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 */ 017package org.apache.commons.configuration2.convert; 018 019import java.util.Collection; 020import java.util.LinkedList; 021import java.util.List; 022 023import org.apache.commons.lang3.StringUtils; 024 025/** 026 * <p> 027 * The default implementation of the {@code ListDelimiterHandler} interface. 028 * </p> 029 * <p> 030 * This class supports list splitting and delimiter escaping using a delimiter 031 * character that can be specified when constructing an instance. Splitting of 032 * strings works by scanning the input for the list delimiter character. The 033 * list delimiter character can be escaped by a backslash. So, provided that a 034 * comma is configured as list delimiter, in the example {@code val1,val2,val3} 035 * three values are recognized. In {@code 3\,1415} the list delimiter is escaped 036 * so that only a single element is detected. (Note that when writing these 037 * examples in Java code, each backslash has to be doubled. This is also true 038 * for all other examples in this documentation.) 039 * </p> 040 * <p> 041 * Because the backslash has a special meaning as escaping character it is 042 * always treated in a special way. If it occurs as a normal character in a 043 * property value, it has to be escaped using another backslash (similar to the 044 * rules of the Java programming language). The following example shows the 045 * correct way to define windows network shares: {@code \\\\Server\\path}. Note 046 * that each backslash is doubled. When combining the list delimiter with 047 * backslashes the same escaping rules apply. For instance, in 048 * {@code C:\\Temp\\,D:\\data\\} the list delimiter is recognized; it is not 049 * escaped by the preceding backslash because this backslash is itself escaped. 050 * In contrast, {@code C:\\Temp\\\,D:\\data\\} defines a single element with a 051 * comma being part of the value; two backslashes after {@code Temp} result in a 052 * single one, the third backslash escapes the list delimiter. 053 * </p> 054 * <p> 055 * As can be seen, there are some constellations which are a bit tricky and 056 * cause a larger number of backslashes in sequence. Nevertheless, the escaping 057 * rules are consistent and do not cause ambiguous results. 058 * </p> 059 * <p> 060 * Implementation node: An instance of this class can safely be shared between 061 * multiple {@code Configuration} instances. 062 * </p> 063 * 064 * @since 2.0 065 */ 066public class DefaultListDelimiterHandler extends AbstractListDelimiterHandler 067{ 068 /** Constant for the escape character. */ 069 private static final char ESCAPE = '\\'; 070 071 /** 072 * Constant for a buffer size for escaping strings. When a character is 073 * escaped the string becomes longer. Therefore, the output buffer is longer 074 * than the original string length. But we assume, that there are not too 075 * many characters that need to be escaped. 076 */ 077 private static final int BUF_SIZE = 16; 078 079 /** Stores the list delimiter character. */ 080 private final char delimiter; 081 082 /** 083 * Creates a new instance of {@code DefaultListDelimiterHandler} and sets 084 * the list delimiter character. 085 * 086 * @param listDelimiter the list delimiter character 087 */ 088 public DefaultListDelimiterHandler(final char listDelimiter) 089 { 090 delimiter = listDelimiter; 091 } 092 093 /** 094 * Returns the list delimiter character used by this instance. 095 * 096 * @return the list delimiter character 097 */ 098 public char getDelimiter() 099 { 100 return delimiter; 101 } 102 103 @Override 104 public Object escapeList(final List<?> values, final ValueTransformer transformer) 105 { 106 final Object[] escapedValues = new String[values.size()]; 107 int idx = 0; 108 for (final Object v : values) 109 { 110 escapedValues[idx++] = escape(v, transformer); 111 } 112 return StringUtils.join(escapedValues, getDelimiter()); 113 } 114 115 @Override 116 protected String escapeString(final String s) 117 { 118 final StringBuilder buf = new StringBuilder(s.length() + BUF_SIZE); 119 for (int i = 0; i < s.length(); i++) 120 { 121 final char c = s.charAt(i); 122 if (c == getDelimiter() || c == ESCAPE) 123 { 124 buf.append(ESCAPE); 125 } 126 buf.append(c); 127 } 128 return buf.toString(); 129 } 130 131 /** 132 * {@inheritDoc} This implementation reverses the escaping done by the 133 * {@code escape()} methods of this class. However, it tries to be tolerant 134 * with unexpected escaping sequences: If after the escape character "\" no 135 * allowed character follows, both the backslash and the following character 136 * are output. 137 */ 138 @Override 139 protected Collection<String> splitString(final String s, final boolean trim) 140 { 141 final List<String> list = new LinkedList<>(); 142 StringBuilder token = new StringBuilder(); 143 boolean inEscape = false; 144 145 for (int i = 0; i < s.length(); i++) 146 { 147 final char c = s.charAt(i); 148 if (inEscape) 149 { 150 // last character was the escape marker 151 // can current character be escaped? 152 if (c != getDelimiter() && c != ESCAPE) 153 { 154 // no, also add escape character 155 token.append(ESCAPE); 156 } 157 token.append(c); 158 inEscape = false; 159 } 160 161 else 162 { 163 if (c == getDelimiter()) 164 { 165 // found a list delimiter -> add token and 166 // reset buffer 167 String t = token.toString(); 168 if (trim) 169 { 170 t = t.trim(); 171 } 172 list.add(t); 173 token = new StringBuilder(); 174 } 175 else if (c == ESCAPE) 176 { 177 // potentially escape next character 178 inEscape = true; 179 } 180 else 181 { 182 token.append(c); 183 } 184 } 185 } 186 187 // Trailing delimiter? 188 if (inEscape) 189 { 190 token.append(ESCAPE); 191 } 192 // Add last token 193 String t = token.toString(); 194 if (trim) 195 { 196 t = t.trim(); 197 } 198 list.add(t); 199 200 return list; 201 } 202}