public static void main(String args[]) throws Exception {
String inputFile = "samplein.txt";
String outputFile = "sampleout.txt";
RandomAccessFile inf = new RandomAccessFile(inputFile, "r");
RandomAccessFile outf = new RandomAccessFile(outputFile, "rw");
long inputLength = new File(inputFile).length();
FileChannel inc = inf.getChannel();
FileChannel outc = outf.getChannel();
MappedByteBuffer inputData = inc.map(FileChannel.MapMode.READ_ONLY, 0, inputLength);
Charset latin1 = Charset.forName("ISO-8859-1");
CharsetDecoder decoder = latin1.newDecoder();
CharsetEncoder encoder = latin1.newEncoder();
CharBuffer cb = decoder.decode(inputData);
// Process char data here
ByteBuffer outputData = encoder.encode(cb);
outc.write(outputData);
inf.close();
outf.close();
}
static boolean check(CharsetDecoder dec, byte[] bytes, boolean direct, int[] flow) {
int inPos = flow[0];
int inLen = flow[1];
int outPos = flow[2];
int outLen = flow[3];
int expedInPos = flow[4];
int expedOutPos = flow[5];
CoderResult expedCR = (flow[6] == 0) ? CoderResult.UNDERFLOW : CoderResult.OVERFLOW;
ByteBuffer bbf;
CharBuffer cbf;
if (direct) {
bbf = ByteBuffer.allocateDirect(inPos + bytes.length);
cbf = ByteBuffer.allocateDirect((outPos + outLen) * 2).asCharBuffer();
} else {
bbf = ByteBuffer.allocate(inPos + bytes.length);
cbf = CharBuffer.allocate(outPos + outLen);
}
bbf.position(inPos);
bbf.put(bytes).flip().position(inPos).limit(inPos + inLen);
cbf.position(outPos);
dec.reset();
CoderResult cr = dec.decode(bbf, cbf, false);
if (cr != expedCR || bbf.position() != expedInPos || cbf.position() != expedOutPos) {
System.out.printf("Expected(direct=%5b): [", direct);
for (int i : flow) System.out.print(" " + i);
System.out.println(
"] CR=" + cr + ", inPos=" + bbf.position() + ", outPos=" + cbf.position());
return false;
}
return true;
}
/**
* Read a 'n' bytes from buffer into a String where n is the framesize - offset so therefore
* cannot use this if there are other objects after it because it has no delimiter.
*
* <p>Must take into account the text encoding defined in the Encoding Object ID3 Text Frames
* often allow multiple strings seperated by the null char appropriate for the encoding.
*
* @param arr this is the buffer for the frame
* @param offset this is where to start reading in the buffer for this field
* @throws NullPointerException
* @throws IndexOutOfBoundsException
*/
public void readByteArray(byte[] arr, int offset) throws InvalidDataTypeException {
logger.finest("Reading from array from offset:" + offset);
// Get the Specified Decoder
String charSetName = getTextEncodingCharSet();
CharsetDecoder decoder = Charset.forName(charSetName).newDecoder();
decoder.reset();
// Decode sliced inBuffer
ByteBuffer inBuffer;
// #302 [dallen] truncating array manually since the decoder.decode() does not honor the offset
// in the in buffer
byte[] truncArr = new byte[arr.length - offset];
System.arraycopy(arr, offset, truncArr, 0, truncArr.length);
inBuffer = ByteBuffer.wrap(truncArr);
CharBuffer outBuffer = CharBuffer.allocate(arr.length - offset);
CoderResult coderResult = decoder.decode(inBuffer, outBuffer, true);
if (coderResult.isError()) {
logger.warning("Decoding error:" + coderResult.toString());
}
decoder.flush(outBuffer);
outBuffer.flip();
// If using UTF16 with BOM we then search through the text removing any BOMs that could exist
// for multiple values, BOM could be Big Endian or Little Endian
if (charSetName.equals(TextEncoding.CHARSET_UTF_16)) {
value = outBuffer.toString().replace("\ufeff", "").replace("\ufffe", "");
} else {
value = outBuffer.toString();
}
// SetSize, important this is correct for finding the next datatype
setSize(arr.length - offset);
logger.config("Read SizeTerminatedString:" + value + " size:" + size);
}
public static void main(String[] arguments) {
try {
// read byte data into a byte buffer
String data = "friends.dat";
FileInputStream inData = new FileInputStream(data);
FileChannel inChannel = inData.getChannel();
long inSize = inChannel.size();
ByteBuffer source = ByteBuffer.allocate((int) inSize);
inChannel.read(source, 0);
source.position(0);
System.out.println("Original byte data:");
for (int i = 0; source.remaining() > 0; i++) {
System.out.print(source.get() + " ");
}
// convert byte data into character data
source.position(0);
Charset ascii = Charset.forName("US-ASCII");
CharsetDecoder toAscii = ascii.newDecoder();
CharBuffer destination = toAscii.decode(source);
destination.position(0);
System.out.println("\n\nNew character data:");
for (int i = 0; destination.remaining() > 0; i++) {
System.out.print(destination.get());
}
System.out.println();
} catch (FileNotFoundException fne) {
System.out.println(fne.getMessage());
} catch (IOException ioe) {
System.out.println(ioe.getMessage());
}
}
static void testMixed(Charset cs) throws Throwable {
CharsetDecoder dec =
cs.newDecoder()
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
CharsetEncoder enc =
cs.newEncoder()
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
List<Integer> cps = new ArrayList<>(0x10000);
int off = 0;
int cp = 0;
while (cp < 0x10000) {
if (enc.canEncode((char) cp)) {
cps.add(cp);
}
cp++;
}
Collections.shuffle(cps);
char[] bmpCA = new char[cps.size()];
for (int i = 0; i < cps.size(); i++) bmpCA[i] = (char) (int) cps.get(i);
String bmpStr = new String(bmpCA);
// getBytes(csn);
byte[] bmpBA = bmpStr.getBytes(cs.name());
ByteBuffer bf = enc.reset().encode(CharBuffer.wrap(bmpCA));
byte[] baNIO = new byte[bf.limit()];
bf.get(baNIO, 0, baNIO.length);
if (!Arrays.equals(bmpBA, baNIO)) {
throw new RuntimeException("getBytes(csn) failed -> " + cs.name());
}
// getBytes(cs);
bmpBA = bmpStr.getBytes(cs);
if (!Arrays.equals(bmpBA, baNIO))
throw new RuntimeException("getBytes(cs) failed -> " + cs.name());
// new String(csn);
String strSC = new String(bmpBA, cs.name());
String strNIO = dec.reset().decode(ByteBuffer.wrap(bmpBA)).toString();
if (!strNIO.equals(strSC)) {
throw new RuntimeException("new String(csn) failed -> " + cs.name());
}
// new String(cs);
strSC = new String(bmpBA, cs);
if (!strNIO.equals(strSC)) throw new RuntimeException("new String(cs) failed -> " + cs.name());
}
/**
* Sets the substitution characters to use when the converter is in substitution mode. The given
* chars must not be longer than the value returned by getMaxCharsPerByte for this converter.
*
* @param chars the substitution chars
* @exception IllegalArgumentException if given byte array is longer than the value returned by
* the method getMaxBytesPerChar.
* @see #setSubstitutionMode
* @see #getMaxBytesPerChar
*/
public void setSubstitutionChars(char[] chars) throws IllegalArgumentException {
if (decoder != null) decoder.replaceWith(new String(chars));
else { // only provided for subclasses
if (chars.length > getMaxCharsPerByte()) throw new IllegalArgumentException();
subChars = new char[chars.length];
System.arraycopy(chars, 0, subChars, 0, chars.length);
}
}
static char[] decode(byte[] bb, Charset cs, boolean testDirect, Time t) throws Exception {
String csn = cs.name();
CharsetDecoder dec = cs.newDecoder();
ByteBuffer bbf;
CharBuffer cbf;
if (testDirect) {
bbf = ByteBuffer.allocateDirect(bb.length);
cbf = ByteBuffer.allocateDirect(bb.length * 2).asCharBuffer();
bbf.put(bb);
} else {
bbf = ByteBuffer.wrap(bb);
cbf = CharBuffer.allocate(bb.length);
}
CoderResult cr = null;
long t1 = System.nanoTime() / 1000;
for (int i = 0; i < iteration; i++) {
bbf.rewind();
cbf.clear();
dec.reset();
cr = dec.decode(bbf, cbf, true);
}
long t2 = System.nanoTime() / 1000;
t.t = (t2 - t1) / iteration;
if (cr != CoderResult.UNDERFLOW) {
System.out.println("DEC-----------------");
int pos = bbf.position();
System.out.printf(
" cr=%s, bbf.pos=%d, bb[pos]=%x,%x,%x,%x%n",
cr.toString(),
pos,
bb[pos++] & 0xff,
bb[pos++] & 0xff,
bb[pos++] & 0xff,
bb[pos++] & 0xff);
throw new RuntimeException("Decoding err: " + csn);
}
char[] cc = new char[cbf.position()];
cbf.flip();
cbf.get(cc);
return cc;
}
public String readAsString(Charset charset) throws CharacterCodingException {
int unreadSize = totalBytesUnread();
if (unreadSize > 0) {
CharsetDecoder decoder =
charset
.newDecoder()
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
CharBuffer charbuffer = CharBuffer.allocate(unreadSize);
ByteBuffer buf = null;
while (prepareRead() != -1) {
buf = currentReadChunk.readToNioBuffer();
boolean endOfInput = (prepareRead() == -1);
CoderResult result = decoder.decode(buf, charbuffer, endOfInput);
if (endOfInput) {
if (!result.isUnderflow()) {
result.throwException();
}
}
}
CoderResult result = decoder.flush(charbuffer);
if (buf.hasRemaining()) {
throw new IllegalStateException("There's a bug here, buffer wasn't read fully.");
}
if (!result.isUnderflow()) result.throwException();
charbuffer.flip();
String str;
if (charbuffer.hasArray()) {
int len = charbuffer.remaining();
char[] ch = charbuffer.array();
if (len != ch.length) {
ch = ArrayUtils.subarray(ch, 0, len);
}
str = StringCharArrayAccessor.createString(ch);
} else {
str = charbuffer.toString();
}
return str;
}
return null;
}
@Override
int read(final TextInput ti) throws IOException {
int c = -1;
while (++c < 4) {
final int ch = ti.readByte();
if (ch < 0) break;
cache[c] = (byte) ch;
outc.position(0);
inc.position(0);
inc.limit(c + 1);
csd.reset();
final CoderResult cr = csd.decode(inc, outc, true);
if (cr.isMalformed()) continue;
// return character
int i = 0;
final int os = outc.position();
for (int o = 0; o < os; ++o) i |= outc.get(o) << (o << 3);
return i;
}
return c == 0 ? -1 : invalid();
}
public static void main(String[] args) throws Exception {
// 创建简体中文对应的Charset
Charset cn = Charset.forName("GBK");
// 获取cn对象对应的编码器和解码器
CharsetEncoder cnEncoder = cn.newEncoder();
CharsetDecoder cnDecoder = cn.newDecoder();
// 创建一个CharBuffer对象
CharBuffer cbuff = CharBuffer.allocate(8);
cbuff.put('孙');
cbuff.put('悟');
cbuff.put('空');
cbuff.flip();
// 将CharBuffer中的字符序列转换成字节序列
ByteBuffer bbuff = cnEncoder.encode(cbuff);
// 循环访问ByteBuffer中的每个字节
for (int i = 0; i < bbuff.capacity(); i++) {
System.out.print(bbuff.get(i) + " ");
}
// 将ByteBuffer的数据解码成字符序列
System.out.println("\n" + cnDecoder.decode(bbuff));
}
static CoderResult decodeCR(byte[] bb, Charset cs, boolean testDirect) throws Exception {
CharsetDecoder dec = cs.newDecoder();
ByteBuffer bbf;
CharBuffer cbf;
if (testDirect) {
bbf = ByteBuffer.allocateDirect(bb.length);
cbf = ByteBuffer.allocateDirect(bb.length * 2).asCharBuffer();
bbf.put(bb).flip();
} else {
bbf = ByteBuffer.wrap(bb);
cbf = CharBuffer.allocate(bb.length);
}
CoderResult cr = null;
for (int i = 0; i < iteration; i++) {
bbf.rewind();
cbf.clear();
dec.reset();
cr = dec.decode(bbf, cbf, true);
}
return cr;
}
/**
* Writes any remaining output to the output buffer and resets the converter to its initial state.
*
* @param output char array to receive flushed output.
* @param outStart start writing to output array at this offset.
* @param outEnd stop writing to output array at this offset (exclusive).
* @exception MalformedInputException if the output to be flushed contained a partial or invalid
* multibyte character sequence. flush will write what it can to the output buffer and reset
* the converter before throwing this exception. An additional call to flush is not required.
* @exception ConversionBufferFullException if output array is filled before all the output can be
* flushed. flush will write what it can to the output buffer and remember its state. An
* additional call to flush with a new output buffer will conclude the operation.
*/
public int flush(char[] output, int outStart, int outEnd)
throws MalformedInputException, ConversionBufferFullException {
byteOff = charOff = 0;
if (outStart >= outEnd || outStart >= output.length) throw new ConversionBufferFullException();
if (dst != null && dst.array() == output) dst.position(outStart).limit(outEnd);
else dst = CharBuffer.wrap(output, outStart, outEnd - outStart);
CoderResult cr = null;
try {
if (src != null) cr = decoder.decode((ByteBuffer) src.clear(), dst, true);
assert !cr.isUnmappable();
if (cr.isMalformed()) {
badInputLength = cr.length();
reset();
throw new MalformedInputException();
}
} catch (IllegalStateException ise) {
if (src != null) cr = decoder.reset().decode(src, dst, true);
}
try {
cr = decoder.flush(dst);
} catch (Exception e) {
assert false;
} finally {
byteOff = 0;
charOff = dst.position();
src = null;
}
if (cr.isOverflow()) throw new ConversionBufferFullException();
// Return the length written to the output buffer
if (cr.isUnderflow()) {
int written = charOff - outStart;
reset();
return written;
}
assert false;
return -1; // should be never reached
}
/**
* Decode file charset.
*
* @param f File to process.
* @return File charset.
* @throws IOException in case of error.
*/
public static Charset decode(File f) throws IOException {
SortedMap<String, Charset> charsets = Charset.availableCharsets();
String[] firstCharsets = {
Charset.defaultCharset().name(), "US-ASCII", "UTF-8", "UTF-16BE", "UTF-16LE"
};
Collection<Charset> orderedCharsets = U.newLinkedHashSet(charsets.size());
for (String c : firstCharsets)
if (charsets.containsKey(c)) orderedCharsets.add(charsets.get(c));
orderedCharsets.addAll(charsets.values());
try (RandomAccessFile raf = new RandomAccessFile(f, "r")) {
FileChannel ch = raf.getChannel();
ByteBuffer buf = ByteBuffer.allocate(4096);
ch.read(buf);
buf.flip();
for (Charset charset : orderedCharsets) {
CharsetDecoder decoder = charset.newDecoder();
decoder.reset();
try {
decoder.decode(buf);
return charset;
} catch (CharacterCodingException ignored) {
}
}
}
return Charset.defaultCharset();
}
/**
* Converts an array of bytes containing characters in an external encoding into an array of
* Unicode characters. This method allows a buffer by buffer conversion of a data stream. The
* state of the conversion is saved between calls to convert. Among other things, this means
* multibyte input sequences can be split between calls. If a call to convert results in an
* exception, the conversion may be continued by calling convert again with suitably modified
* parameters. All conversions should be finished with a call to the flush method.
*
* @return the number of bytes written to output.
* @param input byte array containing text to be converted.
* @param inStart begin conversion at this offset in input array.
* @param inEnd stop conversion at this offset in input array (exclusive).
* @param output character array to receive conversion result.
* @param outStart start writing to output array at this offset.
* @param outEnd stop writing to output array at this offset (exclusive).
* @exception MalformedInputException if the input buffer contains any sequence of bytes that is
* illegal for the input character set.
* @exception UnknownCharacterException for any character that that cannot be converted to
* Unicode. Thrown only when converter is not in substitution mode.
* @exception ConversionBufferFullException if output array is filled prior to converting all the
* input.
*/
public int convert(byte[] input, int inStart, int inEnd, char[] output, int outStart, int outEnd)
throws UnknownCharacterException, MalformedInputException, ConversionBufferFullException {
byteOff = inStart;
charOff = outStart;
// throw exceptions compatible to legacy ByteToCharXxx converters
if (inStart >= inEnd) return 0;
if (inStart >= input.length) throw new ArrayIndexOutOfBoundsException(inStart);
if (outStart >= outEnd || outStart >= output.length) throw new ConversionBufferFullException();
if (src != null && src.array() == input) src.position(inStart).limit(inEnd);
else src = ByteBuffer.wrap(input, inStart, inEnd - inStart);
if (dst != null && dst.array() == output) dst.position(outStart).limit(outEnd);
else dst = CharBuffer.wrap(output, outStart, outEnd - outStart);
CoderResult cr;
try {
cr = decoder.decode(src, dst, false);
} catch (IllegalStateException ise) {
cr = decoder.reset().decode(src, dst, false);
} finally {
byteOff = src.position();
charOff = dst.position();
}
if (cr.isUnmappable()) {
badInputLength = cr.length();
throw new UnknownCharacterException();
}
if (cr.isMalformed()) {
badInputLength = cr.length();
throw new MalformedInputException();
}
if (cr.isOverflow()) throw new ConversionBufferFullException();
// Return the length written to the output buffer
if (cr.isUnderflow()) return charOff - outStart;
return -1; // should be never reached
}
static void test(Charset cs, char[] bmpCA, byte[] sbBA) throws Throwable {
String bmpStr = new String(bmpCA);
CharsetDecoder dec =
cs.newDecoder()
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
CharsetEncoder enc =
cs.newEncoder()
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
// getBytes(csn);
byte[] baSC = bmpStr.getBytes(cs.name());
ByteBuffer bf = enc.reset().encode(CharBuffer.wrap(bmpCA));
byte[] baNIO = new byte[bf.limit()];
bf.get(baNIO, 0, baNIO.length);
if (!Arrays.equals(baSC, baNIO))
throw new RuntimeException("getBytes(csn) failed -> " + cs.name());
// getBytes(cs);
baSC = bmpStr.getBytes(cs);
if (!Arrays.equals(baSC, baNIO))
throw new RuntimeException("getBytes(cs) failed -> " + cs.name());
// new String(csn);
String strSC = new String(sbBA, cs.name());
String strNIO = dec.reset().decode(ByteBuffer.wrap(sbBA)).toString();
if (!strNIO.equals(strSC))
throw new RuntimeException("new String(csn) failed -> " + cs.name());
// new String(cs);
strSC = new String(sbBA, cs);
if (!strNIO.equals(strSC)) throw new RuntimeException("new String(cs) failed -> " + cs.name());
// encode unmappable surrogates
if (enc instanceof sun.nio.cs.ArrayEncoder && cs.contains(Charset.forName("ASCII"))) {
if (cs.name().equals("UTF-8")
|| // utf8 handles surrogates
cs.name().equals("CESU-8")) // utf8 handles surrogates
return;
enc.replaceWith(new byte[] {(byte) 'A'});
sun.nio.cs.ArrayEncoder cae = (sun.nio.cs.ArrayEncoder) enc;
String str = "ab\uD800\uDC00\uD800\uDC00cd";
byte[] ba = new byte[str.length() - 2];
int n = cae.encode(str.toCharArray(), 0, str.length(), ba);
if (n != 6 || !"abAAcd".equals(new String(ba, cs.name())))
throw new RuntimeException("encode1(surrogates) failed -> " + cs.name());
ba = new byte[str.length()];
n = cae.encode(str.toCharArray(), 0, str.length(), ba);
if (n != 6 || !"abAAcd".equals(new String(ba, 0, n, cs.name())))
throw new RuntimeException("encode2(surrogates) failed -> " + cs.name());
str = "ab\uD800B\uDC00Bcd";
ba = new byte[str.length()];
n = cae.encode(str.toCharArray(), 0, str.length(), ba);
if (n != 8 || !"abABABcd".equals(new String(ba, 0, n, cs.name())))
throw new RuntimeException("encode3(surrogates) failed -> " + cs.name());
/* sun.nio.cs.ArrayDeEncoder works on the assumption that the
invoker (StringCoder) allocates enough output buf, utf8
and double-byte coder does not check the output buffer limit.
ba = new byte[str.length() - 1];
n = cae.encode(str.toCharArray(), 0, str.length(), ba);
if (n != 7 || !"abABABc".equals(new String(ba, 0, n, cs.name()))) {
throw new RuntimeException("encode4(surrogates) failed -> "
+ cs.name());
}
*/
}
}
ByteToCharConverter(Charset charset, String encoding) {
super(encoding);
decoder = charset.newDecoder().onUnmappableCharacter(CodingErrorAction.REPLACE);
// for compatibility to old ByteToCharASCII converter:
if (charset.name().equals("US-ASCII")) decoder.onMalformedInput(CodingErrorAction.REPLACE);
}
// we assume out is large enough for this conversion
// returns number of filled chars in out buffer
public int convert(byte[] in, int inOffset, int inLength, char[] out) {
final ByteBuffer inBuffer = ByteBuffer.wrap(in, inOffset, inLength);
final CharBuffer outBuffer = CharBuffer.wrap(out, 0, out.length);
myDecoder.decode(inBuffer, outBuffer, false);
return outBuffer.position();
}
/**
* Sets converter into substitution mode. In substitution mode, the converter will replace
* untranslatable characters in the source encoding with the substitution character set by
* setSubstitionChars. When not in substitution mode, the converter will throw an
* UnknownCharacterException when it encounters untranslatable input.
*
* @param doSub if true, enable substitution mode.
* @see #setSubstitutionChars
*/
public void setSubstitutionMode(boolean doSub) {
super.setSubstitutionMode(doSub);
if (decoder != null)
decoder.onUnmappableCharacter(doSub ? CodingErrorAction.REPLACE : CodingErrorAction.REPORT);
}
/**
* Returns the maximum number of characters needed to convert a byte. Useful for calculating the
* maximum output buffer size needed for a particular input buffer.
*/
public int getMaxCharsPerByte() {
if (decoder != null) return (int) Math.ceil(decoder.maxCharsPerByte());
else // only provided for subclasses
return 1;
// Until UTF-16, this will do for every encoding
}
/** Resets converter to its initial state. */
public void reset() {
super.reset();
decoder.reset();
src = null;
dst = null;
}
public void reset() {
myDecoder.reset();
}
