in src/org/apache/xerces/impl/io/UTF8Reader.java [287:576]
public int read(char ch[], int offset, int length) throws IOException {
// read bytes
int out = offset;
int count = 0;
if (fOffset == 0) {
// adjust length to read
if (length > fBuffer.length) {
length = fBuffer.length;
}
// handle surrogate
if (fSurrogate != -1) {
ch[out++] = (char)fSurrogate;
fSurrogate = -1;
length--;
}
// perform read operation
count = fInputStream.read(fBuffer, 0, length);
if (count == -1) {
return -1;
}
count += out - offset;
}
// skip read; last character was in error
// NOTE: Having an offset value other than zero means that there was
// an error in the last character read. In this case, we have
// skipped the read so we don't consume any bytes past the
// error. By signalling the error on the next block read we
// allow the method to return the most valid characters that
// it can on the previous block read. -Ac
else {
count = fOffset;
fOffset = 0;
}
// convert bytes to characters
final int total = count;
int in;
byte byte1;
final byte byte0 = 0;
for (in = 0; in < total; in++) {
byte1 = fBuffer[in];
if (byte1 >= byte0) {
ch[out++] = (char)byte1;
}
else {
break;
}
}
for ( ; in < total; in++) {
byte1 = fBuffer[in];
// UTF-8: [0xxx xxxx]
// Unicode: [0000 0000] [0xxx xxxx]
if (byte1 >= byte0) {
ch[out++] = (char)byte1;
continue;
}
// UTF-8: [110y yyyy] [10xx xxxx]
// Unicode: [0000 0yyy] [yyxx xxxx]
int b0 = byte1 & 0x0FF;
if ((b0 & 0xE0) == 0xC0 && (b0 & 0x1E) != 0) {
int b1 = -1;
if (++in < total) {
b1 = fBuffer[in] & 0x00FF;
}
else {
b1 = fInputStream.read();
if (b1 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fOffset = 1;
return out - offset;
}
expectedByte(2, 2);
}
count++;
}
if ((b1 & 0xC0) != 0x80) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fOffset = 2;
return out - offset;
}
invalidByte(2, 2, b1);
}
int c = ((b0 << 6) & 0x07C0) | (b1 & 0x003F);
ch[out++] = (char)c;
count -= 1;
continue;
}
// UTF-8: [1110 zzzz] [10yy yyyy] [10xx xxxx]
// Unicode: [zzzz yyyy] [yyxx xxxx]
if ((b0 & 0xF0) == 0xE0) {
int b1 = -1;
if (++in < total) {
b1 = fBuffer[in] & 0x00FF;
}
else {
b1 = fInputStream.read();
if (b1 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fOffset = 1;
return out - offset;
}
expectedByte(2, 3);
}
count++;
}
if ((b1 & 0xC0) != 0x80
|| (b0 == 0xED && b1 >= 0xA0)
|| ((b0 & 0x0F) == 0 && (b1 & 0x20) == 0)) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fOffset = 2;
return out - offset;
}
invalidByte(2, 3, b1);
}
int b2 = -1;
if (++in < total) {
b2 = fBuffer[in] & 0x00FF;
}
else {
b2 = fInputStream.read();
if (b2 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fOffset = 2;
return out - offset;
}
expectedByte(3, 3);
}
count++;
}
if ((b2 & 0xC0) != 0x80) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fBuffer[2] = (byte)b2;
fOffset = 3;
return out - offset;
}
invalidByte(3, 3, b2);
}
int c = ((b0 << 12) & 0xF000) | ((b1 << 6) & 0x0FC0) |
(b2 & 0x003F);
ch[out++] = (char)c;
count -= 2;
continue;
}
// UTF-8: [1111 0uuu] [10uu zzzz] [10yy yyyy] [10xx xxxx]*
// Unicode: [1101 10ww] [wwzz zzyy] (high surrogate)
// [1101 11yy] [yyxx xxxx] (low surrogate)
// * uuuuu = wwww + 1
if ((b0 & 0xF8) == 0xF0) {
int b1 = -1;
if (++in < total) {
b1 = fBuffer[in] & 0x00FF;
}
else {
b1 = fInputStream.read();
if (b1 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fOffset = 1;
return out - offset;
}
expectedByte(2, 4);
}
count++;
}
if ((b1 & 0xC0) != 0x80
|| ((b1 & 0x30) == 0 && (b0 & 0x07) == 0)) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fOffset = 2;
return out - offset;
}
invalidByte(2, 4, b1);
}
int b2 = -1;
if (++in < total) {
b2 = fBuffer[in] & 0x00FF;
}
else {
b2 = fInputStream.read();
if (b2 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fOffset = 2;
return out - offset;
}
expectedByte(3, 4);
}
count++;
}
if ((b2 & 0xC0) != 0x80) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fBuffer[2] = (byte)b2;
fOffset = 3;
return out - offset;
}
invalidByte(3, 4, b2);
}
int b3 = -1;
if (++in < total) {
b3 = fBuffer[in] & 0x00FF;
}
else {
b3 = fInputStream.read();
if (b3 == -1) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fBuffer[2] = (byte)b2;
fOffset = 3;
return out - offset;
}
expectedByte(4, 4);
}
count++;
}
if ((b3 & 0xC0) != 0x80) {
if (out > offset) {
fBuffer[0] = (byte)b0;
fBuffer[1] = (byte)b1;
fBuffer[2] = (byte)b2;
fBuffer[3] = (byte)b3;
fOffset = 4;
return out - offset;
}
invalidByte(4, 4, b2);
}
// decode bytes into surrogate characters
int uuuuu = ((b0 << 2) & 0x001C) | ((b1 >> 4) & 0x0003);
if (uuuuu > 0x10) {
invalidSurrogate(uuuuu);
}
int wwww = uuuuu - 1;
int zzzz = b1 & 0x000F;
int yyyyyy = b2 & 0x003F;
int xxxxxx = b3 & 0x003F;
int hs = 0xD800 | ((wwww << 6) & 0x03C0) | (zzzz << 2) | (yyyyyy >> 4);
int ls = 0xDC00 | ((yyyyyy << 6) & 0x03C0) | xxxxxx;
// set characters
ch[out++] = (char)hs;
if ((count -= 2) <= length) {
ch[out++] = (char)ls;
}
// reached the end of the char buffer; save low surrogate for the next read
else {
fSurrogate = ls;
--count;
}
continue;
}
// error
if (out > offset) {
fBuffer[0] = (byte)b0;
fOffset = 1;
return out - offset;
}
invalidByte(1, 1, b0);
}
// return number of characters converted
if (DEBUG_READ) {
System.out.println("read(char[],"+offset+','+length+"): count="+count);
}
return count;
} // read(char[],int,int)