1 package org.apache.lucene.analysis;
4 * Licensed to the Apache Software Foundation (ASF) under one or more
5 * contributor license agreements. See the NOTICE file distributed with
6 * this work for additional information regarding copyright ownership.
7 * The ASF licenses this file to You under the Apache License, Version 2.0
8 * (the "License"); you may not use this file except in compliance with
9 * the License. You may obtain a copy of the License at
11 * http://www.apache.org/licenses/LICENSE-2.0
13 * Unless required by applicable law or agreed to in writing, software
14 * distributed under the License is distributed on an "AS IS" BASIS,
15 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 * See the License for the specific language governing permissions and
17 * limitations under the License.
20 import java.util.AbstractSet;
21 import java.util.Collection;
22 import java.util.Iterator;
25 import org.apache.lucene.util.Version;
28 * A simple class that stores Strings as char[]'s in a
29 * hash table. Note that this is not a general purpose
30 * class. For example, it cannot remove items from the
31 * set, nor does it resize its hash table to be smaller,
32 * etc. It is designed to be quick to test if a char[]
33 * is in the set without the necessity of converting it
35 * <p>You must specify the required {@link Version}
36 * compatibility when creating {@link CharArraySet}:
38 * <li> As of 3.1, supplementary characters are
39 * properly lowercased.</li>
41 * Before 3.1 supplementary characters could not be
42 * lowercased correctly due to the lack of Unicode 4
43 * support in JDK 1.4. To use instances of
44 * {@link CharArraySet} with the behavior before Lucene
45 * 3.1 pass a {@link Version} < 3.1 to the constructors.
47 * <em>Please note:</em> This class implements {@link java.util.Set Set} but
48 * does not behave like it should in all cases. The generic type is
49 * {@code Set<Object>}, because you can add any object to it,
50 * that has a string representation. The add methods will use
51 * {@link Object#toString} and store the result using a {@code char[]}
52 * buffer. The same behavior have the {@code contains()} methods.
53 * The {@link #iterator()} returns an {@code Iterator<String>}.
54 * For type safety also {@link #stringIterator()} is provided.
56 public class CharArraySet extends AbstractSet<Object> {
57 public static final CharArraySet EMPTY_SET = new CharArraySet(CharArrayMap.<Object>emptyMap());
58 private static final Object PLACEHOLDER = new Object();
60 private final CharArrayMap<Object> map;
63 * Create set with enough capacity to hold startSize terms
66 * compatibility match version see <a href="#version">Version
67 * note</a> above for details.
69 * the initial capacity
71 * <code>false</code> if and only if the set should be case sensitive
72 * otherwise <code>true</code>.
74 public CharArraySet(Version matchVersion, int startSize, boolean ignoreCase) {
75 this(new CharArrayMap<Object>(matchVersion, startSize, ignoreCase));
79 * Creates a set from a Collection of objects.
82 * compatibility match version see <a href="#version">Version
83 * note</a> above for details.
85 * a collection whose elements to be placed into the set
87 * <code>false</code> if and only if the set should be case sensitive
88 * otherwise <code>true</code>.
90 public CharArraySet(Version matchVersion, Collection<?> c, boolean ignoreCase) {
91 this(matchVersion, c.size(), ignoreCase);
96 * Creates a set with enough capacity to hold startSize terms
99 * the initial capacity
101 * <code>false</code> if and only if the set should be case sensitive
102 * otherwise <code>true</code>.
103 * @deprecated use {@link #CharArraySet(Version, int, boolean)} instead
106 public CharArraySet(int startSize, boolean ignoreCase) {
107 this(Version.LUCENE_30, startSize, ignoreCase);
111 * Creates a set from a Collection of objects.
114 * a collection whose elements to be placed into the set
116 * <code>false</code> if and only if the set should be case sensitive
117 * otherwise <code>true</code>.
118 * @deprecated use {@link #CharArraySet(Version, Collection, boolean)} instead
121 public CharArraySet(Collection<?> c, boolean ignoreCase) {
122 this(Version.LUCENE_30, c.size(), ignoreCase);
126 /** Create set from the specified map (internal only), used also by {@link CharArrayMap#keySet()} */
127 CharArraySet(final CharArrayMap<Object> map){
131 /** Clears all entries in this set. This method is supported for reusing, but not {@link Set#remove}. */
133 public void clear() {
137 /** true if the <code>len</code> chars of <code>text</code> starting at <code>off</code>
139 public boolean contains(char[] text, int off, int len) {
140 return map.containsKey(text, off, len);
143 /** true if the <code>CharSequence</code> is in the set */
144 public boolean contains(CharSequence cs) {
145 return map.containsKey(cs);
149 public boolean contains(Object o) {
150 return map.containsKey(o);
154 public boolean add(Object o) {
155 return map.put(o, PLACEHOLDER) == null;
158 /** Add this CharSequence into the set */
159 public boolean add(CharSequence text) {
160 return map.put(text, PLACEHOLDER) == null;
163 /** Add this String into the set */
164 public boolean add(String text) {
165 return map.put(text, PLACEHOLDER) == null;
168 /** Add this char[] directly to the set.
169 * If ignoreCase is true for this Set, the text array will be directly modified.
170 * The user should never modify this text array after calling this method.
172 public boolean add(char[] text) {
173 return map.put(text, PLACEHOLDER) == null;
182 * Returns an unmodifiable {@link CharArraySet}. This allows to provide
183 * unmodifiable views of internal sets for "read-only" use.
186 * a set for which the unmodifiable set is returned.
187 * @return an new unmodifiable {@link CharArraySet}.
188 * @throws NullPointerException
189 * if the given set is <code>null</code>.
191 public static CharArraySet unmodifiableSet(CharArraySet set) {
193 throw new NullPointerException("Given set is null");
194 if (set == EMPTY_SET)
196 if (set.map instanceof CharArrayMap.UnmodifiableCharArrayMap)
198 return new CharArraySet(CharArrayMap.unmodifiableMap(set.map));
202 * Returns a copy of the given set as a {@link CharArraySet}. If the given set
203 * is a {@link CharArraySet} the ignoreCase property will be preserved.
207 * @return a copy of the given set as a {@link CharArraySet}. If the given set
208 * is a {@link CharArraySet} the ignoreCase and matchVersion property will be
210 * @deprecated use {@link #copy(Version, Set)} instead.
213 public static CharArraySet copy(final Set<?> set) {
216 return copy(Version.LUCENE_30, set);
220 * Returns a copy of the given set as a {@link CharArraySet}. If the given set
221 * is a {@link CharArraySet} the ignoreCase property will be preserved.
223 * <b>Note:</b> If you intend to create a copy of another {@link CharArraySet} where
224 * the {@link Version} of the source set differs from its copy
225 * {@link #CharArraySet(Version, Collection, boolean)} should be used instead.
226 * The {@link #copy(Version, Set)} will preserve the {@link Version} of the
227 * source set it is an instance of {@link CharArraySet}.
230 * @param matchVersion
231 * compatibility match version see <a href="#version">Version
232 * note</a> above for details. This argument will be ignored if the
233 * given set is a {@link CharArraySet}.
236 * @return a copy of the given set as a {@link CharArraySet}. If the given set
237 * is a {@link CharArraySet} the ignoreCase property as well as the
238 * matchVersion will be of the given set will be preserved.
240 public static CharArraySet copy(final Version matchVersion, final Set<?> set) {
243 if(set instanceof CharArraySet) {
244 final CharArraySet source = (CharArraySet) set;
245 return new CharArraySet(CharArrayMap.copy(source.map.matchVersion, source.map));
247 return new CharArraySet(matchVersion, set, false);
250 /** The Iterator<String> for this set. Strings are constructed on the fly, so
251 * use <code>nextCharArray</code> for more efficient access.
252 * @deprecated Use the standard iterator, which returns {@code char[]} instances.
255 public class CharArraySetIterator implements Iterator<String> {
258 private CharArraySetIterator() {
262 private void goNext() {
265 while (pos < map.keys.length && (next=map.keys[pos]) == null) pos++;
268 public boolean hasNext() {
272 /** do not modify the returned char[] */
273 public char[] nextCharArray() {
279 /** Returns the next String, as a Set<String> would...
280 * use nextCharArray() for better efficiency. */
281 public String next() {
282 return new String(nextCharArray());
285 public void remove() {
286 throw new UnsupportedOperationException();
290 /** returns an iterator of new allocated Strings (an instance of {@link CharArraySetIterator}).
291 * @deprecated Use {@link #iterator}, which returns {@code char[]} instances.
294 public Iterator<String> stringIterator() {
295 return new CharArraySetIterator();
298 /** Returns an {@link Iterator} depending on the version used:
300 * <li>if {@code matchVersion} ≥ 3.1, it returns {@code char[]} instances in this set.</li>
301 * <li>if {@code matchVersion} is 3.0 or older, it returns new
302 * allocated Strings, so this method violates the Set interface.
303 * It is kept this way for backwards compatibility, normally it should
304 * return {@code char[]} on {@code next()}</li>
307 @Override @SuppressWarnings("unchecked")
308 public Iterator<Object> iterator() {
309 // use the AbstractSet#keySet()'s iterator (to not produce endless recursion)
310 return map.matchVersion.onOrAfter(Version.LUCENE_31) ?
311 map.originalKeySet().iterator() : (Iterator) stringIterator();
315 public String toString() {
316 final StringBuilder sb = new StringBuilder("[");
317 for (Object item : this) {
318 if (sb.length()>1) sb.append(", ");
319 if (item instanceof char[]) {
320 sb.append((char[]) item);
325 return sb.append(']').toString();