summaryrefslogtreecommitdiff
path: root/src/lib/libcrypto/asn1/a_utf8.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/lib/libcrypto/asn1/a_utf8.c')
-rw-r--r--src/lib/libcrypto/asn1/a_utf8.c144
1 files changed, 53 insertions, 91 deletions
diff --git a/src/lib/libcrypto/asn1/a_utf8.c b/src/lib/libcrypto/asn1/a_utf8.c
index c224db4c12..f5e4bec7e0 100644
--- a/src/lib/libcrypto/asn1/a_utf8.c
+++ b/src/lib/libcrypto/asn1/a_utf8.c
@@ -59,11 +59,13 @@
59#include <stdio.h> 59#include <stdio.h>
60#include "cryptlib.h" 60#include "cryptlib.h"
61#include <openssl/asn1.h> 61#include <openssl/asn1.h>
62#include "asn1_locl.h"
62 63
63 64
64/* UTF8 utilities */ 65/* UTF8 utilities */
65 66
66/* This parses a UTF8 string one character at a time. It is passed a pointer 67/*
68 * This parses a UTF8 string one character at a time. It is passed a pointer
67 * to the string and the length of the string. It sets 'value' to the value of 69 * to the string and the length of the string. It sets 'value' to the value of
68 * the current character. It returns the number of characters read or a 70 * the current character. It returns the number of characters read or a
69 * negative error code: 71 * negative error code:
@@ -88,6 +90,8 @@ UTF8_getc(const unsigned char *str, int len, unsigned long *val)
88 value = *p++ & 0x7f; 90 value = *p++ & 0x7f;
89 ret = 1; 91 ret = 1;
90 } else if ((*p & 0xe0) == 0xc0) { 92 } else if ((*p & 0xe0) == 0xc0) {
93 if (*p < 0xc2)
94 return -2;
91 if (len < 2) 95 if (len < 2)
92 return -1; 96 return -1;
93 if ((p[1] & 0xc0) != 0x80) 97 if ((p[1] & 0xc0) != 0x80)
@@ -108,8 +112,11 @@ UTF8_getc(const unsigned char *str, int len, unsigned long *val)
108 value |= *p++ & 0x3f; 112 value |= *p++ & 0x3f;
109 if (value < 0x800) 113 if (value < 0x800)
110 return -4; 114 return -4;
115 /* surrogate pair code points are not valid */
116 if (value >= 0xd800 && value < 0xe000)
117 return -2;
111 ret = 3; 118 ret = 3;
112 } else if ((*p & 0xf8) == 0xf0) { 119 } else if ((*p & 0xf8) == 0xf0 && (*p < 0xf5)) {
113 if (len < 4) 120 if (len < 4)
114 return -1; 121 return -1;
115 if (((p[1] & 0xc0) != 0x80) || 122 if (((p[1] & 0xc0) != 0x80) ||
@@ -122,116 +129,71 @@ UTF8_getc(const unsigned char *str, int len, unsigned long *val)
122 value |= *p++ & 0x3f; 129 value |= *p++ & 0x3f;
123 if (value < 0x10000) 130 if (value < 0x10000)
124 return -4; 131 return -4;
132 if (value > UNICODE_MAX)
133 return -2;
125 ret = 4; 134 ret = 4;
126 } else if ((*p & 0xfc) == 0xf8) { 135 } else
127 if (len < 5) 136 return -2;
128 return -1; 137 *val = value;
129 if (((p[1] & 0xc0) != 0x80) ||
130 ((p[2] & 0xc0) != 0x80) ||
131 ((p[3] & 0xc0) != 0x80) ||
132 ((p[4] & 0xc0) != 0x80))
133 return -3;
134 value = ((unsigned long)(*p++ & 0x3)) << 24;
135 value |= ((unsigned long)(*p++ & 0x3f)) << 18;
136 value |= ((unsigned long)(*p++ & 0x3f)) << 12;
137 value |= (*p++ & 0x3f) << 6;
138 value |= *p++ & 0x3f;
139 if (value < 0x200000)
140 return -4;
141 ret = 5;
142 } else if ((*p & 0xfe) == 0xfc) {
143 if (len < 6)
144 return -1;
145 if (((p[1] & 0xc0) != 0x80) ||
146 ((p[2] & 0xc0) != 0x80) ||
147 ((p[3] & 0xc0) != 0x80) ||
148 ((p[4] & 0xc0) != 0x80) ||
149 ((p[5] & 0xc0) != 0x80))
150 return -3;
151 value = ((unsigned long)(*p++ & 0x1)) << 30;
152 value |= ((unsigned long)(*p++ & 0x3f)) << 24;
153 value |= ((unsigned long)(*p++ & 0x3f)) << 18;
154 value |= ((unsigned long)(*p++ & 0x3f)) << 12;
155 value |= (*p++ & 0x3f) << 6;
156 value |= *p++ & 0x3f;
157 if (value < 0x4000000)
158 return -4;
159 ret = 6;
160 } else return -2;
161 *val = value;
162 return ret; 138 return ret;
163} 139}
164 140
165/* This takes a character 'value' and writes the UTF8 encoded value in 141/* This takes a Unicode code point 'value' and writes its UTF-8 encoded form
166 * 'str' where 'str' is a buffer containing 'len' characters. Returns 142 * in 'str' where 'str' is a buffer of at least length 'len'. If 'str'
167 * the number of characters written or -1 if 'len' is too small. 'str' can 143 * is NULL, then nothing is written and just the return code is determined.
168 * be set to NULL in which case it just returns the number of characters. 144
169 * It will need at most 6 characters. 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.
170 */ 153 */
171 154
172int 155int
173UTF8_putc(unsigned char *str, int len, unsigned long value) 156UTF8_putc(unsigned char *str, int len, unsigned long value)
174{ 157{
175 if (!str)
176 len = 6; /* Maximum we will need */
177 else if (len <= 0)
178 return -1;
179 if (value < 0x80) { 158 if (value < 0x80) {
180 if (str) 159 if (str != NULL) {
181 *str = (unsigned char)value; 160 if (len < 1)
161 return -1;
162 str[0] = (unsigned char)value;
163 }
182 return 1; 164 return 1;
183 } 165 }
184 if (value < 0x800) { 166 if (value < 0x800) {
185 if (len < 2) 167 if (str != NULL) {
186 return -1; 168 if (len < 2)
187 if (str) { 169 return -1;
188 *str++ = (unsigned char)(((value >> 6) & 0x1f) | 0xc0); 170 str[0] = (unsigned char)(((value >> 6) & 0x1f) | 0xc0);
189 *str = (unsigned char)((value & 0x3f) | 0x80); 171 str[1] = (unsigned char)((value & 0x3f) | 0x80);
190 } 172 }
191 return 2; 173 return 2;
192 } 174 }
193 if (value < 0x10000) { 175 if (value < 0x10000) {
194 if (len < 3) 176 if (UNICODE_IS_SURROGATE(value))
195 return -1; 177 return -2;
196 if (str) { 178 if (str != NULL) {
197 *str++ = (unsigned char)(((value >> 12) & 0xf) | 0xe0); 179 if (len < 3)
198 *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80); 180 return -1;
199 *str = (unsigned char)((value & 0x3f) | 0x80); 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);
200 } 184 }
201 return 3; 185 return 3;
202 } 186 }
203 if (value < 0x200000) { 187 if (value <= UNICODE_MAX) {
204 if (len < 4) 188 if (str != NULL) {
205 return -1; 189 if (len < 4)
206 if (str) { 190 return -1;
207 *str++ = (unsigned char)(((value >> 18) & 0x7) | 0xf0); 191 str[0] = (unsigned char)(((value >> 18) & 0x7) | 0xf0);
208 *str++ = (unsigned char)(((value >> 12) & 0x3f) | 0x80); 192 str[1] = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
209 *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80); 193 str[2] = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
210 *str = (unsigned char)((value & 0x3f) | 0x80); 194 str[3] = (unsigned char)((value & 0x3f) | 0x80);
211 } 195 }
212 return 4; 196 return 4;
213 } 197 }
214 if (value < 0x4000000) { 198 return -2;
215 if (len < 5)
216 return -1;
217 if (str) {
218 *str++ = (unsigned char)(((value >> 24) & 0x3) | 0xf8);
219 *str++ = (unsigned char)(((value >> 18) & 0x3f) | 0x80);
220 *str++ = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
221 *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
222 *str = (unsigned char)((value & 0x3f) | 0x80);
223 }
224 return 5;
225 }
226 if (len < 6)
227 return -1;
228 if (str) {
229 *str++ = (unsigned char)(((value >> 30) & 0x1) | 0xfc);
230 *str++ = (unsigned char)(((value >> 24) & 0x3f) | 0x80);
231 *str++ = (unsigned char)(((value >> 18) & 0x3f) | 0x80);
232 *str++ = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
233 *str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
234 *str = (unsigned char)((value & 0x3f) | 0x80);
235 }
236 return 6;
237} 199}
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-08-15 08:32:30 +0000