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.configuration2.convert;
018
019import java.util.ArrayList;
020import java.util.Collection;
021import java.util.Collections;
022import java.util.Iterator;
023import java.util.List;
024
025import org.apache.commons.lang3.Strings;
026
027/**
028 * <p>
029 * A specialized implementation of {@code ListDelimiterHandler} which simulates the list delimiter handling as it was
030 * used by {@code PropertiesConfiguration} in Commons Configuration 1.x.
031 * </p>
032 * <p>
033 * This class mainly exists for compatibility reasons. It is intended to be used by applications which have to deal with
034 * properties files created by an older version of this library.
035 * </p>
036 * <p>
037 * In the 1.x series of Commons Configuration list handling was not fully consistent. The escaping of property values
038 * was done in a different way if they contained a list delimiter or not. From version 2.0 on, escaping is more
039 * stringent which might cause slightly different results when parsing properties files created by or for Configuration
040 * 1.x. If you encounter such problems, you can switch to this {@code ListDelimiterHandler} implementation rather than
041 * the default one. In other cases, this class should not be used!
042 * </p>
043 * <p>
044 * Implementation note: An instance of this class can safely be shared between multiple {@code Configuration} instances.
045 * </p>
046 *
047 * @since 2.0
048 */
049public class LegacyListDelimiterHandler extends AbstractListDelimiterHandler {
050
051    /** Constant for the escaping character. */
052    private static final String ESCAPE = "\\";
053
054    /** Constant for the escaped escaping character. */
055    private static final String DOUBLE_ESC = ESCAPE + ESCAPE;
056
057    /** Constant for a duplicated sequence of escaping characters. */
058    private static final String QUAD_ESC = DOUBLE_ESC + DOUBLE_ESC;
059
060    /**
061     * Returns the number of trailing backslashes. This is sometimes needed for the correct handling of escape characters.
062     *
063     * @param line the string to investigate
064     * @return the number of trailing backslashes
065     */
066    private static int countTrailingBS(final String line) {
067        int bsCount = 0;
068        for (int idx = line.length() - 1; idx >= 0 && line.charAt(idx) == '\\'; idx--) {
069            bsCount++;
070        }
071
072        return bsCount;
073    }
074
075    /** The list delimiter character. */
076    private final char delimiter;
077
078    /**
079     * Creates a new instance of {@code LegacyListDelimiterHandler} and sets the list delimiter character.
080     *
081     * @param listDelimiter the list delimiter character
082     */
083    public LegacyListDelimiterHandler(final char listDelimiter) {
084        delimiter = listDelimiter;
085    }
086
087    /**
088     * {@inheritDoc} This implementation performs delimiter escaping for a single value (which is not part of a list).
089     */
090    @Override
091    public Object escape(final Object value, final ValueTransformer transformer) {
092        return escapeValue(value, false, transformer);
093    }
094
095    /**
096     * Performs the escaping of backslashes in the specified properties value. Because a double backslash is used to escape
097     * the escape character of a list delimiter, double backslashes also have to be escaped if the property is part of a
098     * (single line) list. In addition, because the output is written into a properties file, each occurrence of a backslash
099     * again has to be doubled. This method is called by {@code escapeValue()}.
100     *
101     * @param value the value to be escaped
102     * @param inList a flag whether the value is part of a list
103     * @return the value with escaped backslashes as string
104     */
105    protected String escapeBackslashs(final Object value, final boolean inList) {
106        String strValue = String.valueOf(value);
107
108        if (inList && strValue.contains(DOUBLE_ESC)) {
109            strValue = Strings.CS.replace(strValue, DOUBLE_ESC, QUAD_ESC);
110        }
111
112        return strValue;
113    }
114
115    /**
116     * {@inheritDoc} This implementation performs a special encoding of backslashes at the end of a string so that they are
117     * not interpreted as escape character for a following list delimiter.
118     */
119    @Override
120    public Object escapeList(final List<?> values, final ValueTransformer transformer) {
121        if (!values.isEmpty()) {
122            final Iterator<?> it = values.iterator();
123            String lastValue = escapeValue(it.next(), true, transformer);
124            final StringBuilder buf = new StringBuilder(lastValue);
125            while (it.hasNext()) {
126                // if the last value ended with an escape character, it has
127                // to be escaped itself; otherwise the list delimiter will
128                // be escaped
129                if (lastValue.endsWith(ESCAPE) && countTrailingBS(lastValue) / 2 % 2 != 0) {
130                    buf.append(ESCAPE).append(ESCAPE);
131                }
132                buf.append(getDelimiter());
133                lastValue = escapeValue(it.next(), true, transformer);
134                buf.append(lastValue);
135            }
136            return buf.toString();
137        }
138        return null;
139    }
140
141    /**
142     * {@inheritDoc} This is just a dummy implementation. It is never called.
143     */
144    @Override
145    protected String escapeString(final String s) {
146        return null;
147    }
148
149    /**
150     * Escapes the given property value. This method is called on saving the configuration for each property value. It
151     * ensures a correct handling of backslash characters and also takes care that list delimiter characters in the value
152     * are escaped.
153     *
154     * @param value the property value
155     * @param inList a flag whether the value is part of a list
156     * @param transformer the {@code ValueTransformer}
157     * @return the escaped property value
158     */
159    protected String escapeValue(final Object value, final boolean inList, final ValueTransformer transformer) {
160        String escapedValue = String.valueOf(transformer.transformValue(escapeBackslashs(value, inList)));
161        if (getDelimiter() != 0) {
162            escapedValue = Strings.CS.replace(escapedValue, String.valueOf(getDelimiter()), ESCAPE + getDelimiter());
163        }
164        return escapedValue;
165    }
166
167    /**
168     * Gets the list delimiter character.
169     *
170     * @return the list delimiter character
171     */
172    public char getDelimiter() {
173        return delimiter;
174    }
175
176    /**
177     * {@inheritDoc} This implementation simulates the old splitting algorithm. The string is split at the delimiter
178     * character if it is not escaped. If the delimiter character is not found, the input is returned unchanged.
179     */
180    @Override
181    protected Collection<String> splitString(final String s, final boolean trim) {
182        if (s.indexOf(getDelimiter()) < 0) {
183            return Collections.singleton(s);
184        }
185
186        final List<String> list = new ArrayList<>();
187
188        StringBuilder token = new StringBuilder();
189        int begin = 0;
190        boolean inEscape = false;
191        final char esc = ESCAPE.charAt(0);
192
193        while (begin < s.length()) {
194            final char c = s.charAt(begin);
195            if (inEscape) {
196                // last character was the escape marker
197                // can current character be escaped?
198                if (c != getDelimiter() && c != esc) {
199                    // no, also add escape character
200                    token.append(esc);
201                }
202                token.append(c);
203                inEscape = false;
204            } else if (c == getDelimiter()) {
205                // found a list delimiter -> add token and
206                // resetDefaultFileSystem buffer
207                String t = token.toString();
208                if (trim) {
209                    t = t.trim();
210                }
211                list.add(t);
212                token = new StringBuilder();
213            } else if (c == esc) {
214                // eventually escape next character
215                inEscape = true;
216            } else {
217                token.append(c);
218            }
219
220            begin++;
221        }
222
223        // Trailing delimiter?
224        if (inEscape) {
225            token.append(esc);
226        }
227        // Add last token
228        String t = token.toString();
229        if (trim) {
230            t = t.trim();
231        }
232        list.add(t);
233
234        return list;
235    }
236}