2 * Copyright (c) 2002-2004 Tim J. Robbins
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
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.
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
27 #include <sys/param.h>
28 __FBSDID("$FreeBSD$");
38 extern int __mb_sb_limit;
40 static size_t _UTF8_mbrtowc(wchar_t * __restrict, const char * __restrict,
41 size_t, mbstate_t * __restrict);
42 static int _UTF8_mbsinit(const mbstate_t *);
43 static size_t _UTF8_mbsnrtowcs(wchar_t * __restrict,
44 const char ** __restrict, size_t, size_t,
45 mbstate_t * __restrict);
46 static size_t _UTF8_wcrtomb(char * __restrict, wchar_t,
47 mbstate_t * __restrict);
48 static size_t _UTF8_wcsnrtombs(char * __restrict, const wchar_t ** __restrict,
49 size_t, size_t, mbstate_t * __restrict);
58 _UTF8_init(_RuneLocale *rl)
61 __mbrtowc = _UTF8_mbrtowc;
62 __wcrtomb = _UTF8_wcrtomb;
63 __mbsinit = _UTF8_mbsinit;
64 __mbsnrtowcs = _UTF8_mbsnrtowcs;
65 __wcsnrtombs = _UTF8_wcsnrtombs;
66 _CurrentRuneLocale = rl;
69 * UCS-4 encoding used as the internal representation, so
70 * slots 0x0080-0x00FF are occuped and must be excluded
71 * from the single byte ctype by setting the limit.
79 _UTF8_mbsinit(const mbstate_t *ps)
82 return (ps == NULL || ((const _UTF8State *)ps)->want == 0);
86 _UTF8_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, size_t n,
87 mbstate_t * __restrict ps)
90 int ch, i, mask, want;
93 us = (_UTF8State *)ps;
95 if (us->want < 0 || us->want > 6) {
107 /* Incomplete multibyte sequence */
110 if (us->want == 0 && ((ch = (unsigned char)*s) & ~0x7f) == 0) {
111 /* Fast path for plain ASCII characters. */
114 return (ch != '\0' ? 1 : 0);
119 * Determine the number of octets that make up this character
120 * from the first octet, and a mask that extracts the
121 * interesting bits of the first octet. We already know
122 * the character is at least two bytes long.
124 * We also specify a lower bound for the character code to
125 * detect redundant, non-"shortest form" encodings. For
126 * example, the sequence C0 80 is _not_ a legal representation
127 * of the null character. This enforces a 1-to-1 mapping
128 * between character codes and their multibyte representations.
130 ch = (unsigned char)*s;
131 if ((ch & 0x80) == 0) {
135 } else if ((ch & 0xe0) == 0xc0) {
139 } else if ((ch & 0xf0) == 0xe0) {
143 } else if ((ch & 0xf8) == 0xf0) {
147 } else if ((ch & 0xfc) == 0xf8) {
151 } else if ((ch & 0xfe) == 0xfc) {
157 * Malformed input; input is not UTF-8.
168 * Decode the octet sequence representing the character in chunks
169 * of 6 bits, most significant first.
172 wch = (unsigned char)*s++ & mask;
175 for (i = (us->want == 0) ? 1 : 0; i < MIN(want, n); i++) {
176 if ((*s & 0xc0) != 0x80) {
178 * Malformed input; bad characters in the middle
188 /* Incomplete multibyte sequence. */
196 * Malformed input; redundant encoding.
204 return (wch == L'\0' ? 0 : want);
208 _UTF8_mbsnrtowcs(wchar_t * __restrict dst, const char ** __restrict src,
209 size_t nms, size_t len, mbstate_t * __restrict ps)
217 us = (_UTF8State *)ps;
224 * The fast path in the loop below is not safe if an ASCII
225 * character appears as anything but the first byte of a
226 * multibyte sequence. Check now to avoid doing it in the loop.
228 if (nms > 0 && us->want > 0 && (signed char)*s > 0) {
233 if (nms > 0 && (signed char)*s > 0)
235 * Fast path for plain ASCII characters
239 else if ((nb = _UTF8_mbrtowc(&wc, s, nms, ps)) ==
241 /* Invalid sequence - mbrtowc() sets errno. */
243 else if (nb == 0 || nb == (size_t)-2)
253 * The fast path in the loop below is not safe if an ASCII
254 * character appears as anything but the first byte of a
255 * multibyte sequence. Check now to avoid doing it in the loop.
257 if (nms > 0 && len > 0 && us->want > 0 && (signed char)*s > 0) {
262 if (nms > 0 && (signed char)*s > 0) {
264 * Fast path for plain ASCII characters
269 } else if ((nb = _UTF8_mbrtowc(dst, s, nms, ps)) ==
273 } else if (nb == (size_t)-2) {
276 } else if (nb == 0) {
290 _UTF8_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps)
296 us = (_UTF8State *)ps;
304 /* Reset to initial shift state (no-op) */
307 if ((wc & ~0x7f) == 0) {
308 /* Fast path for plain ASCII characters. */
314 * Determine the number of octets needed to represent this character.
315 * We always output the shortest sequence possible. Also specify the
316 * first few bits of the first octet, which contains the information
317 * about the sequence length.
319 if ((wc & ~0x7f) == 0) {
322 } else if ((wc & ~0x7ff) == 0) {
325 } else if ((wc & ~0xffff) == 0) {
328 } else if ((wc & ~0x1fffff) == 0) {
331 } else if ((wc & ~0x3ffffff) == 0) {
334 } else if ((wc & ~0x7fffffff) == 0) {
343 * Output the octets representing the character in chunks
344 * of 6 bits, least significant last. The first octet is
345 * a special case because it contains the sequence length
348 for (i = len - 1; i > 0; i--) {
349 s[i] = (wc & 0x3f) | 0x80;
352 *s = (wc & 0xff) | lead;
358 _UTF8_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src,
359 size_t nwc, size_t len, mbstate_t * __restrict ps)
362 char buf[MB_LEN_MAX];
367 us = (_UTF8State *)ps;
379 if (0 <= *s && *s < 0x80)
380 /* Fast path for plain ASCII characters. */
382 else if ((nb = _UTF8_wcrtomb(buf, *s, ps)) ==
384 /* Invalid character - wcrtomb() sets errno. */
387 return (nbytes + nb - 1);
394 while (len > 0 && nwc-- > 0) {
395 if (0 <= *s && *s < 0x80) {
396 /* Fast path for plain ASCII characters. */
399 } else if (len > (size_t)MB_CUR_MAX) {
400 /* Enough space to translate in-place. */
401 if ((nb = _UTF8_wcrtomb(dst, *s, ps)) == (size_t)-1) {
407 * May not be enough space; use temp. buffer.
409 if ((nb = _UTF8_wcrtomb(buf, *s, ps)) == (size_t)-1) {
414 /* MB sequence for character won't fit. */
416 memcpy(dst, buf, nb);
420 return (nbytes + nb - 1);