Hadoop的Text类型实现

Hadoop的Text类型是将字符串用UTF-8编码转换成bytes位数组。

  /**

   * Converts the provided String to bytes using the
   * UTF-8 encoding. If <code>replace</code> is true, then
   * malformed input is replaced with the
   * substitution character, which is U+FFFD. Otherwise the
   * method throws a MalformedInputException.
   * @return ByteBuffer: bytes stores at ByteBuffer.array() 
   *                     and length is ByteBuffer.limit()
   */
  public static ByteBuffer encode(String string, boolean replace)
    throws CharacterCodingException {
    CharsetEncoder encoder = ENCODER_FACTORY.get();
    if (replace) {
      encoder.onMalformedInput(CodingErrorAction.REPLACE);
      encoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
    }
    ByteBuffer bytes = 
      encoder.encode(CharBuffer.wrap(string.toCharArray()));
    if (replace) {
      encoder.onMalformedInput(CodingErrorAction.REPORT);
      encoder.onUnmappableCharacter(CodingErrorAction.REPORT);
    }
    return bytes;
  }