xref: /dragonfly/crypto/libressl/crypto/asn1/a_utf8.c (revision e6d22e9b)
1 /* $OpenBSD: a_utf8.c,v 1.8 2014/07/11 08:44:47 jsing Exp $ */
2 /* Copyright (C) 1995-1998 Eric Young (eay@cryptsoft.com)
3  * All rights reserved.
4  *
5  * This package is an SSL implementation written
6  * by Eric Young (eay@cryptsoft.com).
7  * The implementation was written so as to conform with Netscapes SSL.
8  *
9  * This library is free for commercial and non-commercial use as long as
10  * the following conditions are aheared to.  The following conditions
11  * apply to all code found in this distribution, be it the RC4, RSA,
12  * lhash, DES, etc., code; not just the SSL code.  The SSL documentation
13  * included with this distribution is covered by the same copyright terms
14  * except that the holder is Tim Hudson (tjh@cryptsoft.com).
15  *
16  * Copyright remains Eric Young's, and as such any Copyright notices in
17  * the code are not to be removed.
18  * If this package is used in a product, Eric Young should be given attribution
19  * as the author of the parts of the library used.
20  * This can be in the form of a textual message at program startup or
21  * in documentation (online or textual) provided with the package.
22  *
23  * Redistribution and use in source and binary forms, with or without
24  * modification, are permitted provided that the following conditions
25  * are met:
26  * 1. Redistributions of source code must retain the copyright
27  *    notice, this list of conditions and the following disclaimer.
28  * 2. Redistributions in binary form must reproduce the above copyright
29  *    notice, this list of conditions and the following disclaimer in the
30  *    documentation and/or other materials provided with the distribution.
31  * 3. All advertising materials mentioning features or use of this software
32  *    must display the following acknowledgement:
33  *    "This product includes cryptographic software written by
34  *     Eric Young (eay@cryptsoft.com)"
35  *    The word 'cryptographic' can be left out if the rouines from the library
36  *    being used are not cryptographic related :-).
37  * 4. If you include any Windows specific code (or a derivative thereof) from
38  *    the apps directory (application code) you must include an acknowledgement:
39  *    "This product includes software written by Tim Hudson (tjh@cryptsoft.com)"
40  *
41  * THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND
42  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
43  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
44  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
45  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
46  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
47  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
48  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
49  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
50  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
51  * SUCH DAMAGE.
52  *
53  * The licence and distribution terms for any publically available version or
54  * derivative of this code cannot be changed.  i.e. this code cannot simply be
55  * copied and put under another distribution licence
56  * [including the GNU Public Licence.]
57  */
58 
59 #include <stdio.h>
60 
61 #include <openssl/asn1.h>
62 
63 #include "asn1_locl.h"
64 
65 /* UTF8 utilities */
66 
67 /*
68  * This parses a UTF8 string one character at a time. It is passed a pointer
69  * to the string and the length of the string. It sets 'value' to the value of
70  * the current character. It returns the number of characters read or a
71  * negative error code:
72  * -1 = string too short
73  * -2 = illegal character
74  * -3 = subsequent characters not of the form 10xxxxxx
75  * -4 = character encoded incorrectly (not minimal length).
76  */
77 
78 int
79 UTF8_getc(const unsigned char *str, int len, unsigned long *val)
80 {
81 	const unsigned char *p;
82 	unsigned long value;
83 	int ret;
84 	if (len <= 0)
85 		return 0;
86 	p = str;
87 
88 	/* Check syntax and work out the encoded value (if correct) */
89 	if ((*p & 0x80) == 0) {
90 		value = *p++ & 0x7f;
91 		ret = 1;
92 	} else if ((*p & 0xe0) == 0xc0) {
93 		if (*p < 0xc2)
94 			return -2;
95 		if (len < 2)
96 			return -1;
97 		if ((p[1] & 0xc0) != 0x80)
98 			return -3;
99 		value = (*p++ & 0x1f) << 6;
100 		value |= *p++ & 0x3f;
101 		if (value < 0x80)
102 			return -4;
103 		ret = 2;
104 	} else if ((*p & 0xf0) == 0xe0) {
105 		if (len < 3)
106 			return -1;
107 		if (((p[1] & 0xc0) != 0x80) ||
108 		    ((p[2] & 0xc0) != 0x80))
109 			return -3;
110 		value = (*p++ & 0xf) << 12;
111 		value |= (*p++ & 0x3f) << 6;
112 		value |= *p++ & 0x3f;
113 		if (value < 0x800)
114 			return -4;
115 		/* surrogate pair code points are not valid */
116 		if (value >= 0xd800 && value < 0xe000)
117 			return -2;
118 		ret = 3;
119 	} else if ((*p & 0xf8) == 0xf0 && (*p < 0xf5)) {
120 		if (len < 4)
121 			return -1;
122 		if (((p[1] & 0xc0) != 0x80) ||
123 		    ((p[2] & 0xc0) != 0x80) ||
124 		    ((p[3] & 0xc0) != 0x80))
125 			return -3;
126 		value = ((unsigned long)(*p++ & 0x7)) << 18;
127 		value |= (*p++ & 0x3f) << 12;
128 		value |= (*p++ & 0x3f) << 6;
129 		value |= *p++ & 0x3f;
130 		if (value < 0x10000)
131 			return -4;
132 		if (value > UNICODE_MAX)
133 			return -2;
134 		ret = 4;
135 	} else
136 		return -2;
137 	*val = value;
138 	return ret;
139 }
140 
141 /* This takes a Unicode code point 'value' and writes its UTF-8 encoded form
142  * in 'str' where 'str' is a buffer of at least length 'len'.  If 'str'
143  * is NULL, then nothing is written and just the return code is determined.
144 
145  * Returns less than zero on error:
146  *  -1 if 'str' is not NULL and 'len' is too small
147  *  -2 if 'value' is an invalid character (surrogate or out-of-range)
148  *
149  * Otherwise, returns the number of bytes in 'value's encoded form
150  * (i.e., the number of bytes written to 'str' when it's not NULL).
151  *
152  * It will need at most 4 characters.
153  */
154 
155 int
156 UTF8_putc(unsigned char *str, int len, unsigned long value)
157 {
158 	if (value < 0x80) {
159 		if (str != NULL) {
160 			if (len < 1)
161 				return -1;
162 			str[0] = (unsigned char)value;
163 		}
164 		return 1;
165 	}
166 	if (value < 0x800) {
167 		if (str != NULL) {
168 			if (len < 2)
169 				return -1;
170 			str[0] = (unsigned char)(((value >> 6) & 0x1f) | 0xc0);
171 			str[1] = (unsigned char)((value & 0x3f) | 0x80);
172 		}
173 		return 2;
174 	}
175 	if (value < 0x10000) {
176 		if (UNICODE_IS_SURROGATE(value))
177 			return -2;
178 		if (str != NULL) {
179 			if (len < 3)
180 				return -1;
181 			str[0] = (unsigned char)(((value >> 12) & 0xf) | 0xe0);
182 			str[1] = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
183 			str[2] = (unsigned char)((value & 0x3f) | 0x80);
184 		}
185 		return 3;
186 	}
187 	if (value <= UNICODE_MAX) {
188 		if (str != NULL) {
189 			if (len < 4)
190 				return -1;
191 			str[0] = (unsigned char)(((value >> 18) & 0x7) | 0xf0);
192 			str[1] = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
193 			str[2] = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
194 			str[3] = (unsigned char)((value & 0x3f) | 0x80);
195 		}
196 		return 4;
197 	}
198 	return -2;
199 }
200