size_t UnicodeDecoder::decode(Char *to, const char *from, size_t fromLen,
const char **rest)
{
union U {
unsigned short word;
char bytes[2];
};
if (subDecoder_)
return subDecoder_->decode(to, from, fromLen, rest);
if (fromLen < 2) {
*rest = from;
return 0;
}
minBytesPerChar_ = 2;
U u;
u.bytes[0] = from[0];
u.bytes[1] = from[1];
if (u.word == byteOrderMark) {
hadByteOrderMark_ = 1;
from += 2;
fromLen -= 2;
}
else if (u.word == swappedByteOrderMark) {
hadByteOrderMark_ = 1;
from += 2;
fromLen -= 2;
swapBytes_ = 1;
}
if (hadByteOrderMark_ || !subCodingSystem_)
subCodingSystem_ = new UTF16CodingSystem;
subDecoder_ = subCodingSystem_->makeDecoder(swapBytes_);
minBytesPerChar_ = subDecoder_->minBytesPerChar();
return subDecoder_->decode(to, from, fromLen, rest);
}
size_t UnicodeDecoder::decode(Char *to, const char *from, size_t fromLen,
const char **rest)
{
union U {
unsigned short word;
char bytes[2];
};
if (subDecoder_)
return subDecoder_->decode(to, from, fromLen, rest);
if (!hadFirstChar_) {
hadFirstChar_ = 1;
minBytesPerChar_ = 2;
if (fromLen < 2) {
*rest = from;
return 0;
}
U u;
u.bytes[0] = from[0];
u.bytes[1] = from[1];
if (u.word == byteOrderMark) {
hadByteOrderMark_ = 1;
from += 2;
fromLen -= 2;
}
else if (u.word == swappedByteOrderMark) {
hadByteOrderMark_ = 1;
from += 2;
fromLen -= 2;
swapBytes_ = 1;
}
else if (subCodingSystem_) {
subDecoder_ = subCodingSystem_->makeDecoder();
minBytesPerChar_ = subDecoder_->minBytesPerChar();
return subDecoder_->decode(to, from, fromLen, rest);
}
}
fromLen &= ~1;
*rest = from + fromLen;
if (sizeof(Char) == 2) {
if (!swapBytes_) {
if (from != (char *)to)
memmove(to, from, fromLen);
return fromLen/2;
}
}
if (swapBytes_) {
for (size_t n = fromLen; n > 0; n -= 2) {
U u;
u.bytes[1] = *from++;
u.bytes[0] = *from++;
*to++ = u.word;
}
}
else {
for (size_t n = fromLen; n > 0; n -= 2) {
U u;
u.bytes[0] = *from++;
u.bytes[1] = *from++;
*to++ = u.word;
}
}
return fromLen/2;
}