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 static size_t _UTF8_mbrtowc(wchar_t * __restrict, const char * __restrict,
39 size_t, mbstate_t * __restrict);
40 static int _UTF8_mbsinit(const mbstate_t *);
41 static size_t _UTF8_mbsnrtowcs(wchar_t * __restrict,
42 const char ** __restrict, size_t, size_t,
43 mbstate_t * __restrict);
44 static size_t _UTF8_wcrtomb(char * __restrict, wchar_t,
45 mbstate_t * __restrict);
46 static size_t _UTF8_wcsnrtombs(char * __restrict, const wchar_t ** __restrict,
47 size_t, size_t, mbstate_t * __restrict);
56 _UTF8_init(_RuneLocale *rl)
59 __mbrtowc = _UTF8_mbrtowc;
60 __wcrtomb = _UTF8_wcrtomb;
61 __mbsinit = _UTF8_mbsinit;
62 __mbsnrtowcs = _UTF8_mbsnrtowcs;
63 __wcsnrtombs = _UTF8_wcsnrtombs;
64 _CurrentRuneLocale = rl;
71 _UTF8_mbsinit(const mbstate_t *ps)
74 return (ps == NULL || ((const _UTF8State *)ps)->want == 0);
78 _UTF8_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, size_t n,
79 mbstate_t * __restrict ps)
82 int ch, i, mask, want;
85 us = (_UTF8State *)ps;
87 if (us->want < 0 || us->want > 6) {
99 /* Incomplete multibyte sequence */
102 if (us->want == 0 && ((ch = (unsigned char)*s) & ~0x7f) == 0) {
103 /* Fast path for plain ASCII characters. */
106 return (ch != '\0' ? 1 : 0);
111 * Determine the number of octets that make up this character
112 * from the first octet, and a mask that extracts the
113 * interesting bits of the first octet. We already know
114 * the character is at least two bytes long.
116 * We also specify a lower bound for the character code to
117 * detect redundant, non-"shortest form" encodings. For
118 * example, the sequence C0 80 is _not_ a legal representation
119 * of the null character. This enforces a 1-to-1 mapping
120 * between character codes and their multibyte representations.
122 ch = (unsigned char)*s;
123 if ((ch & 0x80) == 0) {
127 } else if ((ch & 0xe0) == 0xc0) {
131 } else if ((ch & 0xf0) == 0xe0) {
135 } else if ((ch & 0xf8) == 0xf0) {
139 } else if ((ch & 0xfc) == 0xf8) {
143 } else if ((ch & 0xfc) == 0xfc) {
149 * Malformed input; input is not UTF-8.
160 * Decode the octet sequence representing the character in chunks
161 * of 6 bits, most significant first.
164 wch = (unsigned char)*s++ & mask;
167 for (i = (us->want == 0) ? 1 : 0; i < MIN(want, n); i++) {
168 if ((*s & 0xc0) != 0x80) {
170 * Malformed input; bad characters in the middle
180 /* Incomplete multibyte sequence. */
188 * Malformed input; redundant encoding.
196 return (wch == L'\0' ? 0 : want);
200 _UTF8_mbsnrtowcs(wchar_t * __restrict dst, const char ** __restrict src,
201 size_t nms, size_t len, mbstate_t * __restrict ps)
209 us = (_UTF8State *)ps;
216 * The fast path in the loop below is not safe if an ASCII
217 * character appears as anything but the first byte of a
218 * multibyte sequence. Check now to avoid doing it in the loop.
220 if (nms > 0 && us->want > 0 && (signed char)*s > 0) {
225 if (nms > 0 && (signed char)*s > 0)
227 * Fast path for plain ASCII characters
231 else if ((nb = _UTF8_mbrtowc(&wc, s, nms, ps)) ==
233 /* Invalid sequence - mbrtowc() sets errno. */
235 else if (nb == 0 || nb == (size_t)-2)
245 * The fast path in the loop below is not safe if an ASCII
246 * character appears as anything but the first byte of a
247 * multibyte sequence. Check now to avoid doing it in the loop.
249 if (nms > 0 && len > 0 && us->want > 0 && (signed char)*s > 0) {
254 if (nms > 0 && (signed char)*s > 0) {
256 * Fast path for plain ASCII characters
261 } else if ((nb = _UTF8_mbrtowc(dst, s, nms, ps)) ==
265 } else if (nb == (size_t)-2) {
268 } else if (nb == 0) {
282 _UTF8_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps)
288 us = (_UTF8State *)ps;
296 /* Reset to initial shift state (no-op) */
299 if ((wc & ~0x7f) == 0) {
300 /* Fast path for plain ASCII characters. */
306 * Determine the number of octets needed to represent this character.
307 * We always output the shortest sequence possible. Also specify the
308 * first few bits of the first octet, which contains the information
309 * about the sequence length.
311 if ((wc & ~0x7f) == 0) {
314 } else if ((wc & ~0x7ff) == 0) {
317 } else if ((wc & ~0xffff) == 0) {
320 } else if ((wc & ~0x1fffff) == 0) {
323 } else if ((wc & ~0x3ffffff) == 0) {
326 } else if ((wc & ~0x7fffffff) == 0) {
335 * Output the octets representing the character in chunks
336 * of 6 bits, least significant last. The first octet is
337 * a special case because it contains the sequence length
340 for (i = len - 1; i > 0; i--) {
341 s[i] = (wc & 0x3f) | 0x80;
344 *s = (wc & 0xff) | lead;
350 _UTF8_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src,
351 size_t nwc, size_t len, mbstate_t * __restrict ps)
354 char buf[MB_LEN_MAX];
359 us = (_UTF8State *)ps;
371 if (0 <= *s && *s < 0x80)
372 /* Fast path for plain ASCII characters. */
374 else if ((nb = _UTF8_wcrtomb(buf, *s, ps)) ==
376 /* Invalid character - wcrtomb() sets errno. */
379 return (nbytes + nb - 1);
386 while (len > 0 && nwc-- > 0) {
387 if (0 <= *s && *s < 0x80) {
388 /* Fast path for plain ASCII characters. */
391 } else if (len > (size_t)MB_CUR_MAX) {
392 /* Enough space to translate in-place. */
393 if ((nb = _UTF8_wcrtomb(dst, *s, ps)) == (size_t)-1) {
399 * May not be enough space; use temp. buffer.
401 if ((nb = _UTF8_wcrtomb(buf, *s, ps)) == (size_t)-1) {
406 /* MB sequence for character won't fit. */
408 memcpy(dst, buf, nb);
412 return (nbytes + nb - 1);