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 * https://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.lang3.text.translate;
18
19 import java.io.IOException;
20 import java.io.Writer;
21 import java.util.Arrays;
22 import java.util.Collections;
23 import java.util.EnumSet;
24
25 import org.apache.commons.lang3.CharUtils;
26
27 /**
28 * Translate XML numeric entities of the form &#[xX]?\d+;? to the specific code point.
29 *
30 * Note that the semicolon is optional.
31 *
32 * @since 3.0
33 * @deprecated As of <a href="https://commons.apache.org/proper/commons-lang/changes-report.html#a3.6">3.6</a>, use Apache Commons Text
34 * <a href="https://commons.apache.org/proper/commons-text/javadocs/api-release/org/apache/commons/text/translate/NumericEntityUnescaper.html">
35 * NumericEntityUnescaper</a>.
36 */
37 @Deprecated
38 public class NumericEntityUnescaper extends CharSequenceTranslator {
39
40 /**
41 * Enumerates NumericEntityUnescaper options for unescaping.
42 *
43 * @deprecated As of 3.18.0, use Apache Commons Text <a href=
44 * "https://commons.apache.org/proper/commons-text/javadocs/api-release/org/apache/commons/text/translate/NumericEntityUnescaper.OPTION.html">
45 * NumericEntityUnescaper.OPTION</a>.
46 */
47 @Deprecated
48 public enum OPTION {
49
50 /**
51 * Require a semicolon.
52 */
53 semiColonRequired,
54
55 /**
56 * Do not require a semicolon.
57 */
58 semiColonOptional,
59
60 /**
61 * Throw an exception if a semicolon is missing.
62 */
63 errorIfNoSemiColon
64 }
65
66 // TODO?: Create an OptionsSet class to hide some of the conditional logic below
67 private final EnumSet<OPTION> options;
68
69 /**
70 * Create a UnicodeUnescaper.
71 *
72 * The constructor takes a list of options, only one type of which is currently
73 * available (whether to allow, error or ignore the semicolon on the end of a
74 * numeric entity to being missing).
75 *
76 * For example, to support numeric entities without a ';':
77 * new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.semiColonOptional)
78 * and to throw an IllegalArgumentException when they're missing:
79 * new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.errorIfNoSemiColon)
80 *
81 * Note that the default behavior is to ignore them.
82 *
83 * @param options to apply to this unescaper.
84 */
85 public NumericEntityUnescaper(final OPTION... options) {
86 if (options.length > 0) {
87 this.options = EnumSet.copyOf(Arrays.asList(options));
88 } else {
89 this.options = EnumSet.copyOf(Collections.singletonList(OPTION.semiColonRequired));
90 }
91 }
92
93 /**
94 * Tests whether the passed in option is currently set.
95 *
96 * @param option to check state of.
97 * @return whether the option is set.
98 */
99 public boolean isSet(final OPTION option) {
100 return options != null && options.contains(option);
101 }
102
103 /**
104 * {@inheritDoc}
105 */
106 @Override
107 public int translate(final CharSequence input, final int index, final Writer out) throws IOException {
108 final int seqEnd = input.length();
109 // Uses -2 to ensure there is something after the &#
110 if (input.charAt(index) == '&' && index < seqEnd - 2 && input.charAt(index + 1) == '#') {
111 int start = index + 2;
112 boolean isHex = false;
113 final char firstChar = input.charAt(start);
114 if (firstChar == 'x' || firstChar == 'X') {
115 start++;
116 isHex = true;
117 // Check there's more than just an x after the &#
118 if (start == seqEnd) {
119 return 0;
120 }
121 }
122 int end = start;
123 // Note that this supports character codes without a ; on the end.
124 // A decimal entity (&#...) only holds decimal digits; a-f are hex digits and must
125 // not be consumed here, otherwise the following text is swallowed into a failed parse.
126 while (end < seqEnd && (isHex ? CharUtils.isHex(input.charAt(end)) : CharUtils.isAsciiNumeric(input.charAt(end)))) {
127 end++;
128 }
129 final boolean semiNext = end != seqEnd && input.charAt(end) == ';';
130 if (!semiNext) {
131 if (isSet(OPTION.semiColonRequired)) {
132 return 0;
133 }
134 if (isSet(OPTION.errorIfNoSemiColon)) {
135 throw new IllegalArgumentException("Semi-colon required at end of numeric entity");
136 }
137 }
138 final int entityValue;
139 try {
140 if (isHex) {
141 entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 16);
142 } else {
143 entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 10);
144 }
145 } catch (final NumberFormatException nfe) {
146 return 0;
147 }
148 if (entityValue < Character.MIN_CODE_POINT || entityValue > Character.MAX_CODE_POINT) {
149 return 0;
150 }
151 if (entityValue > 0xFFFF) {
152 final char[] chars = Character.toChars(entityValue);
153 out.write(chars[0]);
154 out.write(chars[1]);
155 } else {
156 out.write(entityValue);
157 }
158 return 2 + end - start + (isHex ? 1 : 0) + (semiNext ? 1 : 0);
159 }
160 return 0;
161 }
162 }