static char *
scan_to_unichar (const char *str,
bson_unichar_t match,
const char *terminators,
const char **end)
{
bson_unichar_t c;
const char *iter;
for (iter = str;
iter && *iter && (c = bson_utf8_get_char(iter));
iter = bson_utf8_next_char(iter)) {
if (c == match) {
*end = iter;
return bson_strndup(str, iter - str);
} else if (c == '\\') {
iter = bson_utf8_next_char(iter);
if (!bson_utf8_get_char(iter)) {
break;
}
} else {
const char *term_iter;
for (term_iter = terminators; *term_iter; term_iter++) {
if (c == *term_iter) {
return NULL;
}
}
}
}
return NULL;
}
static char *
scan_to_unichar (const char *str,
bson_unichar_t stop,
const char **end)
{
bson_unichar_t c;
const char *iter;
for (iter = str;
iter && *iter && (c = bson_utf8_get_char(iter));
iter = bson_utf8_next_char(iter))
{
if (c == stop) {
*end = iter;
return bson_strndup(str, iter - str);
} else if (c == '\\') {
iter = bson_utf8_next_char(iter);
if (!bson_utf8_get_char(iter)) {
break;
}
}
}
return NULL;
}
static void
test_bson_utf8_get_char (void)
{
static const char *test1 = "asdf";
static const char test2[] = {0xe2, 0x82, 0xac, ' ', 0xe2, 0x82, 0xac, ' ', 0xe2, 0x82, 0xac, 0};
const char *c;
c = test1;
assert(bson_utf8_get_char(c) == 'a');
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == 's');
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == 'd');
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == 'f');
c = bson_utf8_next_char(c);
assert(!*c);
c = test2;
assert(bson_utf8_get_char(c) == 0x20AC);
c = bson_utf8_next_char(c);
assert(c == test2 + 3);
assert(bson_utf8_get_char(c) == ' ');
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == 0x20AC);
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == ' ');
c = bson_utf8_next_char(c);
assert(bson_utf8_get_char(c) == 0x20AC);
c = bson_utf8_next_char(c);
assert(!*c);
}
char *
mongoc_uri_unescape (const char *escaped_string)
{
bson_unichar_t c;
bson_string_t *str;
unsigned int hex = 0;
const char *ptr;
const char *end;
size_t len;
BSON_ASSERT (escaped_string);
len = strlen(escaped_string);
/*
* Double check that this is a UTF-8 valid string. Bail out if necessary.
*/
if (!bson_utf8_validate(escaped_string, len, false)) {
MONGOC_WARNING("%s(): escaped_string contains invalid UTF-8",
BSON_FUNC);
return NULL;
}
ptr = escaped_string;
end = ptr + len;
str = bson_string_new(NULL);
for (; *ptr; ptr = bson_utf8_next_char(ptr)) {
c = bson_utf8_get_char(ptr);
switch (c) {
case '%':
if (((end - ptr) < 2) ||
!isxdigit(ptr[1]) ||
!isxdigit(ptr[2]) ||
#ifdef _MSC_VER
(1 != sscanf_s(&ptr[1], "%02x", &hex)) ||
#else
(1 != sscanf(&ptr[1], "%02x", &hex)) ||
#endif
!isprint(hex)) {
bson_string_free(str, true);
return NULL;
}
bson_string_append_c(str, hex);
ptr += 2;
break;
default:
bson_string_append_unichar(str, c);
break;
}
}
return bson_string_free(str, false);
}
static bool
_bson_json_all_whitespace (const char *utf8)
{
bool all_whitespace = true;
for (; *utf8; utf8 = bson_utf8_next_char (utf8)) {
if (!isspace (bson_utf8_get_char (utf8))) {
all_whitespace = false;
break;
}
}
return all_whitespace;
}
void
mongoc_uri_lowercase_hostname (const char *src,
char *buf /* OUT */,
int len)
{
bson_unichar_t c;
const char *iter;
char *buf_iter;
/* TODO: this code only accepts ascii, and assumes that lowercased
chars are the same width as originals */
for (iter = src, buf_iter = buf;
iter && *iter && (c = bson_utf8_get_char(iter)) && buf_iter - buf < len;
iter = bson_utf8_next_char(iter), buf_iter++) {
assert(c < 128);
*buf_iter = tolower(c);
}
}
int monary_load_length_value(const bson_iter_t* bsonit,
monary_column_item* citem,
int idx)
{
bson_type_t type;
bson_iter_t child;
const char* discard;
uint32_t length;
uint32_t* dest;
type = bson_iter_type(bsonit);
switch (type) {
case BSON_TYPE_UTF8:
case BSON_TYPE_CODE:
discard = bson_iter_utf8(bsonit, &length);
for (length = 0; *discard; length++) {
discard = bson_utf8_next_char(discard);
}
break;
case BSON_TYPE_ARRAY:
case BSON_TYPE_DOCUMENT:
if (!bson_iter_recurse(bsonit, &child)) {
return 0;
}
for (length = 0; bson_iter_next(&child); length++);
break;
case BSON_TYPE_BINARY:
bson_iter_binary(bsonit, NULL, &length, (const uint8_t**) &discard);
break;
default:
return 0;
}
dest = ((uint32_t*) citem->storage) + idx;
memcpy(dest, &length, sizeof(uint32_t));
return 1;
}