View Javadoc
1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements.  See the NOTICE file distributed with
4    * this work for additional information regarding copyright ownership.
5    * The ASF licenses this file to You under the Apache License, Version 2.0
6    * (the "License"); you may not use this file except in compliance with
7    * the License.  You may obtain a copy of the License at
8    * 
9    *      http://www.apache.org/licenses/LICENSE-2.0
10   * 
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS,
13   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14   * See the License for the specific language governing permissions and
15   * limitations under the License.
16   */
17  package org.apache.commons.text.translate;
18  
19  import java.io.IOException;
20  import java.io.Writer;
21  import java.util.Arrays;
22  import java.util.EnumSet;
23  
24  /**
25   * Translate XML numeric entities of the form &#[xX]?\d+;? to 
26   * the specific codepoint.
27   *
28   * Note that the semi-colon is optional.
29   *
30   * @since 1.0
31   */
32  public class NumericEntityUnescaper extends CharSequenceTranslator {
33  
34      public static enum OPTION { semiColonRequired, semiColonOptional, errorIfNoSemiColon }
35  
36      // TODO?: Create an OptionsSet class to hide some of the conditional logic below
37      private final EnumSet<OPTION> options;
38  
39      /**
40       * Create a UnicodeUnescaper.
41       *
42       * The constructor takes a list of options, only one type of which is currently 
43       * available (whether to allow, error or ignore the semi-colon on the end of a 
44       * numeric entity to being missing).
45       *
46       * For example, to support numeric entities without a ';':
47       *    new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.semiColonOptional)
48       * and to throw an IllegalArgumentException when they're missing:
49       *    new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.errorIfNoSemiColon)
50       *
51       * Note that the default behaviour is to ignore them. 
52       *
53       * @param options to apply to this unescaper
54       */
55      public NumericEntityUnescaper(final OPTION... options) {
56          if(options.length > 0) {
57              this.options = EnumSet.copyOf(Arrays.asList(options));
58          } else {
59              this.options = EnumSet.copyOf(Arrays.asList(new OPTION[] { OPTION.semiColonRequired }));
60          }
61      }
62  
63      /**
64       * Whether the passed in option is currently set.
65       *
66       * @param option to check state of
67       * @return whether the option is set
68       */
69      public boolean isSet(final OPTION option) { 
70          return options == null ? false : options.contains(option);
71      }
72  
73      /**
74       * {@inheritDoc}
75       */
76      @Override
77      public int translate(final CharSequence input, final int index, final Writer out) throws IOException {
78          final int seqEnd = input.length();
79          // Uses -2 to ensure there is something after the &#
80          if(input.charAt(index) == '&' && index < seqEnd - 2 && input.charAt(index + 1) == '#') {
81              int start = index + 2;
82              boolean isHex = false;
83  
84              final char firstChar = input.charAt(start);
85              if(firstChar == 'x' || firstChar == 'X') {
86                  start++;
87                  isHex = true;
88  
89                  // Check there's more than just an x after the &#
90                  if(start == seqEnd) {
91                      return 0;
92                  }
93              }
94  
95              int end = start;
96              // Note that this supports character codes without a ; on the end
97              while(end < seqEnd && ( input.charAt(end) >= '0' && input.charAt(end) <= '9' ||
98                                      input.charAt(end) >= 'a' && input.charAt(end) <= 'f' ||
99                                      input.charAt(end) >= 'A' && input.charAt(end) <= 'F' ) )
100             {
101                 end++;
102             }
103 
104             final boolean semiNext = end != seqEnd && input.charAt(end) == ';';
105 
106             if(!semiNext) {
107                 if(isSet(OPTION.semiColonRequired)) {
108                     return 0;
109                 } else
110                 if(isSet(OPTION.errorIfNoSemiColon)) {
111                     throw new IllegalArgumentException("Semi-colon required at end of numeric entity");
112                 }
113             }
114 
115             int entityValue;
116             try {
117                 if(isHex) {
118                     entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 16);
119                 } else {
120                     entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 10);
121                 }
122             } catch(final NumberFormatException nfe) {
123                 return 0;
124             }
125 
126             if(entityValue > 0xFFFF) {
127                 final char[] chrs = Character.toChars(entityValue);
128                 out.write(chrs[0]);
129                 out.write(chrs[1]);
130             } else {
131                 out.write(entityValue);
132             }
133 
134             return 2 + end - start + (isHex ? 1 : 0) + (semiNext ? 1 : 0);
135         }
136         return 0;
137     }
138 }