xref: /illumos-gate/usr/src/lib/libc/port/locale/big5.c (revision 2ad72058bc680e30f084e9f9ba8ceb0f77386821)
1 /*
2  * Copyright (c) 2002-2004 Tim J. Robbins. All rights reserved.
3  * Copyright (c) 1993
4  *	The Regents of the University of California.  All rights reserved.
5  *
6  * This code is derived from software contributed to Berkeley by
7  * Paul Borman at Krystal Technologies.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  * 1. Redistributions of source code must retain the above copyright
13  *    notice, this list of conditions and the following disclaimer.
14  * 2. Redistributions in binary form must reproduce the above copyright
15  *    notice, this list of conditions and the following disclaimer in the
16  *    documentation and/or other materials provided with the distribution.
17  * 3. All advertising materials mentioning features or use of this software
18  *    must display the following acknowledgement:
19  *	This product includes software developed by the University of
20  *	California, Berkeley and its contributors.
21  * 4. Neither the name of the University nor the names of its contributors
22  *    may be used to endorse or promote products derived from this software
23  *    without specific prior written permission.
24  *
25  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
26  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
27  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
28  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
29  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
30  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
31  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
32  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
33  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
34  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
35  * SUCH DAMAGE.
36  */
37 
38 /*
39  * Copyright 2010 Nexenta Systems, Inc.  All rights reserved.
40  * Use is subject to license terms.
41  */
42 
43 #include "lint.h"
44 #include <sys/types.h>
45 #include <errno.h>
46 #include "runetype.h"
47 #include <stdlib.h>
48 #include <string.h>
49 #include <wchar.h>
50 #include "mblocal.h"
51 
52 static size_t	_BIG5_mbrtowc(wchar_t *_RESTRICT_KYWD,
53 		    const char *_RESTRICT_KYWD,
54 		    size_t, mbstate_t *RESTRICT_KYWD);
55 static int	_BIG5_mbsinit(const mbstate_t *);
56 static size_t	_BIG5_wcrtomb(char *_RESTRICT_KYWD, wchar_t,
57 		    mbstate_t *_RESTRICT_KYWD);
58 
59 typedef struct {
60 	wchar_t	ch;
61 } _BIG5State;
62 
63 int
64 _BIG5_init(_RuneLocale *rl)
65 {
66 
67 	__mbrtowc = _BIG5_mbrtowc;
68 	__wcrtomb = _BIG5_wcrtomb;
69 	__mbsinit = _BIG5_mbsinit;
70 	_CurrentRuneLocale = rl;
71 	__ctype[520] = 2;
72 	charset_is_ascii = 0;
73 	return (0);
74 }
75 
76 static int
77 _BIG5_mbsinit(const mbstate_t *ps)
78 {
79 
80 	return (ps == NULL || ((const _BIG5State *)ps)->ch == 0);
81 }
82 
83 static int
84 _big5_check(uint_t c)
85 {
86 
87 	c &= 0xff;
88 	return ((c >= 0xa1 && c <= 0xfe) ? 2 : 1);
89 }
90 
91 static size_t
92 _BIG5_mbrtowc(wchar_t *_RESTRICT_KYWD pwc, const char *_RESTRICT_KYWD s,
93     size_t n, mbstate_t *_RESTRICT_KYWD ps)
94 {
95 	_BIG5State *bs;
96 	wchar_t wc;
97 	size_t len;
98 
99 	bs = (_BIG5State *)ps;
100 
101 	if ((bs->ch & ~0xFF) != 0) {
102 		/* Bad conversion state. */
103 		errno = EINVAL;
104 		return ((size_t)-1);
105 	}
106 
107 	if (s == NULL) {
108 		s = "";
109 		n = 1;
110 		pwc = NULL;
111 	}
112 
113 	if (n == 0)
114 		/* Incomplete multibyte sequence */
115 		return ((size_t)-2);
116 
117 	if (bs->ch != 0) {
118 		if (*s == '\0') {
119 			errno = EILSEQ;
120 			return ((size_t)-1);
121 		}
122 		wc = (bs->ch << 8) | (*s & 0xFF);
123 		if (pwc != NULL)
124 			*pwc = wc;
125 		bs->ch = 0;
126 		return (1);
127 	}
128 
129 	len = (size_t)_big5_check(*s);
130 	wc = *s++ & 0xff;
131 	if (len == 2) {
132 		if (n < 2) {
133 			/* Incomplete multibyte sequence */
134 			bs->ch = wc;
135 			return ((size_t)-2);
136 		}
137 		if (*s == '\0') {
138 			errno = EILSEQ;
139 			return ((size_t)-1);
140 		}
141 		wc = (wc << 8) | (*s++ & 0xff);
142 		if (pwc != NULL)
143 			*pwc = wc;
144 		return (2);
145 	} else {
146 		if (pwc != NULL)
147 			*pwc = wc;
148 		return (wc == L'\0' ? 0 : 1);
149 	}
150 }
151 
152 static size_t
153 _BIG5_wcrtomb(char *_RESTRICT_KYWD s, wchar_t wc, mbstate_t *_RESTRICT_KYWD ps)
154 {
155 	_BIG5State *bs;
156 
157 	bs = (_BIG5State *)ps;
158 
159 	if (bs->ch != 0) {
160 		errno = EINVAL;
161 		return ((size_t)-1);
162 	}
163 
164 	if (s == NULL)
165 		/* Reset to initial shift state (no-op) */
166 		return (1);
167 	if (wc & 0x8000) {
168 		*s++ = (wc >> 8) & 0xff;
169 		*s = wc & 0xff;
170 		return (2);
171 	}
172 	*s = wc & 0xff;
173 	return (1);
174 }
175