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     *      http://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     */
017    package org.apache.commons.lang3.text.translate;
018    
019    import java.io.IOException;
020    import java.io.Writer;
021    import java.util.Arrays;
022    import java.util.EnumSet;
023    
024    /**
025     * Translate XML numeric entities of the form &#[xX]?\d+;? to 
026     * the specific codepoint.
027     *
028     * Note that the semi-colon is optional.
029     * 
030     * @since 3.0
031     * @version $Id: NumericEntityUnescaper.java 1199894 2011-11-09 17:53:59Z ggregory $
032     */
033    public class NumericEntityUnescaper extends CharSequenceTranslator {
034    
035        public static enum OPTION { semiColonRequired, semiColonOptional, errorIfNoSemiColon }
036    
037        // TODO?: Create an OptionsSet class to hide some of the conditional logic below
038        private final EnumSet<OPTION> options;
039    
040        /**
041         * Create a UnicodeUnescaper.
042         *
043         * The constructor takes a list of options, only one type of which is currently 
044         * available (whether to allow, error or ignore the semi-colon on the end of a 
045         * numeric entity to being missing).
046         *
047         * For example, to support numeric entities without a ';':
048         *    new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.semiColonOptional)
049         * and to throw an IllegalArgumentException when they're missing:
050         *    new NumericEntityUnescaper(NumericEntityUnescaper.OPTION.errorIfNoSemiColon)
051         *
052         * Note that the default behaviour is to ignore them. 
053         *
054         * @param options to apply to this unescaper
055         */
056        public NumericEntityUnescaper(OPTION... options) {
057            if(options.length > 0) {
058                this.options = EnumSet.copyOf(Arrays.asList(options));
059            } else {
060                this.options = EnumSet.copyOf(Arrays.asList(new OPTION[] { OPTION.semiColonRequired }));
061            }
062        }
063    
064        /**
065         * Whether the passed in option is currently set.
066         *
067         * @param option to check state of
068         * @return whether the option is set
069         */
070        public boolean isSet(OPTION option) { 
071            return options == null ? false : options.contains(option);
072        }
073    
074        /**
075         * {@inheritDoc}
076         */
077        @Override
078        public int translate(CharSequence input, int index, Writer out) throws IOException {
079            int seqEnd = input.length();
080            // Uses -2 to ensure there is something after the &#
081            if(input.charAt(index) == '&' && index < seqEnd - 2 && input.charAt(index + 1) == '#') {
082                int start = index + 2;
083                boolean isHex = false;
084    
085                char firstChar = input.charAt(start);
086                if(firstChar == 'x' || firstChar == 'X') {
087                    start++;
088                    isHex = true;
089    
090                    // Check there's more than just an x after the &#
091                    if(start == seqEnd) {
092                        return 0;
093                    }
094                }
095    
096                int end = start;
097                // Note that this supports character codes without a ; on the end
098                while(end < seqEnd && ( input.charAt(end) >= '0' && input.charAt(end) <= '9' ||
099                                        input.charAt(end) >= 'a' && input.charAt(end) <= 'f' ||
100                                        input.charAt(end) >= 'A' && input.charAt(end) <= 'F' ) )
101                {
102                    end++;
103                }
104    
105                boolean semiNext = end != seqEnd && input.charAt(end) == ';';
106    
107                if(!semiNext) {
108                    if(isSet(OPTION.semiColonRequired)) {
109                        return 0;
110                    } else
111                    if(isSet(OPTION.errorIfNoSemiColon)) {
112                        throw new IllegalArgumentException("Semi-colon required at end of numeric entity");
113                    }
114                }
115    
116                int entityValue;
117                try {
118                    if(isHex) {
119                        entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 16);
120                    } else {
121                        entityValue = Integer.parseInt(input.subSequence(start, end).toString(), 10);
122                    }
123                } catch(NumberFormatException nfe) {
124                    return 0;
125                }
126    
127                if(entityValue > 0xFFFF) {
128                    char[] chrs = Character.toChars(entityValue);
129                    out.write(chrs[0]);
130                    out.write(chrs[1]);
131                } else {
132                    out.write(entityValue);
133                }
134    
135                return 2 + end - start + (isHex ? 1 : 0) + (semiNext ? 1 : 0);
136            }
137            return 0;
138        }
139    }