 return (0); 103 if (n == 0) 104 /* Incomplete multibyte sequence */ 105 return ((size_t)-2); 106 if (pwc != NULL) 107 *pwc = (unsigned char)*s; 108 return (*s == '\0' ? 0 : 1); 109 } 110 111 static size_t 112 _none_wcrtomb(char * __restrict s, wchar_t wc, 113 mbstate_t * __restrict ps __unused) 114 { 115 116 if (s == NULL) 117 /* Reset to initial shift state (no-op) */ 118 return (1); 119 if (wc < 0 || wc > UCHAR_MAX) { 120 errno = EILSEQ; 121 return ((size_t)-1); 122 } 123 *s = (unsigned char)wc; 124 return (1); 125 } 126 127 static size_t 128 _none_mbsnrtowcs(wchar_t * __restrict dst, const char ** __restrict src, 129 size_t nms, size_t len, mbstate_t * __restrict ps __unused) 130 { 131 const char *s; 132 size_t nchr; 133 134 if (dst == NULL) { 135 s = memchr(*src, '\0', nms); 136 return (s != NULL ? s - *src : nms); 137 } 138 139 s = *src; 140 nchr = 0; 141 while (len-- > 0 && nms-- > 0) { 142 if ((*dst++ = (unsigned char)*s++) == L'\0') { 143 *src = NULL; 144 return (nchr); 145 } 146 nchr++; 147 } 148 *src = s; 149 return (nchr); 150 } 151 152 static size_t 153 _none_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src, 154 size_t nwc, size_t len, mbstate_t * __restrict ps __unused) 155 { 156 const wchar_t *s; 157 size_t nchr; 158 159 if (dst == NULL) { 160 for (s = *src; nwc > 0 && *s != L'\0'; s++, nwc--) { 161 if (*s < 0 || *s > UCHAR_MAX) { 162 errno = EILSEQ; 163 return ((size_t)-1); 164 } 165 } 166 return (s - *src); 167 } 168 169 s = *src; 170 nchr = 0; 171 while (len-- > 0 && nwc-- > 0) { 172 if (*s < 0 || *s > UCHAR_MAX) { 173 errno = EILSEQ; 174 return ((size_t)-1); 175 } 176 if ((*dst++ = *s++) == '\0') { 177 *src = NULL; 178 return (nchr); 179 } 180 nchr++; 181 } 182 *src = s; 183 return (nchr); 184 } 185 186 /* 187 * Multibyte binary data to escaped wchar. 188 * Round-trip match guaranteed, including 0x00 bytes. 189 * 190 * Cannot return an error. *slen bytes is converted to the destination 191 * buffer until one or the other is exhausted. Destination elements returned 192 * and *slen modified with source elements processed. 193 * 194 * Incomplete sequences or partial re-encodings that would overflow the 195 * destination buffer are not processed and will also leave excess *slen. 196 * 197 * Never returns an error. Instead, incomplete sequences are not processed 198 * and *slen will index to the beginning of the incomplete sequence. It is 199 * possible for 0 to be returned and for *slen to be set to 0 due to an 200 * incomplete whole-buffer sequence, unless termination is specified. 201 * 202 * If termination is specified any trailing incomplete sequences are escaped 203 * and *slen will index to the end of the source buffer, unless insufficient 204 * room exists in the destination. If there is insufficient room, *slen may 205 * not be able to index to the end of the source buffer. 206 * 207 * Does not support a NULL dst on purpose - caller is expected to loop 208 * in parts. 209 */ 210 static size_t 211 _none_mbintowcr(wchar_t * __restrict dst, const char * __restrict src, 212 size_t dlen, size_t *slen, int flags) 213 { 214 size_t i; 215 size_t j; 216 size_t n = *slen; 217 218 for (i = j = 0; i < n; ++i) { 219 if (j == dlen) 220 break; 221 if (dst) 222 dst[j] = (unsigned char)src[i]; 223 ++j; 224 } 225 /* no partial sequences so we can ignore flags */ 226 *slen = i; 227 228 return j; 229 } 230 231 /* 232 * Escaped wchar to multibyte binary data. 233 * Round-trip match guaranteed, including 0x00 bytes. 234 * 235 * *slen bytes is converted to the destination buffer until one or the other 236 * is exhausted. Destination elements returned and *slen modified with 237 * source elements processed. 238 * 239 * Can return an error only if the first wchar src[] element is illegal, 240 * otherwise will process up to the illegal wchar and return an error on 241 * the next call (if called with the remainder). 242 * 243 * Never returns -2. Instead, incomplete sequences are not processed and 244 * *slen will index to the beginning of the incomplete sequence. If 245 * termination is specified, incomplete sequences are discarded and *slen 246 * indexes to the end of the input array. 247 * 248 * Does not support a NULL dst on purpose - caller is expected to loop 249 * in parts. 250 */ 251 static size_t 252 _none_wcrtombin(char * __restrict dst, const wchar_t * __restrict src, 253 size_t dlen, size_t *slen, int flags) 254 { 255 size_t i; 256 size_t j; 257 size_t n = *slen; 258 259 for (i = j = 0; i < n; ++i) { 260 if (j == dlen) 261 break; 262 if (src[i] >= 0x100) { 263 if (i == 0) { 264 errno = EILSEQ; 265 return(-1); 266 } 267 break; 268 } 269 if (dst) 270 dst[j] = (unsigned char)src[i]; 271 ++j; 272 } 273 /* no partial sequences so we can ignore flags */ 274 *slen = i; 275 276 return j; 277 } 278 279 /* setup defaults */ 280 281 struct xlocale_ctype __xlocale_global_ctype = { 282 {{0}, "C"}, 283 (_RuneLocale*)&_DefaultRuneLocale, 284 _none_mbrtowc, 285 _none_mbsinit, 286 _none_mbsnrtowcs, 287 _none_wcrtomb, 288 _none_wcsnrtombs, 289 _none_mbintowcr, 290 _none_wcrtombin, 291 1, /* __mb_cur_max, */ 292 256 /* __mb_sb_limit */ 293 }; 294 295 struct xlocale_ctype __xlocale_C_ctype = { 296 {{0}, "C"}, 297 (_RuneLocale*)&_DefaultRuneLocale, 298 _none_mbrtowc, 299 _none_mbsinit, 300 _none_mbsnrtowcs, 301 _none_wcrtomb, 302 _none_wcsnrtombs, 303 _none_mbintowcr, 304 _none_wcrtombin, 305 1, /* __mb_cur_max, */ 306 256 /* __mb_sb_limit */ 307 }; 308