static void test_utf8_encoded_valid_unichar(void) {
assert_se(utf8_encoded_valid_unichar("\342\204\242") == 3);
assert_se(utf8_encoded_valid_unichar("\302\256") == 2);
assert_se(utf8_encoded_valid_unichar("a") == 1);
assert_se(utf8_encoded_valid_unichar("\341\204") < 0);
assert_se(utf8_encoded_valid_unichar("\341\204\341\204") < 0);
}
char *utf8_escape_non_printable(const char *str) {
char *p, *s;
assert(str);
p = s = malloc(strlen(str) * 4 + 1);
if (!p)
return NULL;
while (*str) {
int len;
len = utf8_encoded_valid_unichar(str);
if (len > 0) {
if (utf8_is_printable(str, len)) {
s = mempcpy(s, str, len);
str += len;
} else {
while (len > 0) {
*(s++) = '\\';
*(s++) = 'x';
*(s++) = hexchar((int) *str >> 4);
*(s++) = hexchar((int) *str);
str += 1;
len --;
}
}
} else {
s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER);
str += 1;
}
}
char *utf8_escape_invalid(const char *str) {
char *p, *s;
assert(str);
p = s = malloc(strlen(str) * 4 + 1);
if (!p)
return NULL;
while (*str) {
int len;
len = utf8_encoded_valid_unichar(str);
if (len > 0) {
s = mempcpy(s, str, len);
str += len;
} else {
s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER);
str += 1;
}
}
*s = '\0';
return p;
}
bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
const char *p;
assert(str);
for (p = str; length;) {
int encoded_len, val;
encoded_len = utf8_encoded_valid_unichar(p);
if (encoded_len < 0 ||
(size_t) encoded_len > length)
return false;
val = utf8_encoded_to_unichar(p);
if (val < 0 ||
unichar_is_control(val) ||
(!newline && val == '\n'))
return false;
length -= encoded_len;
p += encoded_len;
}
return true;
}
/* replace everything but whitelisted plain ascii and valid utf8 */
int replace_untrusted_chars(char *str)
{
size_t i = 0;
int replaced = 0;
while (str[i] != '\0') {
int len;
/* valid printable ascii char */
if ((str[i] >= '0' && str[i] <= '9') ||
(str[i] >= 'A' && str[i] <= 'Z') ||
(str[i] >= 'a' && str[i] <= 'z') ||
strchr(" #$%+-./:=?@_,", str[i])) {
i++;
continue;
}
/* valid utf8 is accepted */
len = utf8_encoded_valid_unichar(&str[i]);
if (len > 1) {
i += len;
continue;
}
/* everything else is garbage */
str[i] = '_';
i++;
replaced++;
}
return replaced;
}
int encode_devnode_name(const char *str, char *str_enc, size_t len) {
size_t i, j;
if (str == NULL || str_enc == NULL)
return -EINVAL;
for (i = 0, j = 0; str[i] != '\0'; i++) {
int seqlen;
seqlen = utf8_encoded_valid_unichar(&str[i]);
if (seqlen > 1) {
if (len-j < (size_t)seqlen)
goto err;
memcpy(&str_enc[j], &str[i], seqlen);
j += seqlen;
i += (seqlen-1);
} else if (str[i] == '\\' || !whitelisted_char_for_devnode(str[i], NULL)) {
if (len-j < 4)
goto err;
sprintf(&str_enc[j], "\\x%02x", (unsigned char) str[i]);
j += 4;
} else {
if (len-j < 1)
goto err;
str_enc[j] = str[i];
j++;
}
}
if (len-j < 1)
goto err;
str_enc[j] = '\0';
return 0;
err:
return -EINVAL;
}
/* allow chars in whitelist, plain ascii, hex-escaping and valid utf8 */
int replace_chars(char *str, const char *white)
{
size_t i = 0;
int replaced = 0;
while (str[i] != '\0') {
int len;
/* accept whitelist */
if (white != NULL && strchr(white, str[i]) != NULL) {
i++;
continue;
}
/* accept plain ascii char */
if ((str[i] >= '0' && str[i] <= '9') ||
(str[i] >= 'A' && str[i] <= 'Z') ||
(str[i] >= 'a' && str[i] <= 'z')) {
i++;
continue;
}
/* accept hex encoding */
if (str[i] == '\\' && str[i+1] == 'x') {
i += 2;
continue;
}
/* accept valid utf8 */
len = utf8_encoded_valid_unichar(&str[i]);
if (len > 1) {
i += len;
continue;
}
/* if space is allowed, replace whitespace with ordinary space */
if (isspace(str[i]) && strchr(white, ' ') != NULL) {
str[i] = ' ';
i++;
replaced++;
continue;
}
/* everything else is replaced with '_' */
str[i] = '_';
i++;
replaced++;
}
return replaced;
}
const char *utf8_is_valid(const char *str) {
const uint8_t *p;
assert(str);
for (p = (const uint8_t*) str; *p; ) {
int len;
len = utf8_encoded_valid_unichar((const char *)p);
if (len < 0)
return NULL;
p += len;
}
return str;
}
/* allow chars in whitelist, plain ascii, hex-escaping and valid utf8 */
size_t util_replace_chars(char *str, const char *white) {
size_t i = 0, replaced = 0;
assert(str);
while (str[i] != '\0') {
int len;
if (whitelisted_char_for_devnode(str[i], white)) {
i++;
continue;
}
/* accept hex encoding */
if (str[i] == '\\' && str[i+1] == 'x') {
i += 2;
continue;
}
/* accept valid utf8 */
len = utf8_encoded_valid_unichar(str + i, (size_t) -1);
if (len > 1) {
i += len;
continue;
}
/* if space is allowed, replace whitespace with ordinary space */
if (isspace(str[i]) && white && strchr(white, ' ')) {
str[i] = ' ';
i++;
replaced++;
continue;
}
/* everything else is replaced with '_' */
str[i] = '_';
i++;
replaced++;
}
return replaced;
}
static int json_parse_string(const char **p, char **ret) {
_cleanup_free_ char *s = NULL;
size_t n = 0, allocated = 0;
const char *c;
assert(p);
assert(*p);
assert(ret);
c = *p;
if (*c != '"')
return -EINVAL;
c++;
for (;;) {
int len;
/* Check for EOF */
if (*c == 0)
return -EINVAL;
/* Check for control characters 0x00..0x1f */
if (*c > 0 && *c < ' ')
return -EINVAL;
/* Check for control character 0x7f */
if (*c == 0x7f)
return -EINVAL;
if (*c == '"') {
if (!s) {
s = strdup("");
if (!s)
return -ENOMEM;
} else
s[n] = 0;
*p = c + 1;
*ret = s;
s = NULL;
return JSON_STRING;
}
if (*c == '\\') {
char ch = 0;
c++;
if (*c == 0)
return -EINVAL;
if (IN_SET(*c, '"', '\\', '/'))
ch = *c;
else if (*c == 'b')
ch = '\b';
else if (*c == 'f')
ch = '\f';
else if (*c == 'n')
ch = '\n';
else if (*c == 'r')
ch = '\r';
else if (*c == 't')
ch = '\t';
else if (*c == 'u') {
uint16_t x;
int r;
r = unhex_ucs2(c + 1, &x);
if (r < 0)
return r;
c += 5;
if (!GREEDY_REALLOC(s, allocated, n + 4))
return -ENOMEM;
if (!utf16_is_surrogate(x))
n += utf8_encode_unichar(s + n, x);
else if (utf16_is_trailing_surrogate(x))
return -EINVAL;
else {
uint16_t y;
if (c[0] != '\\' || c[1] != 'u')
return -EINVAL;
r = unhex_ucs2(c + 2, &y);
if (r < 0)
return r;
c += 6;
if (!utf16_is_trailing_surrogate(y))
return -EINVAL;
n += utf8_encode_unichar(s + n, utf16_surrogate_pair_to_unichar(x, y));
}
continue;
} else
return -EINVAL;
if (!GREEDY_REALLOC(s, allocated, n + 2))
return -ENOMEM;
s[n++] = ch;
c ++;
continue;
}
len = utf8_encoded_valid_unichar(c);
if (len < 0)
return len;
if (!GREEDY_REALLOC(s, allocated, n + len + 1))
return -ENOMEM;
memcpy(s + n, c, len);
n += len;
c += len;
}
}