1 /*
2 * Conditions Of Use
3 *
4 * This software was developed by employees of the National Institute of
5 * Standards and Technology (NIST), an agency of the Federal Government.
6 * Pursuant to title 15 Untied States Code Section 105, works of NIST
7 * employees are not subject to copyright protection in the United States
8 * and are considered to be in the public domain.  As a result, a formal
9 * license is not needed to use the software.
10 *
11 * This software is provided by NIST as a service and is expressly
12 * provided "AS IS."  NIST MAKES NO WARRANTY OF ANY KIND, EXPRESS, IMPLIED
13 * OR STATUTORY, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTY OF
14 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NON-INFRINGEMENT
15 * AND DATA ACCURACY.  NIST does not warrant or make any representations
16 * regarding the use of the software or the results thereof, including but
17 * not limited to the correctness, accuracy, reliability or usefulness of
18 * the software.
19 *
20 * Permission to use this software is contingent upon your acceptance
21 * of the terms of this agreement
22 *
23 * .
24 *
25 */
26 package gov.nist.javax.sip.address;
27 
28 import java.io.UnsupportedEncodingException;
29 
30 /**
31  * Copied from Apache Excalibur project.
32  * Source code available at http://www.google.com/codesearch?hl=en&q=+excalibur+decodePath+show:sK_gDY0W5Rw:OTjCHAiSuF0:th3BdHtpX20&sa=N&cd=1&ct=rc&cs_p=http://apache.edgescape.com/excalibur/excalibur-sourceresolve/source/excalibur-sourceresolve-1.1-src.zip&cs_f=excalibur-sourceresolve-1.1/src/java/org/apache/excalibur/source/SourceUtil.java
33  * @author <A HREF="mailto:jean.deruelle@gmail.com">Jean Deruelle</A>
34  *
35  */
36 public class RFC2396UrlDecoder {
37 
38     /**
39      * Decode a path.
40      *
41      * <p>Interprets %XX (where XX is hexadecimal number) as UTF-8 encoded bytes.
42      * <p>The validity of the input path is not checked (i.e. characters that
43      * were not encoded will not be reported as errors).
44      * <p>This method differs from URLDecoder.decode in that it always uses UTF-8
45      * (while URLDecoder uses the platform default encoding, often ISO-8859-1),
46      * and doesn't translate + characters to spaces.
47      *
48      * @param uri the path to decode
49      * @return the decoded path
50      */
decode(String uri)51     public static String decode(String uri) {
52         StringBuffer translatedUri = new StringBuffer(uri.length());
53         byte[] encodedchars = new byte[uri.length() / 3];
54         int i = 0;
55         int length = uri.length();
56         int encodedcharsLength = 0;
57         while (i < length) {
58             if (uri.charAt(i) == '%') {
59                 //we must process all consecutive %-encoded characters in one go, because they represent
60                 //an UTF-8 encoded string, and in UTF-8 one character can be encoded as multiple bytes
61                 while (i < length && uri.charAt(i) == '%') {
62                     if (i + 2 < length) {
63                         try {
64                             byte x = (byte)Integer.parseInt(uri.substring(i + 1, i + 3), 16);
65                             encodedchars[encodedcharsLength] = x;
66                         } catch (NumberFormatException e) {
67                             throw new IllegalArgumentException("Illegal hex characters in pattern %" + uri.substring(i + 1, i + 3));
68                         }
69                         encodedcharsLength++;
70                         i += 3;
71                     } else {
72                         throw new IllegalArgumentException("% character should be followed by 2 hexadecimal characters.");
73                     }
74                 }
75                 try {
76                     String translatedPart = new String(encodedchars, 0, encodedcharsLength, "UTF-8");
77                     translatedUri.append(translatedPart);
78                 } catch (UnsupportedEncodingException e) {
79                     //the situation that UTF-8 is not supported is quite theoretical, so throw a runtime exception
80                     throw new RuntimeException("Problem in decodePath: UTF-8 encoding not supported.");
81                 }
82                 encodedcharsLength = 0;
83             } else {
84                 //a normal character
85                 translatedUri.append(uri.charAt(i));
86                 i++;
87             }
88         }
89         return translatedUri.toString();
90     }
91 }