UTF8 Reader : Char Reader Writer « File Input Output « Java

Home
Java
1.2D Graphics GUI
2.3D
3.Advanced Graphics
4.Ant
5.Apache Common
6.Chart
7.Class
8.Collections Data Structure
9.Data Type
10.Database SQL JDBC
11.Design Pattern
12.Development Class
13.EJB3
14.Email
15.Event
16.File Input Output
17.Game
18.Generics
19.GWT
20.Hibernate
21.I18N
22.J2EE
23.J2ME
24.JDK 6
25.JNDI LDAP
26.JPA
27.JSP
28.JSTL
29.Language Basics
30.Network Protocol
31.PDF RTF
32.Reflection
33.Regular Expressions
34.Scripting
35.Security
36.Servlets
37.Spring
38.Swing Components
39.Swing JFC
40.SWT JFace Eclipse
41.Threads
42.Tiny Application
43.Velocity
44.Web Services SOA
45.XML
Java » File Input Output » Char Reader WriterScreenshots 
UTF8 Reader
 
/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 
 *      http://www.apache.org/licenses/LICENSE-2.0
 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */



import java.io.InputStream;
import java.io.IOException;
import java.io.Reader;
import java.io.UTFDataFormatException;


/**
 @author Andy Clark, IBM
 *
 @version $Id: UTF8Reader.java 515 2008-03-17 21:02:23Z jfrederic.clere@jboss.com $
 */
public class UTF8Reader
    extends Reader {

    
    //
    // Constants
    //

    /** Default byte buffer size (2048). */
    public static final int DEFAULT_BUFFER_SIZE = 2048;

    // debugging

    /** Debug read. */
    private static final boolean DEBUG_READ = false;

    //
    // Data
    //

    /** Input stream. */
    protected InputStream fInputStream;

    /** Byte buffer. */
    protected byte[] fBuffer;

    /** Offset into buffer. */
    protected int fOffset;

    /** Surrogate character. */
    private int fSurrogate = -1;

    //
    // Constructors
    //

    /** 
     * Constructs a UTF-8 reader from the specified input stream, 
     * buffer size and MessageFormatter.
     *
     @param inputStream The input stream.
     @param size        The initial buffer size.
     */
    public UTF8Reader(InputStream inputStream, int size) {
        fInputStream = inputStream;
        fBuffer = new byte[size];
    }

    //
    // Reader methods
    //

    /**
     * Read a single character.  This method will block until a character is
     * available, an I/O error occurs, or the end of the stream is reached.
     *
     * <p> Subclasses that intend to support efficient single-character input
     * should override this method.
     *
     @return     The character read, as an integer in the range 0 to 16383
     *             (<tt>0x00-0xffff</tt>), or -1 if the end of the stream has
     *             been reached
     *
     @exception  IOException  If an I/O error occurs
     */
    public int read() throws IOException {

        // decode character
        int c = fSurrogate;
        if (fSurrogate == -1) {
            // NOTE: We use the index into the buffer if there are remaining
            //       bytes from the last block read. -Ac
            int index = 0;

            // get first byte
            int b0 = index == fOffset 
                   ? fInputStream.read() : fBuffer[index++0x00FF;
            if (b0 == -1) {
                return -1;
            }

            // UTF-8:   [0xxx xxxx]
            // Unicode: [0000 0000] [0xxx xxxx]
            if (b0 < 0x80) {
                c = (char)b0;
            }

            // UTF-8:   [110y yyyy] [10xx xxxx]
            // Unicode: [0000 0yyy] [yyxx xxxx]
            else if ((b0 & 0xE0== 0xC0) {
                int b1 = index == fOffset 
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b1 == -1) {
                    expectedByte(22);
                }
                if ((b1 & 0xC0!= 0x80) {
                    invalidByte(22, b1);
                }
                c = ((b0 << 60x07C0(b1 & 0x003F);
            }

            // UTF-8:   [1110 zzzz] [10yy yyyy] [10xx xxxx]
            // Unicode: [zzzz yyyy] [yyxx xxxx]
            else if ((b0 & 0xF0== 0xE0) {
                int b1 = index == fOffset
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b1 == -1) {
                    expectedByte(23);
                }
                if ((b1 & 0xC0!= 0x80) {
                    invalidByte(23, b1);
                }
                int b2 = index == fOffset 
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b2 == -1) {
                    expectedByte(33);
                }
                if ((b2 & 0xC0!= 0x80) {
                    invalidByte(33, b2);
                }
                c = ((b0 << 120xF000((b1 << 60x0FC0|
                    (b2 & 0x003F);
            }

            // 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
            else if ((b0 & 0xF8== 0xF0) {
                int b1 = index == fOffset 
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b1 == -1) {
                    expectedByte(24);
                }
                if ((b1 & 0xC0!= 0x80) {
                    invalidByte(23, b1);
                }
                int b2 = index == fOffset 
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b2 == -1) {
                    expectedByte(34);
                }
                if ((b2 & 0xC0!= 0x80) {
                    invalidByte(33, b2);
                }
                int b3 = index == fOffset 
                       ? fInputStream.read() : fBuffer[index++0x00FF;
                if (b3 == -1) {
                    expectedByte(44);
                }
                if ((b3 & 0xC0!= 0x80) {
                    invalidByte(44, b3);
                }
                int uuuuu = ((b0 << 20x001C((b1 >> 40x0003);
                if (uuuuu > 0x10) {
                    invalidSurrogate(uuuuu);
                }
                int wwww = uuuuu - 1;
                int hs = 0xD800 
                         ((wwww << 60x03C0((b1 << 20x003C
                         ((b2 >> 40x0003);
                int ls = 0xDC00 ((b2 << 60x03C0(b3 & 0x003F);
                c = hs;
                fSurrogate = ls;
            }

            // error
            else {
                invalidByte(11, b0);
            }
        }

        // use surrogate
        else {
            fSurrogate = -1;
        }

        return c;

    // read():int

    /**
     * Read characters into a portion of an array.  This method will block
     * until some input is available, an I/O error occurs, or the end of the
     * stream is reached.
     *
     @param      ch     Destination buffer
     @param      offset Offset at which to start storing characters
     @param      length Maximum number of characters to read
     *
     @return     The number of characters read, or -1 if the end of the
     *             stream has been reached
     *
     @exception  IOException  If an I/O error occurs
     */
    public int read(char ch[]int offset, int lengththrows IOException {

        // handle surrogate
        int out = offset;
        if (fSurrogate != -1) {
            ch[offset + 1(char)fSurrogate;
            fSurrogate = -1;
            length--;
            out++;
        }

        // read bytes
        int count = 0;
        if (fOffset == 0) {
            // adjust length to read
            if (length > fBuffer.length) {
                length = fBuffer.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;
        for (int in = 0; in < total; in++) {
            int b0 = fBuffer[in0x00FF;

            // UTF-8:   [0xxx xxxx]
            // Unicode: [0000 0000] [0xxx xxxx]
            if (b0 < 0x80) {
                ch[out++(char)b0;
                continue;
            }

            // UTF-8:   [110y yyyy] [10xx xxxx]
            // Unicode: [0000 0yyy] [yyxx xxxx]
            if ((b0 & 0xE0== 0xC0) {
                int b1 = -1;
                if (++in < total) { 
                    b1 = fBuffer[in0x00FF
                }
                else {
                    b1 = fInputStream.read();
                    if (b1 == -1) {
                        if (out > offset) {
                            fBuffer[0(byte)b0;
                            fOffset = 1;
                            return out - offset;
                        }
                        expectedByte(22);
                    }
                    count++;
                }
                if ((b1 & 0xC0!= 0x80) {
                    if (out > offset) {
                        fBuffer[0(byte)b0;
                        fBuffer[1(byte)b1;
                        fOffset = 2;
                        return out - offset;
                    }
                    invalidByte(22, b1);
                }
                int c = ((b0 << 60x07C0(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[in0x00FF
                }
                else {
                    b1 = fInputStream.read();
                    if (b1 == -1) {
                        if (out > offset) {
                            fBuffer[0(byte)b0;
                            fOffset = 1;
                            return out - offset;
                        }
                        expectedByte(23);
                    }
                    count++;
                }
                if ((b1 & 0xC0!= 0x80) {
                    if (out > offset) {
                        fBuffer[0(byte)b0;
                        fBuffer[1(byte)b1;
                        fOffset = 2;
                        return out - offset;
                    }
                    invalidByte(23, b1);
                }
                int b2 = -1;
                if (++in < total) { 
                    b2 = fBuffer[in0x00FF
                }
                else {
                    b2 = fInputStream.read();
                    if (b2 == -1) {
                        if (out > offset) {
                            fBuffer[0(byte)b0;
                            fBuffer[1(byte)b1;
                            fOffset = 2;
                            return out - offset;
                        }
                        expectedByte(33);
                    }
                    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(33, b2);
                }
                int c = ((b0 << 120xF000((b1 << 60x0FC0|
                        (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[in0x00FF
                }
                else {
                    b1 = fInputStream.read();
                    if (b1 == -1) {
                        if (out > offset) {
                            fBuffer[0(byte)b0;
                            fOffset = 1;
                            return out - offset;
                        }
                        expectedByte(24);
                    }
                    count++;
                }
                if ((b1 & 0xC0!= 0x80) {
                    if (out > offset) {
                        fBuffer[0(byte)b0;
                        fBuffer[1(byte)b1;
                        fOffset = 2;
                        return out - offset;
                    }
                    invalidByte(24, b1);
                }
                int b2 = -1;
                if (++in < total) { 
                    b2 = fBuffer[in0x00FF
                }
                else {
                    b2 = fInputStream.read();
                    if (b2 == -1) {
                        if (out > offset) {
                            fBuffer[0(byte)b0;
                            fBuffer[1(byte)b1;
                            fOffset = 2;
                            return out - offset;
                        }
                        expectedByte(34);
                    }
                    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(34, b2);
                }
                int b3 = -1;
                if (++in < total) { 
                    b3 = fBuffer[in0x00FF
                }
                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(44);
                    }
                    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(44, b2);
                }

                // decode bytes into surrogate characters
                int uuuuu = ((b0 << 20x001C((b1 >> 40x0003);
                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 << 60x03C0(zzzz << 2(yyyyyy >> 4);
                int ls = 0xDC00 ((yyyyyy << 60x03C0| xxxxxx;

                // set characters
                ch[out++(char)hs;
                ch[out++(char)ls;
                count -= 2;
                continue;
            }

            // error
            if (out > offset) {
                fBuffer[0(byte)b0;
                fOffset = 1;
                return out - offset;
            }
            invalidByte(11, b0);
        }

        return count;

    // read(char[],int,int)

    /**
     * Skip characters.  This method will block until some characters are
     * available, an I/O error occurs, or the end of the stream is reached.
     *
     @param  n  The number of characters to skip
     *
     @return    The number of characters actually skipped
     *
     @exception  IOException  If an I/O error occurs
     */
    public long skip(long nthrows IOException {

        long remaining = n;
        final char[] ch = new char[fBuffer.length];
        do {
            int length = ch.length < remaining ? ch.length : (int)remaining;
            int count = read(ch, 0, length);
            if (count > 0) {
                remaining -= count;
            }
            else {
                break;
            }
        while (remaining > 0);

        long skipped = n - remaining;
        return skipped;

    // skip(long):long

    /**
     * Tell whether this stream is ready to be read.
     *
     @return True if the next read() is guaranteed not to block for input,
     * false otherwise.  Note that returning false does not guarantee that the
     * next read will block.
     *
     @exception  IOException  If an I/O error occurs
     */
    public boolean ready() throws IOException {
      return false;
    // ready()

    /**
     * Tell whether this stream supports the mark() operation.
     */
    public boolean markSupported() {
      return false;
    // markSupported()

    /**
     * Mark the present position in the stream.  Subsequent calls to reset()
     * will attempt to reposition the stream to this point.  Not all
     * character-input streams support the mark() operation.
     *
     @param  readAheadLimit  Limit on the number of characters that may be
     *                         read while still preserving the mark.  After
     *                         reading this many characters, attempting to
     *                         reset the stream may fail.
     *
     @exception  IOException  If the stream does not support mark(),
     *                          or if some other I/O error occurs
     */
    public void mark(int readAheadLimitthrows IOException {
  throw new IOException("operationNotSupported");
    }

    /**
     * Reset the stream.  If the stream has been marked, then attempt to
     * reposition it at the mark.  If the stream has not been marked, then
     * attempt to reset it in some way appropriate to the particular stream,
     * for example by repositioning it to its starting point.  Not all
     * character-input streams support the reset() operation, and some support
     * reset() without supporting mark().
     *
     @exception  IOException  If the stream has not been marked,
     *                          or if the mark has been invalidated,
     *                          or if the stream does not support reset(),
     *                          or if some other I/O error occurs
     */
    public void reset() throws IOException {
        fOffset = 0;
        fSurrogate = -1;
    // reset()

    /**
     * Close the stream.  Once a stream has been closed, further read(),
     * ready(), mark(), or reset() invocations will throw an IOException.
     * Closing a previously-closed stream, however, has no effect.
     *
     @exception  IOException  If an I/O error occurs
     */
    public void close() throws IOException {
        fInputStream.close();
    // close()

    //
    // Private methods
    //

    /** Throws an exception for expected byte. */
    private void expectedByte(int position, int count)
        throws UTFDataFormatException {

        throw new UTFDataFormatException("expectedByte");

    // expectedByte(int,int,int)

    /** Throws an exception for invalid byte. */
    private void invalidByte(int position, int count, int c
        throws UTFDataFormatException {

        throw new UTFDataFormatException("invalidByte");
    // invalidByte(int,int,int,int)

    /** Throws an exception for invalid surrogate bits. */
    private void invalidSurrogate(int uuuuuthrows UTFDataFormatException {
        
        throw new UTFDataFormatException("invalidHighSurrogate");
    // invalidSurrogate(int)

// class UTF8Reader

   
  
Related examples in the same category
1.Char Array IO
2.uses a pair of CharArrayReaders
3.Demonstrate CharArrayWriter.
4.Safe UTF: 64K serialized size
java2s.com  | Contact Us | Privacy Policy
Copyright 2009 - 12 Demo Source and Support. All rights reserved.
All other trademarks are property of their respective owners.