gb18030.c revision 129153 
1/*-
2 * Copyright (c) 2002-2004 Tim J. Robbins
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
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.
13 *
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
24 * SUCH DAMAGE.
25 */
26/*
27 * PRC National Standard GB 18030-2000 encoding of Chinese text.
28 *
29 * See gb18030(5) for details.
30 */
31
32#include <sys/param.h>
33__FBSDID("$FreeBSD: head/lib/libc/locale/gb18030.c 129153 2004-05-12 14:09:04Z tjr $");
34
35#include <errno.h>
36#include <runetype.h>
37#include <stdlib.h>
38#include <string.h>
39#include <wchar.h>
40#include "mblocal.h"
41
42int	_GB18030_init(_RuneLocale *);
43size_t	_GB18030_mbrtowc(wchar_t * __restrict, const char * __restrict, size_t,
44	    mbstate_t * __restrict);
45int	_GB18030_mbsinit(const mbstate_t *);
46size_t	_GB18030_wcrtomb(char * __restrict, wchar_t, mbstate_t * __restrict);
47
48typedef struct {
49	int	count;
50	u_char	bytes[4];
51} _GB18030State;
52
53int
54_GB18030_init(_RuneLocale *rl)
55{
56
57	__mbrtowc = _GB18030_mbrtowc;
58	__wcrtomb = _GB18030_wcrtomb;
59	__mbsinit = _GB18030_mbsinit;
60	_CurrentRuneLocale = rl;
61	__mb_cur_max = 4;
62
63	return (0);
64}
65
66int
67_GB18030_mbsinit(const mbstate_t *ps)
68{
69
70	return (ps == NULL || ((const _GB18030State *)ps)->count == 0);
71}
72
73size_t
74_GB18030_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s,
75    size_t n, mbstate_t * __restrict ps)
76{
77	_GB18030State *gs;
78	wchar_t wch;
79	int ch, len, ocount;
80	size_t ncopy;
81
82	gs = (_GB18030State *)ps;
83
84	if (gs->count < 0 || gs->count > sizeof(gs->bytes)) {
85		errno = EINVAL;
86		return ((size_t)-1);
87	}
88
89	if (s == NULL) {
90		s = "";
91		n = 1;
92		pwc = NULL;
93	}
94
95	ncopy = MIN(MIN(n, MB_CUR_MAX), sizeof(gs->bytes) - gs->count);
96	memcpy(gs->bytes + gs->count, s, ncopy);
97	ocount = gs->count;
98	gs->count += ncopy;
99	s = (char *)gs->bytes;
100	n = gs->count;
101
102	if (n == 0)
103		/* Incomplete multibyte sequence */
104		return ((size_t)-2);
105
106	/*
107	 * Single byte:		[00-7f]
108	 * Two byte:		[81-fe][40-7e,80-fe]
109	 * Four byte:		[81-fe][30-39][81-fe][30-39]
110	 */
111	ch = (unsigned char)*s++;
112	if (ch <= 0x7f) {
113		len = 1;
114		wch = ch;
115	} else if (ch >= 0x81 && ch <= 0xfe) {
116		wch = ch;
117		if (n < 2)
118			return ((size_t)-2);
119		ch = (unsigned char)*s++;
120		if ((ch >= 0x40 && ch <= 0x7e) || (ch >= 0x80 && ch <= 0xfe)) {
121			wch = (wch << 8) | ch;
122			len = 2;
123		} else if (ch >= 0x30 && ch <= 0x39) {
124			/*
125			 * Strip high bit off the wide character we will
126			 * eventually output so that it is positive when
127			 * cast to wint_t on 32-bit twos-complement machines.
128			 */
129			wch = ((wch & 0x7f) << 8) | ch;
130			if (n < 3)
131				return ((size_t)-2);
132			ch = (unsigned char)*s++;
133			if (ch < 0x81 || ch > 0xfe)
134				goto ilseq;
135			wch = (wch << 8) | ch;
136			if (n < 4)
137				return ((size_t)-2);
138			ch = (unsigned char)*s++;
139			if (ch < 0x30 || ch > 0x39)
140				goto ilseq;
141			wch = (wch << 8) | ch;
142			len = 4;
143		} else
144			goto ilseq;
145	} else
146		goto ilseq;
147
148	if (pwc != NULL)
149		*pwc = wch;
150	gs->count = 0;
151	return (wch == L'\0' ? 0 : len - ocount);
152ilseq:
153	errno = EILSEQ;
154	return ((size_t)-1);
155}
156
157size_t
158_GB18030_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps)
159{
160	_GB18030State *gs;
161	size_t len;
162	int c;
163
164	gs = (_GB18030State *)ps;
165
166	if (gs->count != 0) {
167		errno = EINVAL;
168		return ((size_t)-1);
169	}
170
171	if (s == NULL)
172		/* Reset to initial shift state (no-op) */
173		return (1);
174	if ((wc & ~0x7fffffff) != 0)
175		goto ilseq;
176	if (wc & 0x7f000000) {
177		/* Replace high bit that mbrtowc() removed. */
178		wc |= 0x80000000;
179		c = (wc >> 24) & 0xff;
180		if (c < 0x81 || c > 0xfe)
181			goto ilseq;
182		*s++ = c;
183		c = (wc >> 16) & 0xff;
184		if (c < 0x30 || c > 0x39)
185			goto ilseq;
186		*s++ = c;
187		c = (wc >> 8) & 0xff;
188		if (c < 0x81 || c > 0xfe)
189			goto ilseq;
190		*s++ = c;
191		c = wc & 0xff;
192		if (c < 0x30 || c > 0x39)
193			goto ilseq;
194		*s++ = c;
195		len = 4;
196	} else if (wc & 0x00ff0000)
197		goto ilseq;
198	else if (wc & 0x0000ff00) {
199		c = (wc >> 8) & 0xff;
200		if (c < 0x81 || c > 0xfe)
201			goto ilseq;
202		*s++ = c;
203		c = wc & 0xff;
204		if (c < 0x40 || c == 0x7f || c == 0xff)
205			goto ilseq;
206		*s++ = c;
207		len = 2;
208	} else if (wc <= 0x7f) {
209		*s++ = wc;
210		len = 1;
211	} else
212		goto ilseq;
213
214	return (len);
215ilseq:
216	errno = EILSEQ;
217	return ((size_t)-1);
218}
219