001: /*
002:
003: ============================================================================
004: The Apache Software License, Version 1.1
005: ============================================================================
006:
007: Copyright (C) 1999-2003 The Apache Software Foundation. All rights reserved.
008:
009: Redistribution and use in source and binary forms, with or without modifica-
010: tion, are permitted provided that the following conditions are met:
011:
012: 1. Redistributions of source code must retain the above copyright notice,
013: this list of conditions and the following disclaimer.
014:
015: 2. Redistributions in binary form must reproduce the above copyright notice,
016: this list of conditions and the following disclaimer in the documentation
017: and/or other materials provided with the distribution.
018:
019: 3. The end-user documentation included with the redistribution, if any, must
020: include the following acknowledgment: "This product includes software
021: developed by the Apache Software Foundation (http://www.apache.org/)."
022: Alternately, this acknowledgment may appear in the software itself, if
023: and wherever such third-party acknowledgments normally appear.
024:
025: 4. The names "Batik" and "Apache Software Foundation" must not be
026: used to endorse or promote products derived from this software without
027: prior written permission. For written permission, please contact
028: apache@apache.org.
029:
030: 5. Products derived from this software may not be called "Apache", nor may
031: "Apache" appear in their name, without prior written permission of the
032: Apache Software Foundation.
033:
034: THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES,
035: INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
036: FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
037: APACHE SOFTWARE FOUNDATION OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
038: INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLU-
039: DING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
040: OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
041: ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
042: (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
043: THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
044:
045: This software consists of voluntary contributions made by many individuals
046: on behalf of the Apache Software Foundation. For more information on the
047: Apache Software Foundation, please see <http://www.apache.org/>.
048:
049: */
050:
051: package org.apache.batik.util.io;
052:
053: import java.io.IOException;
054: import java.io.InputStream;
055:
056: /**
057: * This class represents an object which decodes UTF-16 characters from
058: * a stream of bytes.
059: *
060: * @author <a href="mailto:stephane@hillion.org">Stephane Hillion</a>
061: * @version $Id$
062: */
063: public class UTF16Decoder extends AbstractCharDecoder {
064:
065: /**
066: * Whether the stream's byte-order is big-endian.
067: */
068: protected boolean bigEndian;
069:
070: /**
071: * Creates a new UTF16Decoder.
072: * It is assumed that the byte-order mark is present.
073: * @param is The stream to decode.
074: */
075: public UTF16Decoder(InputStream is) throws IOException {
076: super (is);
077: // Byte-order detection.
078: int b1 = is.read();
079: if (b1 == -1) {
080: endOfStreamError("UTF-16");
081: }
082: int b2 = is.read();
083: if (b2 == -1) {
084: endOfStreamError("UTF-16");
085: }
086: int m = (((b1 & 0xff) << 8) | (b2 & 0xff));
087: switch (m) {
088: case 0xfeff:
089: bigEndian = true;
090: break;
091: case 0xfffe:
092: break;
093: default:
094: charError("UTF-16");
095: }
096: }
097:
098: /**
099: * Creates a new UTF16Decoder.
100: * @param is The stream to decode.
101: * @param be Whether or not the given stream's byte-order is
102: * big-endian.
103: */
104: public UTF16Decoder(InputStream is, boolean be) {
105: super (is);
106: bigEndian = be;
107: }
108:
109: /**
110: * Reads the next character.
111: * @return a character or END_OF_STREAM.
112: */
113: public int readChar() throws IOException {
114: if (position == count) {
115: fillBuffer();
116: }
117: if (count == -1) {
118: return END_OF_STREAM;
119: }
120: byte b1 = buffer[position++];
121: if (position == count) {
122: fillBuffer();
123: }
124: if (count == -1) {
125: endOfStreamError("UTF-16");
126: }
127: byte b2 = buffer[position++];
128: int c = (bigEndian) ? (((b1 & 0xff) << 8) | (b2 & 0xff))
129: : (((b2 & 0xff) << 8) | (b1 & 0xff));
130: if (c == 0xfffe) {
131: charError("UTF-16");
132: }
133: return c;
134: }
135: }
|