source: josm/trunk/src/org/glassfish/json/UnicodeDetectingInputStream.java@ 6887

Last change on this file since 6887 was 6756, checked in by Don-vip, 10 years ago

fix #9590 - replace org.json with GPL-compliant jsonp + remove mention of old world image removed in r1680
File size: 6.6 KB
Line 
1/*
2 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER.
3 *
4 * Copyright (c) 2012-2013 Oracle and/or its affiliates. All rights reserved.
5 *
6 * The contents of this file are subject to the terms of either the GNU
7 * General Public License Version 2 only ("GPL") or the Common Development
8 * and Distribution License("CDDL") (collectively, the "License"). You
9 * may not use this file except in compliance with the License. You can
10 * obtain a copy of the License at
11 * https://glassfish.dev.java.net/public/CDDL+GPL_1_1.html
12 * or packager/legal/LICENSE.txt. See the License for the specific
13 * language governing permissions and limitations under the License.
14 *
15 * When distributing the software, include this License Header Notice in each
16 * file and include the License file at packager/legal/LICENSE.txt.
17 *
18 * GPL Classpath Exception:
19 * Oracle designates this particular file as subject to the "Classpath"
20 * exception as provided by Oracle in the GPL Version 2 section of the License
21 * file that accompanied this code.
22 *
23 * Modifications:
24 * If applicable, add the following below the License Header, with the fields
25 * enclosed by brackets [] replaced by your own identifying information:
26 * "Portions Copyright [year] [name of copyright owner]"
27 *
28 * Contributor(s):
29 * If you wish your version of this file to be governed by only the CDDL or
30 * only the GPL Version 2, indicate your decision by adding "[Contributor]
31 * elects to include this software in this distribution under the [CDDL or GPL
32 * Version 2] license." If you don't indicate a single choice of license, a
33 * recipient has the option to distribute your version of this file under
34 * either the CDDL, the GPL Version 2 or to extend the choice of license to
35 * its licensees as provided above. However, if you add GPL Version 2 code
36 * and therefore, elected the GPL Version 2 license, then the option applies
37 * only if the new code is made subject to such option by the copyright
38 * holder.
39 */
40
41package org.glassfish.json;
42
43import javax.json.JsonException;
44import java.io.FilterInputStream;
45import java.io.IOException;
46import java.io.InputStream;
47import java.nio.charset.Charset;
48
49/**
50 * A filter stream that detects the unicode encoding for the original
51 * stream
52 *
53 * @author Jitendra Kotamraju
54 */
55class UnicodeDetectingInputStream extends FilterInputStream {
56 private static final Charset UTF_8 = Charset.forName("UTF-8");
57 private static final Charset UTF_16BE = Charset.forName("UTF-16BE");
58 private static final Charset UTF_16LE = Charset.forName("UTF-16LE");
59 private static final Charset UTF_32LE = Charset.forName("UTF-32LE");
60 private static final Charset UTF_32BE = Charset.forName("UTF-32BE");
61
62 private static final byte FF = (byte)0xFF;
63 private static final byte FE = (byte)0xFE;
64 private static final byte EF = (byte)0xEF;
65 private static final byte BB = (byte)0xBB;
66 private static final byte BF = (byte)0xBF;
67 private static final byte NUL = (byte)0x00;
68
69 private final byte[] buf = new byte[4];
70 private int bufLen;
71 private int curIndex;
72 private final Charset charset;
73
74 UnicodeDetectingInputStream(InputStream is) {
75 super(is);
76 charset = detectEncoding();
77 }
78
79 Charset getCharset() {
80 return charset;
81 }
82
83 private void fillBuf() {
84 int b1;
85 int b2;
86 int b3;
87 int b4;
88
89 try {
90 b1 = in.read();
91 if (b1 == -1) {
92 return;
93 }
94
95 b2 = in.read();
96 if (b2 == -1) {
97 bufLen = 1;
98 buf[0] = (byte)b1;
99 return;
100 }
101
102 b3 = in.read();
103 if (b3 == -1) {
104 bufLen = 2;
105 buf[0] = (byte)b1;
106 buf[1] = (byte)b2;
107 return;
108 }
109
110 b4 = in.read();
111 if (b4 == -1) {
112 bufLen = 3;
113 buf[0] = (byte)b1;
114 buf[1] = (byte)b2;
115 buf[2] = (byte)b3;
116 return;
117 }
118 bufLen = 4;
119 buf[0] = (byte)b1;
120 buf[1] = (byte)b2;
121 buf[2] = (byte)b3;
122 buf[3] = (byte)b4;
123 } catch (IOException ioe) {
124 throw new JsonException("I/O error while auto-detecting the encoding of stream", ioe);
125 }
126 }
127
128 private Charset detectEncoding() {
129 fillBuf();
130 if (bufLen < 2) {
131 throw new JsonException("Cannot auto-detect encoding, not enough chars");
132 } else if (bufLen == 4) {
133 // Use BOM to detect encoding
134 if (buf[0] == NUL && buf[1] == NUL && buf[2] == FE && buf[3] == FF) {
135 curIndex = 4;
136 return UTF_32BE;
137 } else if (buf[0] == FF && buf[1] == FE && buf[2] == NUL && buf[3] == NUL) {
138 curIndex = 4;
139 return UTF_32LE;
140 } else if (buf[0] == FE && buf[1] == FF) {
141 curIndex = 2;
142 return UTF_16BE;
143 } else if (buf[0] == FF && buf[1] == FE) {
144 curIndex = 2;
145 return UTF_16LE;
146 } else if (buf[0] == EF && buf[1] == BB && buf[2] == BF) {
147 curIndex = 3;
148 return UTF_8;
149 }
150 // No BOM, just use JSON RFC's encoding algo to auto-detect
151 if (buf[0] == NUL && buf[1] == NUL && buf[2] == NUL) {
152 return UTF_32BE;
153 } else if (buf[0] == NUL && buf[2] == NUL) {
154 return UTF_16BE;
155 } else if (buf[1] == NUL && buf[2] == NUL && buf[3] == NUL) {
156 return UTF_32LE;
157 } else if (buf[1] == NUL && buf[3] == NUL) {
158 return UTF_16LE;
159 }
160 }
161 return UTF_8;
162 }
163
164 @Override
165 public int read() throws IOException {
166 if (curIndex < bufLen) {
167 return buf[curIndex++];
168 }
169 return in.read();
170 }
171
172 @Override
173 public int read(byte b[], int off, int len) throws IOException {
174 if (curIndex < bufLen) {
175 if (len == 0) {
176 return 0;
177 }
178 if (off < 0 || len < 0 || len > b.length -off) {
179 throw new IndexOutOfBoundsException();
180 }
181 int min = Math.min(bufLen-curIndex, len);
182 System.arraycopy(buf, curIndex, b, off, min);
183 curIndex += min;
184 return min;
185 }
186 return in.read(b, off, len);
187 }
188
189}
Note: See TracBrowser for help on using the repository browser.