void VCodePoint::writeToBinaryStream(VBinaryIOStream& stream) const {
switch (VCodePoint::getUTF8LengthFromCodePointValue(mIntValue)) {
case 1:
stream.writeU8((Vu8) mIntValue); // first byte binary: 0xxxxxxx (with 7 used bits)
break;
case 2:
stream.writeU8((Vu8) UTF8_BYTE_1_OF_2(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_2_OF_2(mIntValue));
break;
case 3:
stream.writeU8((Vu8) UTF8_BYTE_1_OF_3(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_2_OF_3(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_3_OF_3(mIntValue));
break;
case 4:
stream.writeU8((Vu8) UTF8_BYTE_1_OF_4(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_2_OF_4(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_3_OF_4(mIntValue));
stream.writeU8((Vu8) UTF8_BYTE_4_OF_4(mIntValue));
break;
default:
throw VRangeException(VSTRING_FORMAT("VCodePoint::writeToBinaryStream() for an invalid UTF-8 code point 0x%X", mIntValue));
break;
}
}
VString VCodePoint::toString() const {
VString s;
// Use of 0x40 (decimal 64) here is to chop a number into 6-bit parts.
// 0x40 is binary 01000000, so
// n / 0x40 effectively strips off the low 6 bits
// n % 0x40 effectively strips off all but the low 6 bits
// n / 0x40 % 0x40 effectively yields the "next" 6 bits by combining those two operations
switch (VCodePoint::getUTF8LengthFromCodePointValue(mIntValue)) {
case 1:
s += (char) mIntValue; // first byte binary: 0xxxxxxx (with 7 used bits)
break;
case 2:
s += (char) UTF8_BYTE_1_OF_2(mIntValue);
s += (char) UTF8_BYTE_2_OF_2(mIntValue);
break;
case 3:
s += (char) UTF8_BYTE_1_OF_3(mIntValue);
s += (char) UTF8_BYTE_2_OF_3(mIntValue);
s += (char) UTF8_BYTE_3_OF_3(mIntValue);
break;
case 4:
s += (char) UTF8_BYTE_1_OF_4(mIntValue);
s += (char) UTF8_BYTE_2_OF_4(mIntValue);
s += (char) UTF8_BYTE_3_OF_4(mIntValue);
s += (char) UTF8_BYTE_4_OF_4(mIntValue);
break;
default:
throw VRangeException(VSTRING_FORMAT("VCodePoint::toString() for an invalid UTF-8 code point 0x%X", mIntValue));
break;
}
return s;
}
static bstring read_string(char * src, size_t length)
{
int utf8_char;
char hex_num[7] = "0x0000";
char * end = src + length;
bstring s = bfromcstralloc(length, "");
char * dst = bdata(s);
if (* src == '"') src++;
while (src < end) {
if (* src != '\\') {
* dst = * src;
dst++;
} else {
src++;
if (* src == '"') {
* dst = '"';
dst++;
} else if (* src == '\\') {
* dst = '\\';
dst++;
} else if (* src == '/') {
* dst = '/';
dst++;
} else if (* src == 'b') {
* dst = '\b';
dst++;
} else if (* src == 'f') {
* dst = '\f';
dst++;
} else if (* src == 'n') {
* dst = '\n';
dst++;
} else if (* src == 'r') {
* dst = '\r';
dst++;
} else if (* src == 't') {
* dst = '\t';
dst++;
} else if (* src == 'u') {
memcpy(hex_num + 2, src + 1, 4);
sscanf(hex_num, "%x", &utf8_char);
if (utf8_char <= UTF8_2_BYTE_LIMIT) {
* dst = UTF8_BYTE_1_OF_2(utf8_char);
dst++;
* dst = UTF8_BYTE_2_OF_2(utf8_char);
dst++;
} else {
* dst = UTF8_BYTE_1_OF_3(utf8_char);
dst++;
* dst = UTF8_BYTE_2_OF_3(utf8_char);
dst++;
* dst = UTF8_BYTE_3_OF_3(utf8_char);
dst++;
}
src += 4;
}
}
src++;
}
s->slen = dst - bdata(s);
return s;
}