001/* URLEncoder.java -- Class to convert strings to a properly encoded URL
002   Copyright (C) 1998, 1999, 2001, 2002, 2003 Free Software Foundation, Inc.
003
004This file is part of GNU Classpath.
005
006GNU Classpath is free software; you can redistribute it and/or modify
007it under the terms of the GNU General Public License as published by
008the Free Software Foundation; either version 2, or (at your option)
009any later version.
010
011GNU Classpath is distributed in the hope that it will be useful, but
012WITHOUT ANY WARRANTY; without even the implied warranty of
013MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
014General Public License for more details.
015
016You should have received a copy of the GNU General Public License
017along with GNU Classpath; see the file COPYING.  If not, write to the
018Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
01902110-1301 USA.
020
021Linking this library statically or dynamically with other modules is
022making a combined work based on this library.  Thus, the terms and
023conditions of the GNU General Public License cover the whole
024combination.
025
026As a special exception, the copyright holders of this library give you
027permission to link this library with independent modules to produce an
028executable, regardless of the license terms of these independent
029modules, and to copy and distribute the resulting executable under
030terms of your choice, provided that you also meet, for each linked
031independent module, the terms and conditions of the license of that
032module.  An independent module is a module which is not derived from
033or based on this library.  If you modify this library, you may extend
034this exception to your version of the library, but you are not
035obligated to do so.  If you do not wish to do so, delete this
036exception statement from your version. */
037
038package java.net;
039
040import gnu.java.lang.CPStringBuilder;
041
042import java.io.UnsupportedEncodingException;
043
044
045/*
046 * Written using on-line Java Platform 1.2/1.4 API Specification, as well
047 * as "The Java Class Libraries", 2nd edition (Addison-Wesley, 1998).
048 * Status:  Believed complete and correct.
049 */
050
051/**
052 * This utility class contains static methods that converts a
053 * string into a fully encoded URL string in x-www-form-urlencoded
054 * format.  This format replaces certain disallowed characters with
055 * encoded equivalents.  All upper case and lower case letters in the
056 * US alphabet remain as is, the space character (' ') is replaced with
057 * '+' sign, and all other characters are converted to a "%XX" format
058 * where XX is the hexadecimal representation of that character in a
059 * certain encoding (by default, the platform encoding, though the
060 * standard is "UTF-8").
061 * <p>
062 * This method is very useful for encoding strings to be sent to CGI scripts
063 *
064 * @author Aaron M. Renn (arenn@urbanophile.com)
065 * @author Warren Levy (warrenl@cygnus.com)
066 * @author Mark Wielaard (mark@klomp.org)
067 */
068public class URLEncoder
069{
070  /**
071   * This method translates the passed in string into x-www-form-urlencoded
072   * format using the default encoding.  The standard encoding is
073   * "UTF-8", and the two-argument form of this method should be used
074   * instead.
075   *
076   * @param s The String to convert
077   *
078   * @return The converted String
079   *
080   * @deprecated
081   */
082  public static String encode(String s)
083  {
084    try
085      {
086        // We default to 8859_1 for compatibility with the same
087        // default elsewhere in the library.
088        return encode(s, System.getProperty("file.encoding", "8859_1"));
089      }
090    catch (UnsupportedEncodingException uee)
091      {
092        // Should never happen since default should always be supported
093        return s;
094      }
095  }
096
097  /**
098   * This method translates the passed in string into x-www-form-urlencoded
099   * format using the character encoding to hex-encode the unsafe characters.
100   *
101   * @param s The String to convert
102   * @param encoding The encoding to use for unsafe characters
103   *
104   * @return The converted String
105   *
106   * @exception UnsupportedEncodingException If the named encoding is not
107   * supported
108   *
109   * @since 1.4
110   */
111  public static String encode(String s, String encoding)
112    throws UnsupportedEncodingException
113  {
114    int length = s.length();
115    int start = 0;
116    int i = 0;
117
118    CPStringBuilder result = new CPStringBuilder(length);
119    while (true)
120      {
121        while (i < length && isSafe(s.charAt(i)))
122          i++;
123
124        // Safe character can just be added
125        result.append(s.substring(start, i));
126
127        // Are we done?
128        if (i >= length)
129          return result.toString();
130        else if (s.charAt(i) == ' ')
131          {
132            result.append('+'); // Replace space char with plus symbol.
133            i++;
134          }
135        else
136          {
137            // Get all unsafe characters
138            start = i;
139            char c;
140            while (i < length && (c = s.charAt(i)) != ' ' && ! isSafe(c))
141              i++;
142
143            // Convert them to %XY encoded strings
144            String unsafe = s.substring(start, i);
145            byte[] bytes = unsafe.getBytes(encoding);
146            for (int j = 0; j < bytes.length; j++)
147              {
148                result.append('%');
149                int val = bytes[j];
150                result.append(hex.charAt((val & 0xf0) >> 4));
151                result.append(hex.charAt(val & 0x0f));
152              }
153          }
154        start = i;
155      }
156  }
157
158  /**
159   * Private static method that returns true if the given char is either
160   * a uppercase or lowercase letter from 'a' till 'z', or a digit froim
161   * '0' till '9', or one of the characters '-', '_', '.' or '*'. Such
162   * 'safe' character don't have to be url encoded.
163   */
164  private static boolean isSafe(char c)
165  {
166    return ((c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
167           || (c >= '0' && c <= '9') || c == '-' || c == '_' || c == '.'
168           || c == '*');
169  }
170
171  /**
172   * Private constructor that does nothing. Included to avoid a default
173   * public constructor being created by the compiler.
174   */
175  private URLEncoder()
176  {
177  }
178
179  /**
180   * Used to convert to hex.  We don't use Integer.toHexString, since
181   * it converts to lower case (and the Sun docs pretty clearly
182   * specify upper case here), and because it doesn't provide a
183   * leading 0.
184   */
185  private static final String hex = "0123456789ABCDEF";
186}