ConversionResult convertUTF8ToUTF16(
const char** sourceStart, const char* sourceEnd,
UChar** targetStart, UChar* targetEnd, bool strict)
{
ConversionResult result = conversionOK;
const char* source = *sourceStart;
UChar* target = *targetStart;
while (source < sourceEnd) {
int utf8SequenceLength = inlineUTF8SequenceLength(*source);
if (sourceEnd - source < utf8SequenceLength) {
result = sourceExhausted;
break;
}
// Do this check whether lenient or strict
if (!isLegalUTF8(reinterpret_cast<const unsigned char*>(source), utf8SequenceLength)) {
result = sourceIllegal;
break;
}
UChar32 character = readUTF8Sequence(source, utf8SequenceLength);
if (target >= targetEnd) {
source -= utf8SequenceLength; // Back up source pointer!
result = targetExhausted;
break;
}
if (U_IS_BMP(character)) {
// UTF-16 surrogate values are illegal in UTF-32
if (U_IS_SURROGATE(character)) {
if (strict) {
source -= utf8SequenceLength; // return to the illegal value itself
result = sourceIllegal;
break;
} else
*target++ = replacementCharacter;
} else
*target++ = character; // normal case
} else if (U_IS_SUPPLEMENTARY(character)) {
// target is a character in range 0xFFFF - 0x10FFFF
if (target + 1 >= targetEnd) {
source -= utf8SequenceLength; // Back up source pointer!
result = targetExhausted;
break;
}
*target++ = U16_LEAD(character);
*target++ = U16_TRAIL(character);
} else {
if (strict) {
source -= utf8SequenceLength; // return to the start
result = sourceIllegal;
break; // Bail out; shouldn't continue
} else
*target++ = replacementCharacter;
}
}
*sourceStart = source;
*targetStart = target;
return result;
}
// Takes a null-terminated C-style string with a UTF-8 sequence in it and converts it to a character.
// Only allows Unicode characters (U-00000000 to U-0010FFFF).
// Returns -1 if the sequence is not valid (including presence of extra bytes).
int decodeUTF8Sequence(const char *sequence)
{
// Handle 0-byte sequences (never valid).
const unsigned char b0 = sequence[0];
const int length = inlineUTF8SequenceLength(b0);
if (length == 0)
return -1;
// Handle 1-byte sequences (plain ASCII).
const unsigned char b1 = sequence[1];
if (length == 1) {
if (b1)
return -1;
return b0;
}
// Handle 2-byte sequences.
if ((b1 & 0xC0) != 0x80)
return -1;
const unsigned char b2 = sequence[2];
if (length == 2) {
if (b2)
return -1;
const int c = ((b0 & 0x1F) << 6) | (b1 & 0x3F);
if (c < 0x80)
return -1;
return c;
}
// Handle 3-byte sequences.
if ((b2 & 0xC0) != 0x80)
return -1;
const unsigned char b3 = sequence[3];
if (length == 3) {
if (b3)
return -1;
const int c = ((b0 & 0xF) << 12) | ((b1 & 0x3F) << 6) | (b2 & 0x3F);
if (c < 0x800)
return -1;
// UTF-16 surrogates should never appear in UTF-8 data.
if (c >= 0xD800 && c <= 0xDFFF)
return -1;
// Backwards BOM and U+FFFF should never appear in UTF-8 data.
if (c == 0xFFFE || c == 0xFFFF)
return -1;
return c;
}
// Handle 4-byte sequences.
if ((b3 & 0xC0) != 0x80)
return -1;
const unsigned char b4 = sequence[4];
if (length == 4) {
if (b4)
return -1;
const int c = ((b0 & 0x7) << 18) | ((b1 & 0x3F) << 12) | ((b2 & 0x3F) << 6) | (b3 & 0x3F);
if (c < 0x10000 || c > 0x10FFFF)
return -1;
return c;
}
return -1;
}