1 /*
2 * $Header$
3 * $Revision$
4 * $Date$
5 *
6 * ====================================================================
7 *
8 * Copyright 2000-2002 bob mcwhirter & James Strachan.
9 * All rights reserved.
10 *
11 *
12 * Redistribution and use in source and binary forms, with or without
13 * modification, are permitted provided that the following conditions are
14 * met:
15 *
16 * * Redistributions of source code must retain the above copyright
17 * notice, this list of conditions and the following disclaimer.
18 *
19 * * Redistributions in binary form must reproduce the above copyright
20 * notice, this list of conditions and the following disclaimer in the
21 * documentation and/or other materials provided with the distribution.
22 *
23 * * Neither the name of the Jaxen Project nor the names of its
24 * contributors may be used to endorse or promote products derived
25 * from this software without specific prior written permission.
26 *
27 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
28 * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
29 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
30 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
31 * OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
32 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
33 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
34 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
35 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
36 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
37 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
38 *
39 * ====================================================================
40 * This software consists of voluntary contributions made by many
41 * individuals on behalf of the Jaxen Project and was originally
42 * created by bob mcwhirter <bob@werken.com> and
43 * James Strachan <jstrachan@apache.org>. For more information on the
44 * Jaxen Project, please see <http://www.jaxen.org/>.
45 *
46 */
47 package org.jaxen.function;
48
49 import java.util.List;
50
51 import org.jaxen.Context;
52 import org.jaxen.Function;
53 import org.jaxen.FunctionCallException;
54 import org.jaxen.Navigator;
55 /**
56 * <p>
57 * <b>4.2</b>
58 * <code><i>string</i> substring(<i>string</i>,<i>number</i>,<i>number?</i>)</code>
59 * </p>
60 *
61 * <blockquote cite="http://www.w3.org/TR/xpath">
62 * <p>The <b>substring</b> function returns the
63 * substring of the first argument starting at the position specified in
64 * the second argument with length specified in the third argument. For
65 * example,
66 *
67 * <code>substring("12345",2,3)</code> returns <code>"234"</code>.
68 * If the third argument is not specified, it returns the substring
69 * starting at the position specified in the second argument and
70 * continuing to the end of the string. For example,
71 * <code>substring("12345",2)</code> returns <code>"2345"</code>.
72 * </p>
73 *
74 * <p>
75 * More precisely, each character in the string (see <a
76 * href="https://www.w3.org/TR/xpath#strings">[<b>3.6 Strings</b>]</a>) is considered to have a
77 * numeric position: the position of the first character is 1, the
78 * position of the second character is 2 and so on.
79 * </p>
80 *
81 * <blockquote> <b>NOTE: </b>This differs from Java and ECMAScript, in
82 * which the <code>String.substring</code> method treats the position
83 * of the first character as 0.</blockquote>
84 *
85 * <p>
86 * The returned substring contains those characters for which the
87 * position of the character is greater than or equal to the rounded
88 * value of the second argument and, if the third argument is specified,
89 * less than the sum of the rounded value of the second argument and the
90 * rounded value of the third argument; the comparisons and addition
91 * used for the above follow the standard IEEE 754 rules; rounding is
92 * done as if by a call to the <b><a href="#function-round">round</a></b>
93 * function. The following examples illustrate various unusual cases:
94 * </p>
95 *
96 * <ul>
97 *
98 * <li>
99 * <p>
100 * <code>substring("12345", 1.5, 2.6)</code> returns
101 * <code>"234"</code>
102 * </p>
103 * </li>
104 *
105 * <li>
106 * <p>
107 * <code>substring("12345", 0, 3)</code> returns <code>"12"</code>
108 *
109 * </p>
110 * </li>
111 *
112 * <li>
113 * <p>
114 * <code>substring("12345", 0 div 0, 3)</code> returns <code>""</code>
115 * </p>
116 * </li>
117 *
118 * <li>
119 * <p>.
120 * <code>substring("12345", 1, 0 div 0)</code> returns
121 *
122 * <code>""</code>
123 * </p>
124 * </li>
125 *
126 * <li>
127 * <p>
128 * <code>substring("12345", -42, 1 div 0)</code> returns
129 * <code>"12345"</code>
130 * </p>
131 * </li>
132 *
133 * </ul>
134 *
135 * <code>substring("12345", -1 div 0, 1 div 0)</code> returns
136 * <code>""</code> </blockquote>
137 *
138 * @author bob mcwhirter (bob @ werken.com)
139 *
140 * @see <a href="https://www.w3.org/TR/xpath#function-substring"
141 * target="_top">Section 4.2 of the XPath Specification</a>
142 */
143 public class SubstringFunction implements Function
144 {
145
146 /**
147 * Create a new <code>SubstringFunction</code> object.
148 */
149 public SubstringFunction() {}
150
151
152 /** Returns a substring of an XPath string-value by character index.
153 *
154 * @param context the context at the point in the
155 * expression when the function is called
156 * @param args a list that contains two or three items
157 *
158 * @return a <code>String</code> containing the specifed character subsequence of
159 * the original string or the string-value of the context node
160 *
161 * @throws FunctionCallException if <code>args</code> has more than three
162 * or less than two items
163 */
164 public Object call(Context context,
165 List args) throws FunctionCallException
166 {
167 final int argc = args.size();
168 if (argc < 2 || argc > 3){
169 throw new FunctionCallException( "substring() requires two or three arguments." );
170 }
171
172 final Navigator nav = context.getNavigator();
173
174 final String str = StringFunction.evaluate(args.get(0), nav );
175 // The spec doesn't really address this case
176 if (str == null) {
177 return "";
178 }
179
180 final int stringLength = (StringLengthFunction.evaluate(args.get(0), nav )).intValue();
181
182 if (stringLength == 0) {
183 return "";
184 }
185
186 Double d1 = NumberFunction.evaluate(args.get(1), nav);
187
188 if (d1.isNaN()){
189 return "";
190 }
191 // Round the value and subtract 1 as Java strings are zero based
192 int start = RoundFunction.evaluate(d1, nav).intValue() - 1;
193
194 int substringLength = stringLength;
195 if (argc == 3){
196 Double d2 = NumberFunction.evaluate(args.get(2), nav);
197
198 if (!d2.isNaN()){
199 substringLength = RoundFunction.evaluate(d2, nav ).intValue();
200 }
201 else {
202 substringLength = 0;
203 }
204 }
205
206 if (substringLength < 0) return "";
207
208 int end = start + substringLength;
209 if (argc == 2) end = stringLength;
210
211 // negative start is treated as 0
212 if ( start < 0){
213 start = 0;
214 }
215 else if (start > stringLength){
216 return "";
217 }
218
219 if (end > stringLength){
220 end = stringLength;
221 }
222 else if (end < start) return "";
223
224 if (stringLength == str.length()) {
225 // easy case; no surrogate pairs
226 return str.substring(start, end);
227 }
228 else {
229 return unicodeSubstring(str, start, end);
230 }
231
232 }
233
234 private static String unicodeSubstring(String s, int start, int end) {
235
236 StringBuffer result = new StringBuffer(s.length());
237 for (int jChar = 0, uChar=0; uChar < end; jChar++, uChar++) {
238 char c = s.charAt(jChar);
239 if (uChar >= start) result.append(c);
240 if (c >= 0xD800) { // get the low surrogate
241 // ???? we could check here that this is indeed a low surroagte
242 // we could also catch StringIndexOutOfBoundsException
243 jChar++;
244 if (uChar >= start) result.append(s.charAt(jChar));
245 }
246 }
247 return result.toString();
248 }
249 }