xref: /netbsd-src/sys/fs/hfs/unicode.c (revision 0df165c04d0a9ca1adde9ed2b890344c937954a6)
1 /* $NetBSD: unicode.c,v 1.1.1.1 2007/03/06 00:10:39 dillo Exp $ */
2 
3 /*-
4  * Copyright (c) 2007 The NetBSD Foundation, Inc.
5  * All rights reserved.
6  *
7  * This code is derived from software contributed to The NetBSD Foundation
8  * by Dieter Baron.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
20  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
21  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
22  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
23  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
24  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
25  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
26  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
27  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
28  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
29  * POSSIBILITY OF SUCH DAMAGE.
30  */
31 
32 #include <sys/null.h>
33 
34 #include "unicode.h"
35 
36 size_t
37 utf8_to_utf16(uint16_t *dst, size_t dst_len,
38 	      const char *src, size_t src_len,
39 	      int flags, int *errp)
40 {
41     const unsigned char *s;
42     size_t spos, dpos;
43     int error;
44     uint16_t c;
45 
46 #define IS_CONT(c)	(((c)&0xc0) == 0x80)
47 
48     error = 0;
49     s = (const unsigned char *)src;
50     spos = dpos = 0;
51     while (spos<src_len) {
52 	if (s[spos] < 0x80)
53 	    c = s[spos++];
54 	else if ((flags & UNICODE_UTF8_LATIN1_FALLBACK)
55 		 && (spos >= src_len || !IS_CONT(s[spos+1]))
56 		 && s[spos]>=0xa0) {
57 	    /* not valid UTF-8, assume ISO 8859-1 */
58 	    c = s[spos++];
59 	}
60 	else if (s[spos] < 0xc0 || s[spos] >= 0xf5) {
61 	    /* continuation byte without lead byte
62 	       or lead byte for codepoint above 0x10ffff */
63 	    error++;
64 	    spos++;
65 	    continue;
66 	}
67 	else if (s[spos] < 0xe0) {
68 	    if (spos >= src_len || !IS_CONT(s[spos+1])) {
69 		spos++;
70 		error++;
71 		continue;
72 	    }
73 	    c = ((s[spos] & 0x3f) << 6) | (s[spos+1] & 0x3f);
74 	    spos += 2;
75 	    if (c < 0x80) {
76 		/* overlong encoding */
77 		error++;
78 		continue;
79 	    }
80 	}
81 	else if (s[spos] < 0xf0) {
82 	    if (spos >= src_len-2
83 		|| !IS_CONT(s[spos+1]) || !IS_CONT(s[spos+2])) {
84 		spos++;
85 		error++;
86 		continue;
87 	    }
88 	    c = ((s[spos] & 0x0f) << 12) | ((s[spos+1] & 0x3f) << 6)
89 		| (s[spos+2] & 0x3f);
90 	    spos += 3;
91 	    if (c < 0x800 || (c & 0xdf00) == 0xd800 ) {
92 		/* overlong encoding or encoded surrogate */
93 		error++;
94 		continue;
95 	    }
96 	}
97 	else {
98 	    uint32_t cc;
99 	    /* UTF-16 surrogate pair */
100 
101 	    if (spos >= src_len-3 || !IS_CONT(s[spos+1])
102 		|| !IS_CONT(s[spos+2]) || !IS_CONT(s[spos+3])) {
103 		spos++;
104 		error++;
105 
106 		continue;
107 	    }
108 	    cc = ((s[spos] & 0x03) << 18) | ((s[spos+1] & 0x3f) << 12)
109 		 | ((s[spos+2] & 0x3f) << 6) | (s[spos+3] & 0x3f);
110 	    spos += 4;
111 	    if (cc < 0x10000) {
112 		/* overlong encoding */
113 		error++;
114 		continue;
115 	    }
116 	    if (dst && dpos < dst_len)
117 		dst[dpos] = (0xd800 | ((cc-0x10000)>>10));
118 	    dpos++;
119 	    c = 0xdc00 | ((cc-0x10000) & 0x3ffff);
120 	}
121 
122 	if (dst && dpos < dst_len)
123 	    dst[dpos] = c;
124 	dpos++;
125     }
126 
127     if (errp)
128 	*errp = error;
129 
130     return dpos;
131 
132 #undef IS_CONT
133 }
134 
135 
136 size_t
137 utf16_to_utf8(char *dst, size_t dst_len,
138 	      const uint16_t *src, size_t src_len,
139 	      int flags, int *errp)
140 {
141     uint8_t spos, dpos;
142     int error;
143 
144 #define CHECK_LENGTH(l)	(dpos > dst_len-(l) ? dst=NULL : NULL)
145 #define ADD_BYTE(b)	(dst ? dst[dpos] = (b) : 0, dpos++)
146 
147     error = 0;
148     dpos = 0;
149     for (spos=0; spos<src_len; spos++) {
150 	if (src[spos] < 0x80) {
151 	    CHECK_LENGTH(1);
152 	    ADD_BYTE(src[spos]);
153 	}
154 	else if (src[spos] < 0x800) {
155 	    CHECK_LENGTH(2);
156 	    ADD_BYTE(0xc0 | (src[spos]>>6));
157 	    ADD_BYTE(0x80 | (src[spos] & 0x3f));
158 	}
159 	else if ((src[spos] & 0xdc00) == 0xd800) {
160 	    uint32_t c;
161 	    /* first surrogate */
162 	    if (spos == src_len - 1 || (src[spos] & 0xdc00) != 0xdc00) {
163 		/* no second surrogate present */
164 		error++;
165 		continue;
166 	    }
167 	    spos++;
168 	    CHECK_LENGTH(4);
169 	    c = (((src[spos]&0x3ff) << 10) | (src[spos+1]&0x3ff)) + 0x10000;
170 	    ADD_BYTE(0xf0 | (c>>18));
171 	    ADD_BYTE(0x80 | ((c>>12) & 0x3f));
172 	    ADD_BYTE(0x80 | ((c>>6) & 0x3f));
173 	    ADD_BYTE(0x80 | (c & 0x3f));
174 	}
175 	else if ((src[spos] & 0xdc00) == 0xdc00) {
176 	    /* second surrogate without preceding first surrogate */
177 	    error++;
178 	}
179 	else {
180 	    CHECK_LENGTH(3);
181 	    ADD_BYTE(0xe0 | src[spos]>>12);
182 	    ADD_BYTE(0x80 | ((src[spos]>>6) & 0x3f));
183 	    ADD_BYTE(0x80 | (src[spos] & 0x3f));
184 	}
185     }
186 
187     if (errp)
188 	*errp = error;
189 
190     return dpos;
191 
192 #undef ADD_BYTE
193 #undef CHECK_LENGTH
194 }
195