xref: /dragonfly/lib/libc/locale/gbk.c (revision cb40c8cc)
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  * 3. 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 
41 #include <sys/types.h>
42 #include <errno.h>
43 #include <runetype.h>
44 #include <stdlib.h>
45 #include <string.h>
46 #include <wchar.h>
47 #include "mblocal.h"
48 
49 static size_t	_GBK_mbrtowc(wchar_t * __restrict, const char * __restrict,
50 		    size_t, mbstate_t * __restrict);
51 static int	_GBK_mbsinit(const mbstate_t *);
52 static size_t	_GBK_wcrtomb(char * __restrict, wchar_t,
53 		    mbstate_t * __restrict);
54 static size_t	_GBK_mbsnrtowcs(wchar_t * __restrict,
55 		    const char ** __restrict, size_t, size_t,
56 		    mbstate_t * __restrict);
57 static size_t	_GBK_wcsnrtombs(char * __restrict,
58 		    const wchar_t ** __restrict, size_t, size_t,
59 		    mbstate_t * __restrict);
60 
61 typedef struct {
62 	wchar_t	ch;
63 } _GBKState;
64 
65 int
_GBK_init(struct xlocale_ctype * l,_RuneLocale * rl)66 _GBK_init(struct xlocale_ctype *l, _RuneLocale *rl)
67 {
68 
69 	l->__mbrtowc = _GBK_mbrtowc;
70 	l->__wcrtomb = _GBK_wcrtomb;
71 	l->__mbsinit = _GBK_mbsinit;
72 	l->__mbsnrtowcs = _GBK_mbsnrtowcs;
73 	l->__wcsnrtombs = _GBK_wcsnrtombs;
74 	l->runes = rl;
75 	l->__mb_cur_max = 2;
76 	l->__mb_sb_limit = 128;
77 	return (0);
78 }
79 
80 static int
_GBK_mbsinit(const mbstate_t * ps)81 _GBK_mbsinit(const mbstate_t *ps)
82 {
83 
84 	return (ps == NULL || ((const _GBKState *)ps)->ch == 0);
85 }
86 
87 static int
_gbk_check(u_int c)88 _gbk_check(u_int c)
89 {
90 
91 	c &= 0xff;
92 	return ((c >= 0x81 && c <= 0xfe) ? 2 : 1);
93 }
94 
95 static size_t
_GBK_mbrtowc(wchar_t * __restrict pwc,const char * __restrict s,size_t n,mbstate_t * __restrict ps)96 _GBK_mbrtowc(wchar_t * __restrict pwc, const char * __restrict s, size_t n,
97     mbstate_t * __restrict ps)
98 {
99 	_GBKState *gs;
100 	wchar_t wc;
101 	size_t len;
102 
103 	gs = (_GBKState *)ps;
104 
105 	if ((gs->ch & ~0xFF) != 0) {
106 		/* Bad conversion state. */
107 		errno = EINVAL;
108 		return ((size_t)-1);
109 	}
110 
111 	if (s == NULL) {
112 		s = "";
113 		n = 1;
114 		pwc = NULL;
115 	}
116 
117 	if (n == 0)
118 		/* Incomplete multibyte sequence */
119 		return ((size_t)-2);
120 
121 	if (gs->ch != 0) {
122 		if (*s == '\0') {
123 			errno = EILSEQ;
124 			return ((size_t)-1);
125 		}
126 		wc = (gs->ch << 8) | (*s & 0xFF);
127 		if (pwc != NULL)
128 			*pwc = wc;
129 		gs->ch = 0;
130 		return (1);
131 	}
132 
133 	len = (size_t)_gbk_check(*s);
134 	wc = *s++ & 0xff;
135 	if (len == 2) {
136 		if (n < 2) {
137 			/* Incomplete multibyte sequence */
138 			gs->ch = wc;
139 			return ((size_t)-2);
140 		}
141 		if (*s == '\0') {
142 			errno = EILSEQ;
143 			return ((size_t)-1);
144 		}
145 		wc = (wc << 8) | (*s++ & 0xff);
146 		if (pwc != NULL)
147 			*pwc = wc;
148 		return (2);
149 	} else {
150 		if (pwc != NULL)
151 			*pwc = wc;
152 		return (wc == L'\0' ? 0 : 1);
153 	}
154 }
155 
156 static size_t
_GBK_wcrtomb(char * __restrict s,wchar_t wc,mbstate_t * __restrict ps)157 _GBK_wcrtomb(char * __restrict s, wchar_t wc, mbstate_t * __restrict ps)
158 {
159 	_GBKState *gs;
160 
161 	gs = (_GBKState *)ps;
162 
163 	if (gs->ch != 0) {
164 		errno = EINVAL;
165 		return ((size_t)-1);
166 	}
167 
168 	if (s == NULL)
169 		/* Reset to initial shift state (no-op) */
170 		return (1);
171 	if (wc & 0x8000) {
172 		*s++ = (wc >> 8) & 0xff;
173 		*s = wc & 0xff;
174 		return (2);
175 	}
176 	*s = wc & 0xff;
177 	return (1);
178 }
179 
180 static size_t
_GBK_mbsnrtowcs(wchar_t * __restrict dst,const char ** __restrict src,size_t nms,size_t len,mbstate_t * __restrict ps)181 _GBK_mbsnrtowcs(wchar_t * __restrict dst, const char ** __restrict src,
182     size_t nms, size_t len, mbstate_t * __restrict ps)
183 {
184 	return (__mbsnrtowcs_std(dst, src, nms, len, ps, _GBK_mbrtowc));
185 }
186 
187 static size_t
_GBK_wcsnrtombs(char * __restrict dst,const wchar_t ** __restrict src,size_t nwc,size_t len,mbstate_t * __restrict ps)188 _GBK_wcsnrtombs(char * __restrict dst, const wchar_t ** __restrict src,
189     size_t nwc, size_t len, mbstate_t * __restrict ps)
190 {
191 	return (__wcsnrtombs_std(dst, src, nwc, len, ps, _GBK_wcrtomb));
192 }
193