1 /*
2 * ====================================================================
3 * Licensed to the Apache Software Foundation (ASF) under one
4 * or more contributor license agreements. See the NOTICE file
5 * distributed with this work for additional information
6 * regarding copyright ownership. The ASF licenses this file
7 * to you under the Apache License, Version 2.0 (the
8 * "License"); you may not use this file except in compliance
9 * with the License. You may obtain a copy of the License at
10 *
11 * http://www.apache.org/licenses/LICENSE-2.0
12 *
13 * Unless required by applicable law or agreed to in writing,
14 * software distributed under the License is distributed on an
15 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
16 * KIND, either express or implied. See the License for the
17 * specific language governing permissions and limitations
18 * under the License.
19 * ====================================================================
20 *
21 * This software consists of voluntary contributions made by many
22 * individuals on behalf of the Apache Software Foundation. For more
23 * information on the Apache Software Foundation, please see
24 * <http://www.apache.org/>.
25 *
26 */
27
28 package org.apache.hc.core5.net;
29
30 import java.net.URI;
31 import java.nio.charset.Charset;
32 import java.nio.charset.StandardCharsets;
33 import java.util.ArrayList;
34 import java.util.Arrays;
35 import java.util.BitSet;
36 import java.util.List;
37
38 import org.apache.hc.core5.http.NameValuePair;
39 import org.apache.hc.core5.http.message.BasicNameValuePair;
40 import org.apache.hc.core5.util.Args;
41 import org.apache.hc.core5.util.Tokenizer;
42
43 /**
44 * A collection of utilities for encoding URLs.
45 *
46 * @since 4.0
47 *
48 * @deprecated Use {@link URIBuilder} to parse and format {@link URI}s and
49 * {@link WWWFormCodec} to parse and format {@code application/x-www-form-urlencoded} forms.
50 */
51 @Deprecated
52 public class URLEncodedUtils {
53
54 private static final char QP_SEP_A = '&';
55 private static final char QP_SEP_S = ';';
56
57 /**
58 * Returns a list of {@link NameValuePair}s URI query parameters.
59 * By convention, {@code '&'} and {@code ';'} are accepted as parameter separators.
60 *
61 * @param uri input URI.
62 * @param charset parameter charset.
63 * @return list of query parameters.
64 *
65 * @since 4.5
66 */
67 public static List<NameValuePair> parse(final URI uri, final Charset charset) {
68 Args.notNull(uri, "URI");
69 final String query = uri.getRawQuery();
70 if (query != null && !query.isEmpty()) {
71 return parse(query, charset);
72 }
73 return new ArrayList<>(0);
74 }
75
76 /**
77 * Returns a list of {@link NameValuePair}s URI query parameters.
78 * By convention, {@code '&'} and {@code ';'} are accepted as parameter separators.
79 *
80 * @param s URI query component.
81 * @param charset charset to use when decoding the parameters.
82 * @return list of query parameters.
83 *
84 * @since 4.2
85 */
86 public static List<NameValuePair> parse(final CharSequence s, final Charset charset) {
87 if (s == null) {
88 return new ArrayList<>(0);
89 }
90 return parse(s, charset, QP_SEP_A, QP_SEP_S);
91 }
92
93 /**
94 * Returns a list of {@link NameValuePair}s parameters.
95 *
96 * @param s input text.
97 * @param charset parameter charset.
98 * @param separators parameter separators.
99 * @return list of query parameters.
100 *
101 * @since 4.4
102 */
103 public static List<NameValuePair> parse(
104 final CharSequence s, final Charset charset, final char... separators) {
105 Args.notNull(s, "Char sequence");
106 final Tokenizer tokenParser = Tokenizer.INSTANCE;
107 final BitSet delimSet = new BitSet();
108 for (final char separator: separators) {
109 delimSet.set(separator);
110 }
111 final Tokenizer.Cursor cursor = new Tokenizer.Cursor(0, s.length());
112 final List<NameValuePair> list = new ArrayList<>();
113 while (!cursor.atEnd()) {
114 delimSet.set('=');
115 final String name = tokenParser.parseToken(s, cursor, delimSet);
116 String value = null;
117 if (!cursor.atEnd()) {
118 final int delim = s.charAt(cursor.getPos());
119 cursor.updatePos(cursor.getPos() + 1);
120 if (delim == '=') {
121 delimSet.clear('=');
122 value = tokenParser.parseToken(s, cursor, delimSet);
123 if (!cursor.atEnd()) {
124 cursor.updatePos(cursor.getPos() + 1);
125 }
126 }
127 }
128 if (!name.isEmpty()) {
129 list.add(new BasicNameValuePair(
130 PercentCodec.decode(name, charset, true),
131 PercentCodec.decode(value, charset, true)));
132 }
133 }
134 return list;
135 }
136
137 /**
138 * Returns a list of URI path segments.
139 *
140 * @param s URI path component.
141 * @param charset parameter charset.
142 * @return list of segments.
143 *
144 * @since 4.5
145 */
146 public static List<String> parsePathSegments(final CharSequence s, final Charset charset) {
147 return URIBuilder.parsePath(s, charset);
148 }
149
150 /**
151 * Returns a list of URI path segments.
152 *
153 * @param s URI path component.
154 * @return list of segments.
155 *
156 * @since 4.5
157 */
158 public static List<String> parsePathSegments(final CharSequence s) {
159 return parsePathSegments(s, StandardCharsets.UTF_8);
160 }
161
162 /**
163 * Returns a string consisting of joint encoded path segments.
164 *
165 * @param segments the segments.
166 * @param charset parameter charset.
167 * @return URI path component
168 *
169 * @since 4.5
170 */
171 public static String formatSegments(final Iterable<String> segments, final Charset charset) {
172 Args.notNull(segments, "Segments");
173 final StringBuilder buf = new StringBuilder();
174 URIBuilder.formatPath(buf, segments, false, charset);
175 return buf.toString();
176 }
177
178 /**
179 * Returns a string consisting of joint encoded path segments.
180 *
181 * @param segments the segments.
182 * @return URI path component
183 *
184 * @since 4.5
185 */
186 public static String formatSegments(final String... segments) {
187 return formatSegments(Arrays.asList(segments), StandardCharsets.UTF_8);
188 }
189
190 /**
191 * Returns a String that is suitable for use as an {@code application/x-www-form-urlencoded}
192 * list of parameters in an HTTP PUT or HTTP POST.
193 *
194 * @param parameters The parameters to include.
195 * @param parameterSeparator The parameter separator, by convention, {@code '&'} or {@code ';'}.
196 * @param charset The encoding to use.
197 * @return An {@code application/x-www-form-urlencoded} string
198 *
199 * @since 4.3
200 */
201 public static String format(
202 final Iterable<? extends NameValuePair> parameters,
203 final char parameterSeparator,
204 final Charset charset) {
205 Args.notNull(parameters, "Parameters");
206 final StringBuilder buf = new StringBuilder();
207 int i = 0;
208 for (final NameValuePair parameter : parameters) {
209 if (i > 0) {
210 buf.append(parameterSeparator);
211 }
212 PercentCodec.encode(buf, parameter.getName(), charset, URLENCODER, true);
213 if (parameter.getValue() != null) {
214 buf.append('=');
215 PercentCodec.encode(buf, parameter.getValue(), charset, URLENCODER, true);
216 }
217 i++;
218 }
219 return buf.toString();
220 }
221
222 /**
223 * Returns a String that is suitable for use as an {@code application/x-www-form-urlencoded}
224 * list of parameters in an HTTP PUT or HTTP POST.
225 *
226 * @param parameters The parameters to include.
227 * @param charset The encoding to use.
228 * @return An {@code application/x-www-form-urlencoded} string
229 *
230 * @since 4.2
231 */
232 public static String format(
233 final Iterable<? extends NameValuePair> parameters,
234 final Charset charset) {
235 return format(parameters, QP_SEP_A, charset);
236 }
237
238 private static final BitSet URLENCODER = new BitSet(256);
239
240 static {
241 // unreserved chars
242 // alpha characters
243 for (int i = 'a'; i <= 'z'; i++) {
244 URLENCODER.set(i);
245 }
246 for (int i = 'A'; i <= 'Z'; i++) {
247 URLENCODER.set(i);
248 }
249 // numeric characters
250 for (int i = '0'; i <= '9'; i++) {
251 URLENCODER.set(i);
252 }
253 URLENCODER.set('_'); // these are the characters of the "mark" list
254 URLENCODER.set('-');
255 URLENCODER.set('.');
256 URLENCODER.set('*');
257 }
258
259 }