string.java

来自「linux下建立JAVA虚拟机的源码KAFFE」· Java 代码 · 共 1,934 行 · 第 1/5 页

   * @throws UnsupportedEncodingException if encoding is not found
   * @throws Error if the decoding fails
   * @see #String(byte[], int, int, String)
   * @since 1.1
   */
  public String(byte[] data, String encoding)
    throws UnsupportedEncodingException
  {
    this(data, 0, data.length, encoding);
  }

  /**
   * Creates a new String using the portion of the byte array starting at the
   * offset and ending at offset + count. Uses the encoding of the platform's
   * default charset, so the resulting string may be longer or shorter than
   * the byte array. For more decoding control, use
   * {@link java.nio.charset.CharsetDecoder}.  The behavior is not specified
   * if the decoder encounters invalid characters; this implementation throws
   * an Error.
   *
   * @param data byte array to copy
   * @param offset the offset to start at
   * @param count the number of bytes in the array to use
   * @throws NullPointerException if data is null
   * @throws IndexOutOfBoundsException if offset or count is incorrect
   * @throws Error if the decoding fails
   * @see #String(byte[], int, int, String)
   * @since 1.1
   */
  public String(byte[] data, int offset, int count)
  {
    if (offset < 0)
      throw new StringIndexOutOfBoundsException("offset: " + offset);
    if (count < 0)
      throw new StringIndexOutOfBoundsException("count: " + count);
    // equivalent to: offset + count < 0 || offset + count > data.length
    if (data.length - offset < count)
      throw new StringIndexOutOfBoundsException("offset + count: "
						+ (offset + count));
    int o, c;
    char[] v;
    String encoding;
    try 
	{
	  encoding = System.getProperty("file.encoding");
	  CharsetDecoder csd = Charset.forName(encoding).newDecoder();
	  csd.onMalformedInput(CodingErrorAction.REPLACE);
	  csd.onUnmappableCharacter(CodingErrorAction.REPLACE);
	  CharBuffer cbuf = csd.decode(ByteBuffer.wrap(data, offset, count));
	  if(cbuf.hasArray())
	    {
              v = cbuf.array();
	      o = cbuf.position();
	      c = cbuf.remaining();
	    } else {
	      // Doubt this will happen. But just in case.
	      v = new char[cbuf.remaining()];
	      cbuf.get(v);
	      o = 0;
	      c = v.length;
	    }
	} catch(Exception ex){
	    // If anything goes wrong (System property not set,
	    // NIO provider not available, etc)
	    // Default to the 'safe' encoding ISO8859_1
	    v = new char[count];
	    o = 0;
	    c = count;
	    for (int i=0;i<count;i++)
	      v[i] = (char)data[offset+i];
	}
    this.value = v;
    this.offset = o;
    this.count = c;
  }

  /**
   * Creates a new String using the byte array. Uses the encoding of the
   * platform's default charset, so the resulting string may be longer or
   * shorter than the byte array. For more decoding control, use
   * {@link java.nio.charset.CharsetDecoder}.  The behavior is not specified
   * if the decoder encounters invalid characters; this implementation throws
   * an Error.
   *
   * @param data byte array to copy
   * @throws NullPointerException if data is null
   * @throws Error if the decoding fails
   * @see #String(byte[], int, int)
   * @see #String(byte[], int, int, String)
   * @since 1.1
   */
  public String(byte[] data)
  {
    this(data, 0, data.length);
  }

  /**
   * Creates a new String using the character sequence represented by
   * the StringBuffer. Subsequent changes to buf do not affect the String.
   *
   * @param buffer StringBuffer to copy
   * @throws NullPointerException if buffer is null
   */
  public String(StringBuffer buffer)
  {
    synchronized (buffer)
      {
        offset = 0;
        count = buffer.count;
        // Share unless buffer is 3/4 empty.
        if ((count << 2) < buffer.value.length)
          {
            value = new char[count];
            VMSystem.arraycopy(buffer.value, 0, value, 0, count);
          }
        else
          {
            buffer.shared = true;
            value = buffer.value;
          }
      }
  }

  /**
   * Creates a new String using the character sequence represented by
   * the StringBuilder. Subsequent changes to buf do not affect the String.
   *
   * @param buffer StringBuilder to copy
   * @throws NullPointerException if buffer is null
   */
  public String(StringBuilder buffer)
  {
    this(buffer.value, 0, buffer.count);
  }

  /**
   * Special constructor which can share an array when safe to do so.
   *
   * @param data the characters to copy
   * @param offset the location to start from
   * @param count the number of characters to use
   * @param dont_copy true if the array is trusted, and need not be copied
   * @throws NullPointerException if chars is null
   * @throws StringIndexOutOfBoundsException if bounds check fails
   */
  String(char[] data, int offset, int count, boolean dont_copy)
  {
    if (offset < 0)
      throw new StringIndexOutOfBoundsException("offset: " + offset);
    if (count < 0)
      throw new StringIndexOutOfBoundsException("count: " + count);
    // equivalent to: offset + count < 0 || offset + count > data.length
    if (data.length - offset < count)
      throw new StringIndexOutOfBoundsException("offset + count: "
						+ (offset + count));
    if (dont_copy)
      {
        value = data;
        this.offset = offset;
      }
    else
      {
        value = new char[count];
        VMSystem.arraycopy(data, offset, value, 0, count);
        this.offset = 0;
      }
    this.count = count;
  }

  /**
   * Creates a new String containing the characters represented in the
   * given subarray of Unicode code points.
   * @param codePoints the entire array of code points
   * @param offset the start of the subarray
   * @param count the length of the subarray
   * 
   * @throws IllegalArgumentException if an invalid code point is found
   * in the codePoints array
   * @throws IndexOutOfBoundsException if offset is negative or offset + count
   * is greater than the length of the array.
   */
  public String(int[] codePoints, int offset, int count)
  {
    // FIXME: This implementation appears to give correct internal
    // representation of the String because: 
    //   - length() is correct
    //   - getting a char[] from toCharArray() and testing 
    //     Character.codePointAt() on all the characters in that array gives
    //     the appropriate results
    // however printing the String gives incorrect results.  This may be 
    // due to printing method errors (such as incorrectly looping through
    // the String one char at a time rather than one "character" at a time.
    
    if (offset < 0)
      throw new IndexOutOfBoundsException();
    int end = offset + count;
    int pos = 0;
    // This creates a char array that is long enough for all of the code
    // points to represent supplementary characters.  This is more than likely
    // a waste of storage, so we use it only temporarily and then copy the 
    // used portion into the value array.
    char[] temp = new char[2 * codePoints.length];
    for (int i = offset; i < end; i++)
      {
        pos += Character.toChars(codePoints[i], temp, pos);        
      }
    this.count = pos;
    this.value = new char[pos];
    System.arraycopy(temp, 0, value, 0, pos);
    this.offset = 0;
  }
  
  /**
   * Returns the number of characters contained in this String.
   *
   * @return the length of this String
   */
  public int length()
  {
    return count;
  }

  /**
   * Returns the character located at the specified index within this String.
   *
   * @param index position of character to return (base 0)
   * @return character located at position index
   * @throws IndexOutOfBoundsException if index &lt; 0 || index &gt;= length()
   *         (while unspecified, this is a StringIndexOutOfBoundsException)
   */
  public char charAt(int index)
  {
    if (index < 0 || index >= count)
      throw new StringIndexOutOfBoundsException(index);
    return value[offset + index];
  }

  /**
   * Get the code point at the specified index.  This is like #charAt(int),
   * but if the character is the start of a surrogate pair, and the
   * following character completes the pair, then the corresponding
   * supplementary code point is returned.
   * @param index the index of the codepoint to get, starting at 0
   * @return the codepoint at the specified index
   * @throws IndexOutOfBoundsException if index is negative or &gt;= length()
   * @since 1.5
   */
  public synchronized int codePointAt(int index)
  {
    // Use the CharSequence overload as we get better range checking
    // this way.
    return Character.codePointAt(this, index);
  }

  /**
   * Get the code point before the specified index.  This is like
   * #codePointAt(int), but checks the characters at <code>index-1</code> and
   * <code>index-2</code> to see if they form a supplementary code point.
   * @param index the index just past the codepoint to get, starting at 0
   * @return the codepoint at the specified index
   * @throws IndexOutOfBoundsException if index is negative or &gt;= length()
   *         (while unspecified, this is a StringIndexOutOfBoundsException)
   * @since 1.5
   */
  public synchronized int codePointBefore(int index)
  {
    // Use the CharSequence overload as we get better range checking
    // this way.
    return Character.codePointBefore(this, index);
  }

  /**
   * Copies characters from this String starting at a specified start index,
   * ending at a specified stop index, to a character array starting at
   * a specified destination begin index.
   *
   * @param srcBegin index to begin copying characters from this String
   * @param srcEnd index after the last character to be copied from this String
   * @param dst character array which this String is copied into
   * @param dstBegin index to start writing characters into dst
   * @throws NullPointerException if dst is null
   * @throws IndexOutOfBoundsException if any indices are out of bounds
   *         (while unspecified, source problems cause a
   *         StringIndexOutOfBoundsException, and dst problems cause an
   *         ArrayIndexOutOfBoundsException)
   */
  public void getChars(int srcBegin, int srcEnd, char dst[], int dstBegin)
  {
    if (srcBegin < 0 || srcBegin > srcEnd || srcEnd > count)
      throw new StringIndexOutOfBoundsException();
    VMSystem.arraycopy(value, srcBegin + offset,
                     dst, dstBegin, srcEnd - srcBegin);
  }

  /**
   * Copies the low byte of each character from this String starting at a
   * specified start index, ending at a specified stop index, to a byte array
   * starting at a specified destination begin index.
   *
   * @param srcBegin index to being copying characters from this String
   * @param srcEnd index after the last character to be copied from this String
   * @param dst byte array which each low byte of this String is copied into
   * @param dstBegin index to start writing characters into dst
   * @throws NullPointerException if dst is null and copy length is non-zero
   * @throws IndexOutOfBoundsException if any indices are out of bounds
   *         (while unspecified, source problems cause a
   *         StringIndexOutOfBoundsException, and dst problems cause an
   *         ArrayIndexOutOfBoundsException)
   * @see #getBytes()
   * @see #getBytes(String)
   * @deprecated use {@link #getBytes()}, which uses a char to byte encoder
   */
  public void getBytes(int srcBegin, int srcEnd, byte dst[], int dstBegin)
  {
    if (srcBegin < 0 || srcBegin > srcEnd || srcEnd > count)
      throw new StringIndexOutOfBoundsException();
    int i = srcEnd - srcBegin;
    srcBegin += offset;
    while (--i >= 0)
      dst[dstBegin++] = (byte) value[srcBegin++];
  }

  /**
   * Converts the Unicode characters in this String to a byte array. Uses the
   * specified encoding method, so the result may be longer or shorter than
   * the String. For more encoding control, use
   * {@link java.nio.charset.CharsetEncoder}, and for valid character sets,
   * see {@link java.nio.charset.Charset}. Unsupported characters get
   * replaced by an encoding specific byte.
   *
   * @param enc encoding name
   * @return the resulting byte array
   * @throws NullPointerException if enc is null
   * @throws UnsupportedEncodingException if encoding is not supported
   * @since 1.1
   */
  public byte[] getBytes(String enc) throws UnsupportedEncodingException
  {
    try 
      {
	CharsetEncoder cse = Charset.forName(enc).newEncoder();
	cse.onMalformedInput(CodingErrorAction.REPLACE);
	cse.onUnmappableCharacter(CodingErrorAction.REPLACE);
	ByteBuffer bbuf = cse.encode(CharBuffer.wrap(value, offset, count));
	if(bbuf.hasArray())
	  return bbuf.array();
	
	// Doubt this will happen. But just in case.
	byte[] bytes = new byte[bbuf.remaining()];
	bbuf.get(bytes);
	return bytes;
      } 
    catch(IllegalCharsetNameException e)
      {
	throw new UnsupportedEncodingException("Encoding: " + enc
					       + " not found.");
      } 
    catch(UnsupportedCharsetException e)
      {
	throw new UnsupportedEncodingException("Encoding: " + enc
					       + " not found.");
      } 
    catch(CharacterCodingException e)
      {
	// This shouldn't ever happen.
	throw (InternalError) new InternalError().initCause(e);
      }	  
  }

  /**
   * Converts the Unicode characters in this String to a byte array. Uses the
   * encoding of the platform's default charset, so the result may be longer
   * or shorter than the String. For more encoding control, use
   * {@link java.nio.charset.CharsetEncoder}. Unsupported characters get
   * replaced by an encoding specific byte.
   *
   * @return the resulting byte array, or null on a problem
   * @since 1.1
   */
  public byte[] getBytes()
  { 
      try 
	  {
	      return getBytes(System.getProperty("file.encoding"));
	  } catch(Exception e) {
	      // XXX - Throw an error here? 
	      // For now, default to the 'safe' encoding.