001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      https://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.text.translate;
018
019import java.io.IOException;
020import java.io.Writer;
021import java.util.Arrays;
022import java.util.Collections;
023import java.util.EnumSet;
024
025import org.apache.commons.lang3.ArrayUtils;
026import org.apache.commons.lang3.CharUtils;
027
028/**
029 * Translates XML numeric entities of the form &#[xX]?\d+;? to
030 * the specific code point.
031 *
032 * Note that the semicolon is optional.
033 *
034 * @since 1.0
035 */
036public class NumericEntityUnescaper extends CharSequenceTranslator {
037
038    /**
039     * Enumerates NumericEntityUnescaper options for unescaping.
040     */
041    public enum OPTION {
042
043        /**
044         * Requires a semicolon.
045         */
046        semiColonRequired,
047
048        /**
049         * Does not require a semicolon.
050         */
051        semiColonOptional,
052
053        /**
054         * Throws an exception if a semicolon is missing.
055         */
056        errorIfNoSemiColon
057    }
058
059    /** Default options. */
060    private static final EnumSet<OPTION> DEFAULT_OPTIONS = EnumSet
061        .copyOf(Collections.singletonList(OPTION.semiColonRequired));
062
063    /** EnumSet of OPTIONS, given from the constructor, read-only. */
064    private final EnumSet<OPTION> options;
065
066    /**
067     * Constructs a new instance.
068     *
069     * The constructor takes a list of options, only one type of which is currently available (whether to allow, error or ignore the semicolon on the end of a
070     * numeric entity to being missing).
071     * <p>
072     * For example, to support numeric entities without a ';':
073     * </p>
074     *
075     * <pre>
076     * new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.semiColonOptional)
077     * </pre>
078     * <p>
079     * and to throw an IllegalArgumentException when they're missing:
080     * </p>
081     *
082     * <pre>
083     * new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.errorIfNoSemiColon)
084     * </pre>
085     * <p>
086     * Note that the default behavior is to ignore them.
087     * </p>
088     *
089     * @param options to apply to this unescaper
090     */
091    public NumericEntityUnescaper(final OPTION... options) {
092        this.options = ArrayUtils.isEmpty(options) ? DEFAULT_OPTIONS : EnumSet.copyOf(Arrays.asList(options));
093    }
094
095    /**
096     * Tests whether the passed in option is currently set.
097     *
098     * @param option to check state of.
099     * @return whether the option is set.
100     */
101    public boolean isSet(final OPTION option) {
102        return options.contains(option);
103    }
104
105    /**
106     * {@inheritDoc}
107     */
108    @Override
109    public int translate(final CharSequence input, final int index, final Writer writer) throws IOException {
110        final int seqEnd = input.length();
111        // Uses -2 to ensure there is something after the &#
112        if (input.charAt(index) == '&' && index < seqEnd - 2 && input.charAt(index + 1) == '#') {
113            int start = index + 2;
114            boolean isHex = false;
115            final char firstChar = input.charAt(start);
116            if (firstChar == 'x' || firstChar == 'X') {
117                start++;
118                isHex = true;
119                // Check there's more than just an x after the &#
120                if (start == seqEnd) {
121                    return 0;
122                }
123            }
124            int end = start;
125            // Note that this supports character codes without a ; on the end
126            while (end < seqEnd && CharUtils.isHex(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 (!Character.isValidCodePoint(entityValue)) {
149                return 0;
150            }
151            if (entityValue > 0xFFFF) {
152                final char[] chars = Character.toChars(entityValue);
153                writer.write(chars[0]);
154                writer.write(chars[1]);
155            } else {
156                writer.write(entityValue);
157            }
158            return 2 + end - start + (isHex ? 1 : 0) + (semiNext ? 1 : 0);
159        }
160        return 0;
161    }
162}