001 /*
002 * Copyright 2000-2006 Sun Microsystems, Inc. All Rights Reserved.
003 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
004 *
005 * This code is free software; you can redistribute it and/or modify it
006 * under the terms of the GNU General Public License version 2 only, as
007 * published by the Free Software Foundation. Sun designates this
008 * particular file as subject to the "Classpath" exception as provided
009 * by Sun in the LICENSE file that accompanied this code.
010 *
011 * This code is distributed in the hope that it will be useful, but WITHOUT
012 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
013 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
014 * version 2 for more details (a copy is included in the LICENSE file that
015 * accompanied this code).
016 *
017 * You should have received a copy of the GNU General Public License version
018 * 2 along with this work; if not, write to the Free Software Foundation,
019 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
020 *
021 * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
022 * CA 95054 USA or visit www.sun.com if you need additional information or
023 * have any questions.
024 */
025
026 package java.util;
027
028 import java.io.*;
029
030 /**
031 * <p>Hash table and linked list implementation of the <tt>Map</tt> interface,
032 * with predictable iteration order. This implementation differs from
033 * <tt>HashMap</tt> in that it maintains a doubly-linked list running through
034 * all of its entries. This linked list defines the iteration ordering,
035 * which is normally the order in which keys were inserted into the map
036 * (<i>insertion-order</i>). Note that insertion order is not affected
037 * if a key is <i>re-inserted</i> into the map. (A key <tt>k</tt> is
038 * reinserted into a map <tt>m</tt> if <tt>m.put(k, v)</tt> is invoked when
039 * <tt>m.containsKey(k)</tt> would return <tt>true</tt> immediately prior to
040 * the invocation.)
041 *
042 * <p>This implementation spares its clients from the unspecified, generally
043 * chaotic ordering provided by {@link HashMap} (and {@link Hashtable}),
044 * without incurring the increased cost associated with {@link TreeMap}. It
045 * can be used to produce a copy of a map that has the same order as the
046 * original, regardless of the original map's implementation:
047 * <pre>
048 * void foo(Map m) {
049 * Map copy = new LinkedHashMap(m);
050 * ...
051 * }
052 * </pre>
053 * This technique is particularly useful if a module takes a map on input,
054 * copies it, and later returns results whose order is determined by that of
055 * the copy. (Clients generally appreciate having things returned in the same
056 * order they were presented.)
057 *
058 * <p>A special {@link #LinkedHashMap(int,float,boolean) constructor} is
059 * provided to create a linked hash map whose order of iteration is the order
060 * in which its entries were last accessed, from least-recently accessed to
061 * most-recently (<i>access-order</i>). This kind of map is well-suited to
062 * building LRU caches. Invoking the <tt>put</tt> or <tt>get</tt> method
063 * results in an access to the corresponding entry (assuming it exists after
064 * the invocation completes). The <tt>putAll</tt> method generates one entry
065 * access for each mapping in the specified map, in the order that key-value
066 * mappings are provided by the specified map's entry set iterator. <i>No
067 * other methods generate entry accesses.</i> In particular, operations on
068 * collection-views do <i>not</i> affect the order of iteration of the backing
069 * map.
070 *
071 * <p>The {@link #removeEldestEntry(Map.Entry)} method may be overridden to
072 * impose a policy for removing stale mappings automatically when new mappings
073 * are added to the map.
074 *
075 * <p>This class provides all of the optional <tt>Map</tt> operations, and
076 * permits null elements. Like <tt>HashMap</tt>, it provides constant-time
077 * performance for the basic operations (<tt>add</tt>, <tt>contains</tt> and
078 * <tt>remove</tt>), assuming the hash function disperses elements
079 * properly among the buckets. Performance is likely to be just slightly
080 * below that of <tt>HashMap</tt>, due to the added expense of maintaining the
081 * linked list, with one exception: Iteration over the collection-views
082 * of a <tt>LinkedHashMap</tt> requires time proportional to the <i>size</i>
083 * of the map, regardless of its capacity. Iteration over a <tt>HashMap</tt>
084 * is likely to be more expensive, requiring time proportional to its
085 * <i>capacity</i>.
086 *
087 * <p>A linked hash map has two parameters that affect its performance:
088 * <i>initial capacity</i> and <i>load factor</i>. They are defined precisely
089 * as for <tt>HashMap</tt>. Note, however, that the penalty for choosing an
090 * excessively high value for initial capacity is less severe for this class
091 * than for <tt>HashMap</tt>, as iteration times for this class are unaffected
092 * by capacity.
093 *
094 * <p><strong>Note that this implementation is not synchronized.</strong>
095 * If multiple threads access a linked hash map concurrently, and at least
096 * one of the threads modifies the map structurally, it <em>must</em> be
097 * synchronized externally. This is typically accomplished by
098 * synchronizing on some object that naturally encapsulates the map.
099 *
100 * If no such object exists, the map should be "wrapped" using the
101 * {@link Collections#synchronizedMap Collections.synchronizedMap}
102 * method. This is best done at creation time, to prevent accidental
103 * unsynchronized access to the map:<pre>
104 * Map m = Collections.synchronizedMap(new LinkedHashMap(...));</pre>
105 *
106 * A structural modification is any operation that adds or deletes one or more
107 * mappings or, in the case of access-ordered linked hash maps, affects
108 * iteration order. In insertion-ordered linked hash maps, merely changing
109 * the value associated with a key that is already contained in the map is not
110 * a structural modification. <strong>In access-ordered linked hash maps,
111 * merely querying the map with <tt>get</tt> is a structural
112 * modification.</strong>)
113 *
114 * <p>The iterators returned by the <tt>iterator</tt> method of the collections
115 * returned by all of this class's collection view methods are
116 * <em>fail-fast</em>: if the map is structurally modified at any time after
117 * the iterator is created, in any way except through the iterator's own
118 * <tt>remove</tt> method, the iterator will throw a {@link
119 * ConcurrentModificationException}. Thus, in the face of concurrent
120 * modification, the iterator fails quickly and cleanly, rather than risking
121 * arbitrary, non-deterministic behavior at an undetermined time in the future.
122 *
123 * <p>Note that the fail-fast behavior of an iterator cannot be guaranteed
124 * as it is, generally speaking, impossible to make any hard guarantees in the
125 * presence of unsynchronized concurrent modification. Fail-fast iterators
126 * throw <tt>ConcurrentModificationException</tt> on a best-effort basis.
127 * Therefore, it would be wrong to write a program that depended on this
128 * exception for its correctness: <i>the fail-fast behavior of iterators
129 * should be used only to detect bugs.</i>
130 *
131 * <p>This class is a member of the
132 * <a href="{@docRoot}/../technotes/guides/collections/index.html">
133 * Java Collections Framework</a>.
134 *
135 * @param <K> the type of keys maintained by this map
136 * @param <V> the type of mapped values
137 *
138 * @author Josh Bloch
139 * @version 1.32, 05/05/07
140 * @see Object#hashCode()
141 * @see Collection
142 * @see Map
143 * @see HashMap
144 * @see TreeMap
145 * @see Hashtable
146 * @since 1.4
147 */
148
149 public class LinkedHashMap<K, V> extends HashMap<K, V> implements
150 Map<K, V> {
151
152 private static final long serialVersionUID = 3801124242820219131L;
153
154 /**
155 * The head of the doubly linked list.
156 */
157 private transient Entry<K, V> header;
158
159 /**
160 * The iteration ordering method for this linked hash map: <tt>true</tt>
161 * for access-order, <tt>false</tt> for insertion-order.
162 *
163 * @serial
164 */
165 private final boolean accessOrder;
166
167 /**
168 * Constructs an empty insertion-ordered <tt>LinkedHashMap</tt> instance
169 * with the specified initial capacity and load factor.
170 *
171 * @param initialCapacity the initial capacity
172 * @param loadFactor the load factor
173 * @throws IllegalArgumentException if the initial capacity is negative
174 * or the load factor is nonpositive
175 */
176 public LinkedHashMap(int initialCapacity, float loadFactor) {
177 super (initialCapacity, loadFactor);
178 accessOrder = false;
179 }
180
181 /**
182 * Constructs an empty insertion-ordered <tt>LinkedHashMap</tt> instance
183 * with the specified initial capacity and a default load factor (0.75).
184 *
185 * @param initialCapacity the initial capacity
186 * @throws IllegalArgumentException if the initial capacity is negative
187 */
188 public LinkedHashMap(int initialCapacity) {
189 super (initialCapacity);
190 accessOrder = false;
191 }
192
193 /**
194 * Constructs an empty insertion-ordered <tt>LinkedHashMap</tt> instance
195 * with the default initial capacity (16) and load factor (0.75).
196 */
197 public LinkedHashMap() {
198 super ();
199 accessOrder = false;
200 }
201
202 /**
203 * Constructs an insertion-ordered <tt>LinkedHashMap</tt> instance with
204 * the same mappings as the specified map. The <tt>LinkedHashMap</tt>
205 * instance is created with a default load factor (0.75) and an initial
206 * capacity sufficient to hold the mappings in the specified map.
207 *
208 * @param m the map whose mappings are to be placed in this map
209 * @throws NullPointerException if the specified map is null
210 */
211 public LinkedHashMap(Map<? extends K, ? extends V> m) {
212 super (m);
213 accessOrder = false;
214 }
215
216 /**
217 * Constructs an empty <tt>LinkedHashMap</tt> instance with the
218 * specified initial capacity, load factor and ordering mode.
219 *
220 * @param initialCapacity the initial capacity
221 * @param loadFactor the load factor
222 * @param accessOrder the ordering mode - <tt>true</tt> for
223 * access-order, <tt>false</tt> for insertion-order
224 * @throws IllegalArgumentException if the initial capacity is negative
225 * or the load factor is nonpositive
226 */
227 public LinkedHashMap(int initialCapacity, float loadFactor,
228 boolean accessOrder) {
229 super (initialCapacity, loadFactor);
230 this .accessOrder = accessOrder;
231 }
232
233 /**
234 * Called by superclass constructors and pseudoconstructors (clone,
235 * readObject) before any entries are inserted into the map. Initializes
236 * the chain.
237 */
238 void init() {
239 header = new Entry<K, V>(-1, null, null, null);
240 header.before = header.after = header;
241 }
242
243 /**
244 * Transfers all entries to new table array. This method is called
245 * by superclass resize. It is overridden for performance, as it is
246 * faster to iterate using our linked list.
247 */
248 void transfer(HashMap.Entry[] newTable) {
249 int newCapacity = newTable.length;
250 for (Entry<K, V> e = header.after; e != header; e = e.after) {
251 int index = indexFor(e.hash, newCapacity);
252 e.next = newTable[index];
253 newTable[index] = e;
254 }
255 }
256
257 /**
258 * Returns <tt>true</tt> if this map maps one or more keys to the
259 * specified value.
260 *
261 * @param value value whose presence in this map is to be tested
262 * @return <tt>true</tt> if this map maps one or more keys to the
263 * specified value
264 */
265 public boolean containsValue(Object value) {
266 // Overridden to take advantage of faster iterator
267 if (value == null) {
268 for (Entry e = header.after; e != header; e = e.after)
269 if (e.value == null)
270 return true;
271 } else {
272 for (Entry e = header.after; e != header; e = e.after)
273 if (value.equals(e.value))
274 return true;
275 }
276 return false;
277 }
278
279 /**
280 * Returns the value to which the specified key is mapped,
281 * or {@code null} if this map contains no mapping for the key.
282 *
283 * <p>More formally, if this map contains a mapping from a key
284 * {@code k} to a value {@code v} such that {@code (key==null ? k==null :
285 * key.equals(k))}, then this method returns {@code v}; otherwise
286 * it returns {@code null}. (There can be at most one such mapping.)
287 *
288 * <p>A return value of {@code null} does not <i>necessarily</i>
289 * indicate that the map contains no mapping for the key; it's also
290 * possible that the map explicitly maps the key to {@code null}.
291 * The {@link #containsKey containsKey} operation may be used to
292 * distinguish these two cases.
293 */
294 public V get(Object key) {
295 Entry<K, V> e = (Entry<K, V>) getEntry(key);
296 if (e == null)
297 return null;
298 e.recordAccess(this );
299 return e.value;
300 }
301
302 /**
303 * Removes all of the mappings from this map.
304 * The map will be empty after this call returns.
305 */
306 public void clear() {
307 super .clear();
308 header.before = header.after = header;
309 }
310
311 /**
312 * LinkedHashMap entry.
313 */
314 private static class Entry<K, V> extends HashMap.Entry<K, V> {
315 // These fields comprise the doubly linked list used for iteration.
316 Entry<K, V> before, after;
317
318 Entry(int hash, K key, V value, HashMap.Entry<K, V> next) {
319 super (hash, key, value, next);
320 }
321
322 /**
323 * Removes this entry from the linked list.
324 */
325 private void remove() {
326 before.after = after;
327 after.before = before;
328 }
329
330 /**
331 * Inserts this entry before the specified existing entry in the list.
332 */
333 private void addBefore(Entry<K, V> existingEntry) {
334 after = existingEntry;
335 before = existingEntry.before;
336 before.after = this ;
337 after.before = this ;
338 }
339
340 /**
341 * This method is invoked by the superclass whenever the value
342 * of a pre-existing entry is read by Map.get or modified by Map.set.
343 * If the enclosing Map is access-ordered, it moves the entry
344 * to the end of the list; otherwise, it does nothing.
345 */
346 void recordAccess(HashMap<K, V> m) {
347 LinkedHashMap<K, V> lm = (LinkedHashMap<K, V>) m;
348 if (lm.accessOrder) {
349 lm.modCount++;
350 remove();
351 addBefore(lm.header);
352 }
353 }
354
355 void recordRemoval(HashMap<K, V> m) {
356 remove();
357 }
358 }
359
360 private abstract class LinkedHashIterator<T> implements Iterator<T> {
361 Entry<K, V> nextEntry = header.after;
362 Entry<K, V> lastReturned = null;
363
364 /**
365 * The modCount value that the iterator believes that the backing
366 * List should have. If this expectation is violated, the iterator
367 * has detected concurrent modification.
368 */
369 int expectedModCount = modCount;
370
371 public boolean hasNext() {
372 return nextEntry != header;
373 }
374
375 public void remove() {
376 if (lastReturned == null)
377 throw new IllegalStateException();
378 if (modCount != expectedModCount)
379 throw new ConcurrentModificationException();
380
381 LinkedHashMap.this .remove(lastReturned.key);
382 lastReturned = null;
383 expectedModCount = modCount;
384 }
385
386 Entry<K, V> nextEntry() {
387 if (modCount != expectedModCount)
388 throw new ConcurrentModificationException();
389 if (nextEntry == header)
390 throw new NoSuchElementException();
391
392 Entry<K, V> e = lastReturned = nextEntry;
393 nextEntry = e.after;
394 return e;
395 }
396 }
397
398 private class KeyIterator extends LinkedHashIterator<K> {
399 public K next() {
400 return nextEntry().getKey();
401 }
402 }
403
404 private class ValueIterator extends LinkedHashIterator<V> {
405 public V next() {
406 return nextEntry().value;
407 }
408 }
409
410 private class EntryIterator extends
411 LinkedHashIterator<Map.Entry<K, V>> {
412 public Map.Entry<K, V> next() {
413 return nextEntry();
414 }
415 }
416
417 // These Overrides alter the behavior of superclass view iterator() methods
418 Iterator<K> newKeyIterator() {
419 return new KeyIterator();
420 }
421
422 Iterator<V> newValueIterator() {
423 return new ValueIterator();
424 }
425
426 Iterator<Map.Entry<K, V>> newEntryIterator() {
427 return new EntryIterator();
428 }
429
430 /**
431 * This override alters behavior of superclass put method. It causes newly
432 * allocated entry to get inserted at the end of the linked list and
433 * removes the eldest entry if appropriate.
434 */
435 void addEntry(int hash, K key, V value, int bucketIndex) {
436 createEntry(hash, key, value, bucketIndex);
437
438 // Remove eldest entry if instructed, else grow capacity if appropriate
439 Entry<K, V> eldest = header.after;
440 if (removeEldestEntry(eldest)) {
441 removeEntryForKey(eldest.key);
442 } else {
443 if (size >= threshold)
444 resize(2 * table.length);
445 }
446 }
447
448 /**
449 * This override differs from addEntry in that it doesn't resize the
450 * table or remove the eldest entry.
451 */
452 void createEntry(int hash, K key, V value, int bucketIndex) {
453 HashMap.Entry<K, V> old = table[bucketIndex];
454 Entry<K, V> e = new Entry<K, V>(hash, key, value, old);
455 table[bucketIndex] = e;
456 e.addBefore(header);
457 size++;
458 }
459
460 /**
461 * Returns <tt>true</tt> if this map should remove its eldest entry.
462 * This method is invoked by <tt>put</tt> and <tt>putAll</tt> after
463 * inserting a new entry into the map. It provides the implementor
464 * with the opportunity to remove the eldest entry each time a new one
465 * is added. This is useful if the map represents a cache: it allows
466 * the map to reduce memory consumption by deleting stale entries.
467 *
468 * <p>Sample use: this override will allow the map to grow up to 100
469 * entries and then delete the eldest entry each time a new entry is
470 * added, maintaining a steady state of 100 entries.
471 * <pre>
472 * private static final int MAX_ENTRIES = 100;
473 *
474 * protected boolean removeEldestEntry(Map.Entry eldest) {
475 * return size() > MAX_ENTRIES;
476 * }
477 * </pre>
478 *
479 * <p>This method typically does not modify the map in any way,
480 * instead allowing the map to modify itself as directed by its
481 * return value. It <i>is</i> permitted for this method to modify
482 * the map directly, but if it does so, it <i>must</i> return
483 * <tt>false</tt> (indicating that the map should not attempt any
484 * further modification). The effects of returning <tt>true</tt>
485 * after modifying the map from within this method are unspecified.
486 *
487 * <p>This implementation merely returns <tt>false</tt> (so that this
488 * map acts like a normal map - the eldest element is never removed).
489 *
490 * @param eldest The least recently inserted entry in the map, or if
491 * this is an access-ordered map, the least recently accessed
492 * entry. This is the entry that will be removed it this
493 * method returns <tt>true</tt>. If the map was empty prior
494 * to the <tt>put</tt> or <tt>putAll</tt> invocation resulting
495 * in this invocation, this will be the entry that was just
496 * inserted; in other words, if the map contains a single
497 * entry, the eldest entry is also the newest.
498 * @return <tt>true</tt> if the eldest entry should be removed
499 * from the map; <tt>false</tt> if it should be retained.
500 */
501 protected boolean removeEldestEntry(Map.Entry<K, V> eldest) {
502 return false;
503 }
504 }
|