View Javadoc
1   /*
2    * Copyright (C) 2009 The Guava Authors
3    *
4    * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
5    * in compliance with the License. You may obtain a copy of the License at
6    *
7    * http://www.apache.org/licenses/LICENSE-2.0
8    *
9    * Unless required by applicable law or agreed to in writing, software distributed under the License
10   * is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
11   * or implied. See the License for the specific language governing permissions and limitations under
12   * the License.
13   */
14  
15  package com.google.common.escape;
16  
17  import static com.google.common.base.Preconditions.checkNotNull;
18  
19  import com.google.common.annotations.Beta;
20  import com.google.common.annotations.GwtCompatible;
21  import java.util.Map;
22  
23  /**
24   * A {@link CharEscaper} that uses an array to quickly look up replacement characters for a given
25   * {@code char} value. An additional safe range is provided that determines whether {@code char}
26   * values without specific replacements are to be considered safe and left unescaped or should be
27   * escaped in a general way.
28   *
29   * <p>A good example of usage of this class is for Java source code escaping where the replacement
30   * array contains information about special ASCII characters such as {@code \\t} and {@code \\n}
31   * while {@link #escapeUnsafe} is overridden to handle general escaping of the form {@code \\uxxxx}.
32   *
33   * <p>The size of the data structure used by {@link ArrayBasedCharEscaper} is proportional to the
34   * highest valued character that requires escaping. For example a replacement map containing the
35   * single character '{@code \}{@code u1000}' will require approximately 16K of memory. If you need
36   * to create multiple escaper instances that have the same character replacement mapping consider
37   * using {@link ArrayBasedEscaperMap}.
38   *
39   * @author Sven Mawson
40   * @author David Beaumont
41   * @since 15.0
42   */
43  @Beta
44  @GwtCompatible
45  public abstract class ArrayBasedCharEscaper extends CharEscaper {
46    // The replacement array (see ArrayBasedEscaperMap).
47    private final char[][] replacements;
48    // The number of elements in the replacement array.
49    private final int replacementsLength;
50    // The first character in the safe range.
51    private final char safeMin;
52    // The last character in the safe range.
53    private final char safeMax;
54  
55    /**
56     * Creates a new ArrayBasedCharEscaper instance with the given replacement map and specified safe
57     * range. If {@code safeMax < safeMin} then no characters are considered safe.
58     *
59     * <p>If a character has no mapped replacement then it is checked against the safe range. If it
60     * lies outside that, then {@link #escapeUnsafe} is called, otherwise no escaping is performed.
61     *
62     * @param replacementMap a map of characters to their escaped representations
63     * @param safeMin the lowest character value in the safe range
64     * @param safeMax the highest character value in the safe range
65     */
66    protected ArrayBasedCharEscaper(
67        Map<Character, String> replacementMap, char safeMin, char safeMax) {
68  
69      this(ArrayBasedEscaperMap.create(replacementMap), safeMin, safeMax);
70    }
71  
72    /**
73     * Creates a new ArrayBasedCharEscaper instance with the given replacement map and specified safe
74     * range. If {@code safeMax < safeMin} then no characters are considered safe. This initializer is
75     * useful when explicit instances of ArrayBasedEscaperMap are used to allow the sharing of large
76     * replacement mappings.
77     *
78     * <p>If a character has no mapped replacement then it is checked against the safe range. If it
79     * lies outside that, then {@link #escapeUnsafe} is called, otherwise no escaping is performed.
80     *
81     * @param escaperMap the mapping of characters to be escaped
82     * @param safeMin the lowest character value in the safe range
83     * @param safeMax the highest character value in the safe range
84     */
85    protected ArrayBasedCharEscaper(ArrayBasedEscaperMap escaperMap, char safeMin, char safeMax) {
86  
87      checkNotNull(escaperMap); // GWT specific check (do not optimize)
88      this.replacements = escaperMap.getReplacementArray();
89      this.replacementsLength = replacements.length;
90      if (safeMax < safeMin) {
91        // If the safe range is empty, set the range limits to opposite extremes
92        // to ensure the first test of either value will (almost certainly) fail.
93        safeMax = Character.MIN_VALUE;
94        safeMin = Character.MAX_VALUE;
95      }
96      this.safeMin = safeMin;
97      this.safeMax = safeMax;
98    }
99  
100   /*
101    * This is overridden to improve performance. Rough benchmarking shows that this almost doubles
102    * the speed when processing strings that do not require any escaping.
103    */
104   @Override
105   public final String escape(String s) {
106     checkNotNull(s); // GWT specific check (do not optimize).
107     for (int i = 0; i < s.length(); i++) {
108       char c = s.charAt(i);
109       if ((c < replacementsLength && replacements[c] != null) || c > safeMax || c < safeMin) {
110         return escapeSlow(s, i);
111       }
112     }
113     return s;
114   }
115 
116   /**
117    * Escapes a single character using the replacement array and safe range values. If the given
118    * character does not have an explicit replacement and lies outside the safe range then
119    * {@link #escapeUnsafe} is called.
120    */
121   @Override
122   protected final char[] escape(char c) {
123     if (c < replacementsLength) {
124       char[] chars = replacements[c];
125       if (chars != null) {
126         return chars;
127       }
128     }
129     if (c >= safeMin && c <= safeMax) {
130       return null;
131     }
132     return escapeUnsafe(c);
133   }
134 
135   /**
136    * Escapes a {@code char} value that has no direct explicit value in the replacement array and
137    * lies outside the stated safe range. Subclasses should override this method to provide
138    * generalized escaping for characters.
139    *
140    * <p>Note that arrays returned by this method must not be modified once they have been returned.
141    * However it is acceptable to return the same array multiple times (even for different input
142    * characters).
143    *
144    * @param c the character to escape
145    * @return the replacement characters, or {@code null} if no escaping was required
146    */
147   // TODO(user,cpovirk): Rename this something better once refactoring done
148   protected abstract char[] escapeUnsafe(char c);
149 }