openssl/crypto/asn1/a_utf8.c
Beat Bolli ba64e5a92a ASN1: limit the Unicode code point range in UTF8_getc() and UTF8_putc()
Since the Unicode 4.0.0 standard, the valid code point range is U+0000
to U+10FFFF. Make code points outside this range invalid when converting
from/to UTF-8.

Signed-off-by: Beat Bolli <dev@drbeat.li>

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14185)
2021-03-18 14:12:48 +10:00

138 lines
4.0 KiB
C

/*
* Copyright 1995-2021 The OpenSSL Project Authors. All Rights Reserved.
*
* Licensed under the Apache License 2.0 (the "License"). You may not use
* this file except in compliance with the License. You can obtain a copy
* in the file LICENSE in the source distribution or at
* https://www.openssl.org/source/license.html
*/
#include <stdio.h>
#include "internal/cryptlib.h"
#include "internal/unicode.h"
#include <openssl/asn1.h>
/* UTF8 utilities */
/*-
* This parses a UTF8 string one character at a time. It is passed a pointer
* to the string and the length of the string. It sets 'value' to the value of
* the current character. It returns the number of characters read or a
* negative error code:
* -1 = string too short
* -2 = illegal character
* -3 = subsequent characters not of the form 10xxxxxx
* -4 = character encoded incorrectly (not minimal length).
*/
int UTF8_getc(const unsigned char *str, int len, unsigned long *val)
{
const unsigned char *p;
unsigned long value;
int ret;
if (len <= 0)
return 0;
p = str;
/* Check syntax and work out the encoded value (if correct) */
if ((*p & 0x80) == 0) {
value = *p++ & 0x7f;
ret = 1;
} else if ((*p & 0xe0) == 0xc0) {
if (len < 2)
return -1;
if ((p[1] & 0xc0) != 0x80)
return -3;
value = (*p++ & 0x1f) << 6;
value |= *p++ & 0x3f;
if (value < 0x80)
return -4;
ret = 2;
} else if ((*p & 0xf0) == 0xe0) {
if (len < 3)
return -1;
if (((p[1] & 0xc0) != 0x80)
|| ((p[2] & 0xc0) != 0x80))
return -3;
value = (*p++ & 0xf) << 12;
value |= (*p++ & 0x3f) << 6;
value |= *p++ & 0x3f;
if (value < 0x800)
return -4;
if (is_unicode_surrogate(value))
return -2;
ret = 3;
} else if ((*p & 0xf8) == 0xf0) {
if (len < 4)
return -1;
if (((p[1] & 0xc0) != 0x80)
|| ((p[2] & 0xc0) != 0x80)
|| ((p[3] & 0xc0) != 0x80))
return -3;
value = ((unsigned long)(*p++ & 0x7)) << 18;
value |= (*p++ & 0x3f) << 12;
value |= (*p++ & 0x3f) << 6;
value |= *p++ & 0x3f;
if (value < 0x10000)
return -4;
ret = 4;
} else
return -2;
*val = value;
return ret;
}
/*
* This takes a character 'value' and writes the UTF8 encoded value in 'str'
* where 'str' is a buffer containing 'len' characters. Returns the number of
* characters written, -1 if 'len' is too small or -2 if 'value' is out of
* range. 'str' can be set to NULL in which case it just returns the number of
* characters. It will need at most 4 characters.
*/
int UTF8_putc(unsigned char *str, int len, unsigned long value)
{
if (!str)
len = 4; /* Maximum we will need */
else if (len <= 0)
return -1;
if (value < 0x80) {
if (str)
*str = (unsigned char)value;
return 1;
}
if (value < 0x800) {
if (len < 2)
return -1;
if (str) {
*str++ = (unsigned char)(((value >> 6) & 0x1f) | 0xc0);
*str = (unsigned char)((value & 0x3f) | 0x80);
}
return 2;
}
if (value < 0x10000) {
if (is_unicode_surrogate(value))
return -2;
if (len < 3)
return -1;
if (str) {
*str++ = (unsigned char)(((value >> 12) & 0xf) | 0xe0);
*str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
*str = (unsigned char)((value & 0x3f) | 0x80);
}
return 3;
}
if (value < UNICODE_LIMIT) {
if (len < 4)
return -1;
if (str) {
*str++ = (unsigned char)(((value >> 18) & 0x7) | 0xf0);
*str++ = (unsigned char)(((value >> 12) & 0x3f) | 0x80);
*str++ = (unsigned char)(((value >> 6) & 0x3f) | 0x80);
*str = (unsigned char)((value & 0x3f) | 0x80);
}
return 4;
}
return -2;
}