/**
* Like "strlen", but for strings encoded with Java's modified UTF-8.
*
* The value returned is the number of UTF-16 characters required
* to represent this string.
*/
extern size_t strlen8to16 (const char* utf8Str)
{
size_t len = 0;
int ic;
int expected = 0;
while ((ic = *utf8Str++) != '\0') {
/* bytes that start 0? or 11 are lead bytes and count as characters.*/
/* bytes that start 10 are extention bytes and are not counted */
if ((ic & 0xc0) == 0x80) {
/* count the 0x80 extention bytes. if we have more than
* expected, then start counting them because strcpy8to16
* will insert UTF16_REPLACEMENT_CHAR's
*/
expected--;
if (expected < 0) {
len++;
}
} else {
len++;
expected = UTF8_SEQ_LENGTH(ic) - 1;
/* this will result in a surrogate pair */
if (expected == 3) {
len++;
}
}
}
return len;
}
int UString::_utf_string_len(const char* buf, size_t size)
{
int len = 0;
int i = 0;
if (buf == NULL){
return 0;
}
while(i < (int)size){
int sl = UTF8_SEQ_LENGTH(buf);
if (sl){
len++;
buf += sl;
}
else{ // Something wrong here, failure recover.
// utf8_data++;
break;
}
i += sl;
}
if(i > (int)size){ // Overflow, invalid data stream.
return -1;
}
return len;
}
/*
* Retrieve the next UTF-32 character from a UTF-8 string.
*
* Stops at inner \0's
*
* Returns UTF16_REPLACEMENT_CHAR if an invalid sequence is encountered
*
* Advances "*pUtf8Ptr" to the start of the next character.
*/
static inline uint32_t getUtf32FromUtf8(const char** pUtf8Ptr)
{
uint32_t ret;
int seq_len;
int i;
/* Mask for leader byte for lengths 1, 2, 3, and 4 respectively*/
static const char leaderMask[4] = {0xff, 0x1f, 0x0f, 0x07};
/* Bytes that start with bits "10" are not leading characters. */
if (((**pUtf8Ptr) & 0xc0) == 0x80) {
(*pUtf8Ptr)++;
return UTF16_REPLACEMENT_CHAR;
}
/* note we tolerate invalid leader 11111xxx here */
seq_len = UTF8_SEQ_LENGTH(**pUtf8Ptr);
ret = (**pUtf8Ptr) & leaderMask [seq_len - 1];
if (**pUtf8Ptr == '\0') return ret;
(*pUtf8Ptr)++;
for (i = 1; i < seq_len ; i++, (*pUtf8Ptr)++) {
if ((**pUtf8Ptr) == '\0') return UTF16_REPLACEMENT_CHAR;
if (((**pUtf8Ptr) & 0xc0) != 0x80) return UTF16_REPLACEMENT_CHAR;
UTF8_SHIFT_AND_MASK(ret, **pUtf8Ptr);
}
return ret;
}
UString UString::substr(size_t startIndex, size_t endIndex) const
{
endIndex = endIndex > m_strLen ? m_strLen : endIndex;
UString sRet;
const char* str = (const char *)m_mallocBuffer;
if(endIndex - startIndex != 0){
size_t n = 0;
int start = -1, end = -1, index = 0;
while(n < m_mallocBufferLen){
int len = UTF8_SEQ_LENGTH(str);
if(index == startIndex){
start = n;
}
if(index == endIndex - 1){
end = n + len;
break;
}
n += len;
str += len;
index++;
}
if(start != -1 && end != -1){
sRet.init( (const char*)(m_mallocBuffer+start), end-start);
}
}
return sRet;
}