1 | /**********************************************************************
|
---|
2 | *
|
---|
3 | *
|
---|
4 | * URLUTF8Encoder.java -- encode UTF8 String into a URL form
|
---|
5 | *
|
---|
6 | * W3C® SOFTWARE NOTICE AND LICENSE
|
---|
7 | *
|
---|
8 | * Copyright © 1994-2000 World Wide Web Consortium,
|
---|
9 | * (Massachusetts Institute of Technology, Institut National de
|
---|
10 | * Recherche en Informatique et en Automatique, Keio University).
|
---|
11 | * All Rights Reserved. http://www.w3.org/Consortium/Legal/
|
---|
12 | *
|
---|
13 | * This W3C work (including software, documents, or other related items) is
|
---|
14 | * being provided by the copyright holders under the following license. By
|
---|
15 | * obtaining, using and/or copying this work, you (the licensee) agree that
|
---|
16 | * you have read, understood, and will comply with the following terms and
|
---|
17 | * conditions:
|
---|
18 | *
|
---|
19 | * Permission to use, copy, modify, and distribute this software and its
|
---|
20 | * documentation, with or without modification, for any purpose and without
|
---|
21 | * fee or royalty is hereby granted, provided that you include the following
|
---|
22 | * on ALL copies of the software and documentation or portions thereof,
|
---|
23 | * including modifications, that you make:
|
---|
24 | *
|
---|
25 | * 1. The full text of this NOTICE in a location viewable to users of the
|
---|
26 | * redistributed or derivative work.
|
---|
27 | *
|
---|
28 | * 2. Any pre-existing intellectual property disclaimers, notices, or terms
|
---|
29 | * and conditions. If none exist, a short notice of the following form
|
---|
30 | * (hypertext is preferred, text is permitted) should be used within the
|
---|
31 | * body of any redistributed or derivative code: "Copyright ©
|
---|
32 | * [$date-of-software] World Wide Web Consortium, (Massachusetts
|
---|
33 | * Institute of Technology, Institut National de Recherche en
|
---|
34 | * Informatique et en Automatique, Keio University). All Rights
|
---|
35 | * Reserved. http://www.w3.org/Consortium/Legal/"
|
---|
36 | *
|
---|
37 | * 3. Notice of any changes or modifications to the W3C files, including the
|
---|
38 | * date changes were made. (We recommend you provide URIs to the location
|
---|
39 | * from which the code is derived.)
|
---|
40 | *
|
---|
41 | * THIS SOFTWARE AND DOCUMENTATION IS PROVIDED "AS IS," AND COPYRIGHT HOLDERS
|
---|
42 | * MAKE NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT
|
---|
43 | * NOT LIMITED TO, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR
|
---|
44 | * PURPOSE OR THAT THE USE OF THE SOFTWARE OR DOCUMENTATION WILL NOT INFRINGE
|
---|
45 | * ANY THIRD PARTY PATENTS, COPYRIGHTS, TRADEMARKS OR OTHER RIGHTS.
|
---|
46 | *
|
---|
47 | * COPYRIGHT HOLDERS WILL NOT BE LIABLE FOR ANY DIRECT, INDIRECT, SPECIAL OR
|
---|
48 | * CONSEQUENTIAL DAMAGES ARISING OUT OF ANY USE OF THE SOFTWARE OR
|
---|
49 | * DOCUMENTATION.
|
---|
50 | *
|
---|
51 | * The name and trademarks of copyright holders may NOT be used in advertising
|
---|
52 | * or publicity pertaining to the software without specific, written prior
|
---|
53 | * permission. Title to copyright in this software and any associated
|
---|
54 | * documentation will at all times remain with copyright holders.
|
---|
55 | *
|
---|
56 | *********************************************************************/
|
---|
57 |
|
---|
58 | /**
|
---|
59 | * Provides a method to encode any string into a URL-safe
|
---|
60 | * form, the so-called "x-www-form-urlencoded" form.
|
---|
61 | * Non-ASCII characters are first encoded as sequences of
|
---|
62 | * two or three bytes, using the UTF-8 algorithm, before being
|
---|
63 | * encoded in "x-www-form-urlencoded".
|
---|
64 | */
|
---|
65 |
|
---|
66 | //package org.nzdl.gsdl.Phind;
|
---|
67 | package org.greenstone.applet.phind;
|
---|
68 | public class URLUTF8Encoder
|
---|
69 | {
|
---|
70 |
|
---|
71 | final static String[] hex = {
|
---|
72 | "%00", "%01", "%02", "%03", "%04", "%05", "%06", "%07",
|
---|
73 | "%08", "%09", "%0a", "%0b", "%0c", "%0d", "%0e", "%0f",
|
---|
74 | "%10", "%11", "%12", "%13", "%14", "%15", "%16", "%17",
|
---|
75 | "%18", "%19", "%1a", "%1b", "%1c", "%1d", "%1e", "%1f",
|
---|
76 | "%20", "%21", "%22", "%23", "%24", "%25", "%26", "%27",
|
---|
77 | "%28", "%29", "%2a", "%2b", "%2c", "%2d", "%2e", "%2f",
|
---|
78 | "%30", "%31", "%32", "%33", "%34", "%35", "%36", "%37",
|
---|
79 | "%38", "%39", "%3a", "%3b", "%3c", "%3d", "%3e", "%3f",
|
---|
80 | "%40", "%41", "%42", "%43", "%44", "%45", "%46", "%47",
|
---|
81 | "%48", "%49", "%4a", "%4b", "%4c", "%4d", "%4e", "%4f",
|
---|
82 | "%50", "%51", "%52", "%53", "%54", "%55", "%56", "%57",
|
---|
83 | "%58", "%59", "%5a", "%5b", "%5c", "%5d", "%5e", "%5f",
|
---|
84 | "%60", "%61", "%62", "%63", "%64", "%65", "%66", "%67",
|
---|
85 | "%68", "%69", "%6a", "%6b", "%6c", "%6d", "%6e", "%6f",
|
---|
86 | "%70", "%71", "%72", "%73", "%74", "%75", "%76", "%77",
|
---|
87 | "%78", "%79", "%7a", "%7b", "%7c", "%7d", "%7e", "%7f",
|
---|
88 | "%80", "%81", "%82", "%83", "%84", "%85", "%86", "%87",
|
---|
89 | "%88", "%89", "%8a", "%8b", "%8c", "%8d", "%8e", "%8f",
|
---|
90 | "%90", "%91", "%92", "%93", "%94", "%95", "%96", "%97",
|
---|
91 | "%98", "%99", "%9a", "%9b", "%9c", "%9d", "%9e", "%9f",
|
---|
92 | "%a0", "%a1", "%a2", "%a3", "%a4", "%a5", "%a6", "%a7",
|
---|
93 | "%a8", "%a9", "%aa", "%ab", "%ac", "%ad", "%ae", "%af",
|
---|
94 | "%b0", "%b1", "%b2", "%b3", "%b4", "%b5", "%b6", "%b7",
|
---|
95 | "%b8", "%b9", "%ba", "%bb", "%bc", "%bd", "%be", "%bf",
|
---|
96 | "%c0", "%c1", "%c2", "%c3", "%c4", "%c5", "%c6", "%c7",
|
---|
97 | "%c8", "%c9", "%ca", "%cb", "%cc", "%cd", "%ce", "%cf",
|
---|
98 | "%d0", "%d1", "%d2", "%d3", "%d4", "%d5", "%d6", "%d7",
|
---|
99 | "%d8", "%d9", "%da", "%db", "%dc", "%dd", "%de", "%df",
|
---|
100 | "%e0", "%e1", "%e2", "%e3", "%e4", "%e5", "%e6", "%e7",
|
---|
101 | "%e8", "%e9", "%ea", "%eb", "%ec", "%ed", "%ee", "%ef",
|
---|
102 | "%f0", "%f1", "%f2", "%f3", "%f4", "%f5", "%f6", "%f7",
|
---|
103 | "%f8", "%f9", "%fa", "%fb", "%fc", "%fd", "%fe", "%ff"
|
---|
104 | };
|
---|
105 |
|
---|
106 | /**
|
---|
107 | * Encode a string to the "x-www-form-urlencoded" form, enhanced
|
---|
108 | * with the UTF-8-in-URL proposal. This is what happens:
|
---|
109 | *
|
---|
110 | * <ul>
|
---|
111 | * <li><p>The ASCII characters 'a' through 'z', 'A' through 'Z',
|
---|
112 | * and '0' through '9' remain the same.
|
---|
113 | *
|
---|
114 | * <li><p>The space character ' ' is converted into a plus sign '+'.
|
---|
115 | *
|
---|
116 | * <li><p>All other ASCII characters are converted into the
|
---|
117 | * 3-character string "%xy", where xy is
|
---|
118 | * the two-digit hexadecimal representation of the character
|
---|
119 | * code
|
---|
120 | *
|
---|
121 | * <li><p>All non-ASCII characters are encoded in two steps: first
|
---|
122 | * to a sequence of 2 or 3 bytes, using the UTF-8 algorithm;
|
---|
123 | * secondly each of these bytes is encoded as "%xx".
|
---|
124 | * </ul>
|
---|
125 | *
|
---|
126 | * @param s The string to be encoded
|
---|
127 | * @return The encoded string
|
---|
128 | */
|
---|
129 | public static String encode(String s)
|
---|
130 | {
|
---|
131 | StringBuffer sbuf = new StringBuffer();
|
---|
132 | int len = s.length();
|
---|
133 | for (int i = 0; i < len; i++) {
|
---|
134 | int ch = s.charAt(i);
|
---|
135 | if ('A' <= ch && ch <= 'Z') { // 'A'..'Z'
|
---|
136 | sbuf.append((char)ch);
|
---|
137 | } else if ('a' <= ch && ch <= 'z') { // 'a'..'z'
|
---|
138 | sbuf.append((char)ch);
|
---|
139 | } else if ('0' <= ch && ch <= '9') { // '0'..'9'
|
---|
140 | sbuf.append((char)ch);
|
---|
141 | } else if (ch == ' ') { // space
|
---|
142 | sbuf.append('+');
|
---|
143 | } else if (ch <= 0x007f) { // other ASCII
|
---|
144 | sbuf.append(hex[ch]);
|
---|
145 | } else if (ch <= 0x07FF) { // non-ASCII <= 0x7FF
|
---|
146 | sbuf.append(hex[0xc0 | (ch >> 6)]);
|
---|
147 | sbuf.append(hex[0x80 | (ch & 0x3F)]);
|
---|
148 | } else { // 0x7FF < ch <= 0xFFFF
|
---|
149 | sbuf.append(hex[0xe0 | (ch >> 12)]);
|
---|
150 | sbuf.append(hex[0x80 | ((ch >> 6) & 0x3F)]);
|
---|
151 | sbuf.append(hex[0x80 | (ch & 0x3F)]);
|
---|
152 | }
|
---|
153 | }
|
---|
154 | return sbuf.toString();
|
---|
155 | }
|
---|
156 |
|
---|
157 | }
|
---|
158 |
|
---|
159 |
|
---|