コード例 #1
0
 static void checkMalformed(Charset cs, byte[][] malformed) throws Exception {
   boolean failed = false;
   String csn = cs.name();
   System.out.printf("Check malformed <%s>...%n", csn);
   for (boolean direct : new boolean[] {false, true}) {
     for (byte[] bins : malformed) {
       int mlen = bins[0];
       byte[] bin = Arrays.copyOfRange(bins, 1, bins.length);
       CoderResult cr = decodeCR(bin, cs, direct);
       String ashex = "";
       for (int i = 0; i < bin.length; i++) {
         if (i > 0) ashex += " ";
         ashex += Integer.toString((int) bin[i] & 0xff, 16);
       }
       if (!cr.isMalformed()) {
         System.out.printf(
             "        FAIL(direct=%b): [%s] not malformed. -->cr=%s\n",
             direct, ashex, cr.toString());
         failed = true;
       } else if (cr.length() != mlen) {
         System.out.printf(
             "        FAIL(direct=%b): [%s] malformed[len=%d].\n", direct, ashex, cr.length());
         failed = true;
       }
     }
   }
   if (failed) throw new RuntimeException("Check malformed failed " + csn);
 }
コード例 #2
0
 static byte[] encode(char[] cc, Charset cs, boolean testDirect, Time t) throws Exception {
   ByteBuffer bbf;
   CharBuffer cbf;
   CharsetEncoder enc = cs.newEncoder();
   String csn = cs.name();
   if (testDirect) {
     bbf = ByteBuffer.allocateDirect(cc.length * 4);
     cbf = ByteBuffer.allocateDirect(cc.length * 2).asCharBuffer();
     cbf.put(cc).flip();
   } else {
     bbf = ByteBuffer.allocate(cc.length * 4);
     cbf = CharBuffer.wrap(cc);
   }
   CoderResult cr = null;
   long t1 = System.nanoTime() / 1000;
   for (int i = 0; i < iteration; i++) {
     cbf.rewind();
     bbf.clear();
     enc.reset();
     cr = enc.encode(cbf, bbf, true);
   }
   long t2 = System.nanoTime() / 1000;
   t.t = (t2 - t1) / iteration;
   if (cr != CoderResult.UNDERFLOW) {
     System.out.println("ENC-----------------");
     int pos = cbf.position();
     System.out.printf("  cr=%s, cbf.pos=%d, cc[pos]=%x%n", cr.toString(), pos, cc[pos] & 0xffff);
     throw new RuntimeException("Encoding err: " + csn);
   }
   byte[] bb = new byte[bbf.position()];
   bbf.flip();
   bbf.get(bb);
   return bb;
 }
コード例 #3
0
 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;
 }
コード例 #4
0
ファイル: TextDecoder.java プロジェクト: dimitarp/basex
 @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();
 }
コード例 #5
0
  /**
   * 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
  }
コード例 #6
0
  /**
   * 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
  }