1 /*
   2  * Copyright (c) 1997, 2019, Oracle and/or its affiliates. All rights reserved.
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * This code is free software; you can redistribute it and/or modify it
   6  * under the terms of the GNU General Public License version 2 only, as
   7  * published by the Free Software Foundation.  Oracle designates this
   8  * particular file as subject to the "Classpath" exception as provided
   9  * by Oracle in the LICENSE file that accompanied this code.
  10  *
  11  * This code is distributed in the hope that it will be useful, but WITHOUT
  12  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  13  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  14  * version 2 for more details (a copy is included in the LICENSE file that
  15  * accompanied this code).
  16  *
  17  * You should have received a copy of the GNU General Public License version
  18  * 2 along with this work; if not, write to the Free Software Foundation,
  19  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  20  *
  21  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  22  * or visit www.oracle.com if you need additional information or have any
  23  * questions.
  24  */
  25 
  26 package java.util;
  27 
  28 import java.io.InvalidObjectException;
  29 import jdk.internal.access.SharedSecrets;
  30 
  31 /**
  32  * This class implements the {@code Set} interface, backed by a hash table
  33  * (actually a {@code HashMap} instance).  It makes no guarantees as to the
  34  * iteration order of the set; in particular, it does not guarantee that the
  35  * order will remain constant over time.  This class permits the {@code null}
  36  * element.
  37  *
  38  * <p>This class offers constant time performance for the basic operations
  39  * ({@code add}, {@code remove}, {@code contains} and {@code size}),
  40  * assuming the hash function disperses the elements properly among the
  41  * buckets.  Iterating over this set requires time proportional to the sum of
  42  * the {@code HashSet} instance's size (the number of elements) plus the
  43  * "capacity" of the backing {@code HashMap} instance (the number of
  44  * buckets).  Thus, it's very important not to set the initial capacity too
  45  * high (or the load factor too low) if iteration performance is important.
  46  *
  47  * <p><strong>Note that this implementation is not synchronized.</strong>
  48  * If multiple threads access a hash set concurrently, and at least one of
  49  * the threads modifies the set, it <i>must</i> be synchronized externally.
  50  * This is typically accomplished by synchronizing on some object that
  51  * naturally encapsulates the set.
  52  *
  53  * If no such object exists, the set should be "wrapped" using the
  54  * {@link Collections#synchronizedSet Collections.synchronizedSet}
  55  * method.  This is best done at creation time, to prevent accidental
  56  * unsynchronized access to the set:<pre>
  57  *   Set s = Collections.synchronizedSet(new HashSet(...));</pre>
  58  *
  59  * <p>The iterators returned by this class's {@code iterator} method are
  60  * <i>fail-fast</i>: if the set is modified at any time after the iterator is
  61  * created, in any way except through the iterator's own {@code remove}
  62  * method, the Iterator throws a {@link ConcurrentModificationException}.
  63  * Thus, in the face of concurrent modification, the iterator fails quickly
  64  * and cleanly, rather than risking arbitrary, non-deterministic behavior at
  65  * an undetermined time in the future.
  66  *
  67  * <p>Note that the fail-fast behavior of an iterator cannot be guaranteed
  68  * as it is, generally speaking, impossible to make any hard guarantees in the
  69  * presence of unsynchronized concurrent modification.  Fail-fast iterators
  70  * throw {@code ConcurrentModificationException} on a best-effort basis.
  71  * Therefore, it would be wrong to write a program that depended on this
  72  * exception for its correctness: <i>the fail-fast behavior of iterators
  73  * should be used only to detect bugs.</i>
  74  *
  75  * <p>This class is a member of the
  76  * <a href="{@docRoot}/java.base/java/util/package-summary.html#CollectionsFramework">
  77  * Java Collections Framework</a>.
  78  *
  79  * @param <E> the type of elements maintained by this set
  80  *
  81  * @author  Josh Bloch
  82  * @author  Neal Gafter
  83  * @see     Collection
  84  * @see     Set
  85  * @see     TreeSet
  86  * @see     HashMap
  87  * @since   1.2
  88  */
  89 
  90 public class HashSet<E>
  91     extends AbstractSet<E>
  92     implements Set<E>, Cloneable, java.io.Serializable
  93 {
  94     @java.io.Serial
  95     static final long serialVersionUID = -5024744406713321676L;
  96 
  97     private transient HashMap<E,Object> map;
  98 
  99     // Dummy value to associate with an Object in the backing Map
 100     private static final Object PRESENT = new Object();
 101 
 102     /**
 103      * Constructs a new, empty set; the backing {@code HashMap} instance has
 104      * default initial capacity (16) and load factor (0.75).
 105      */
 106     public HashSet() {
 107         map = new HashMap<>();
 108     }
 109 
 110     /**
 111      * Constructs a new set containing the elements in the specified
 112      * collection.  The {@code HashMap} is created with default load factor
 113      * (0.75) and an initial capacity sufficient to contain the elements in
 114      * the specified collection.
 115      *
 116      * @param c the collection whose elements are to be placed into this set
 117      * @throws NullPointerException if the specified collection is null
 118      */
 119     public HashSet(Collection<? extends E> c) {
 120         map = new HashMap<>(Math.max((int) (c.size()/.75f) + 1, 16));
 121         addAll(c);
 122     }
 123 
 124     /**
 125      * Constructs a new, empty set; the backing {@code HashMap} instance has
 126      * the specified initial capacity and the specified load factor.
 127      *
 128      * @param      initialCapacity   the initial capacity of the hash map
 129      * @param      loadFactor        the load factor of the hash map
 130      * @throws     IllegalArgumentException if the initial capacity is less
 131      *             than zero, or if the load factor is nonpositive
 132      */
 133     public HashSet(int initialCapacity, float loadFactor) {
 134         map = new HashMap<>(initialCapacity, loadFactor);
 135     }
 136 
 137     /**
 138      * Constructs a new, empty set; the backing {@code HashMap} instance has
 139      * the specified initial capacity and default load factor (0.75).
 140      *
 141      * @param      initialCapacity   the initial capacity of the hash table
 142      * @throws     IllegalArgumentException if the initial capacity is less
 143      *             than zero
 144      */
 145     public HashSet(int initialCapacity) {
 146         map = new HashMap<>(initialCapacity);
 147     }
 148 
 149     /**
 150      * Constructs a new, empty linked hash set.  (This package private
 151      * constructor is only used by LinkedHashSet.) The backing
 152      * HashMap instance is a LinkedHashMap with the specified initial
 153      * capacity and the specified load factor.
 154      *
 155      * @param      initialCapacity   the initial capacity of the hash map
 156      * @param      loadFactor        the load factor of the hash map
 157      * @param      dummy             ignored (distinguishes this
 158      *             constructor from other int, float constructor.)
 159      * @throws     IllegalArgumentException if the initial capacity is less
 160      *             than zero, or if the load factor is nonpositive
 161      */
 162     HashSet(int initialCapacity, float loadFactor, boolean dummy) {
 163         map = new LinkedHashMap<>(initialCapacity, loadFactor);
 164     }
 165 
 166     /**
 167      * Returns an iterator over the elements in this set.  The elements
 168      * are returned in no particular order.
 169      *
 170      * @return an Iterator over the elements in this set
 171      * @see ConcurrentModificationException
 172      */
 173     public Iterator<E> iterator() {
 174         return map.keySet().iterator();
 175     }
 176 
 177     /**
 178      * Returns the number of elements in this set (its cardinality).
 179      *
 180      * @return the number of elements in this set (its cardinality)
 181      */
 182     public int size() {
 183         return map.size();
 184     }
 185 
 186     /**
 187      * Returns {@code true} if this set contains no elements.
 188      *
 189      * @return {@code true} if this set contains no elements
 190      */
 191     public boolean isEmpty() {
 192         return map.isEmpty();
 193     }
 194 
 195     /**
 196      * Returns {@code true} if this set contains the specified element.
 197      * More formally, returns {@code true} if and only if this set
 198      * contains an element {@code e} such that
 199      * {@code Objects.equals(o, e)}.
 200      *
 201      * @param o element whose presence in this set is to be tested
 202      * @return {@code true} if this set contains the specified element
 203      */
 204     public boolean contains(Object o) {
 205         return map.containsKey(o);
 206     }
 207 
 208     /**
 209      * Adds the specified element to this set if it is not already present.
 210      * More formally, adds the specified element {@code e} to this set if
 211      * this set contains no element {@code e2} such that
 212      * {@code Objects.equals(e, e2)}.
 213      * If this set already contains the element, the call leaves the set
 214      * unchanged and returns {@code false}.
 215      *
 216      * @param e element to be added to this set
 217      * @return {@code true} if this set did not already contain the specified
 218      * element
 219      */
 220     public boolean add(E e) {
 221         return map.put(e, PRESENT)==null;
 222     }
 223 
 224     /**
 225      * Removes the specified element from this set if it is present.
 226      * More formally, removes an element {@code e} such that
 227      * {@code Objects.equals(o, e)},
 228      * if this set contains such an element.  Returns {@code true} if
 229      * this set contained the element (or equivalently, if this set
 230      * changed as a result of the call).  (This set will not contain the
 231      * element once the call returns.)
 232      *
 233      * @param o object to be removed from this set, if present
 234      * @return {@code true} if the set contained the specified element
 235      */
 236     public boolean remove(Object o) {
 237         return map.remove(o)==PRESENT;
 238     }
 239 
 240     /**
 241      * Removes all of the elements from this set.
 242      * The set will be empty after this call returns.
 243      */
 244     public void clear() {
 245         map.clear();
 246     }
 247 
 248     /**
 249      * Returns a shallow copy of this {@code HashSet} instance: the elements
 250      * themselves are not cloned.
 251      *
 252      * @return a shallow copy of this set
 253      */
 254     @SuppressWarnings("unchecked")
 255     public Object clone() {
 256         try {
 257             HashSet<E> newSet = (HashSet<E>) super.clone();
 258             newSet.map = (HashMap<E, Object>) map.clone();
 259             return newSet;
 260         } catch (CloneNotSupportedException e) {
 261             throw new InternalError(e);
 262         }
 263     }
 264 
 265     /**
 266      * Save the state of this {@code HashSet} instance to a stream (that is,
 267      * serialize it).
 268      *
 269      * @serialData The capacity of the backing {@code HashMap} instance
 270      *             (int), and its load factor (float) are emitted, followed by
 271      *             the size of the set (the number of elements it contains)
 272      *             (int), followed by all of its elements (each an Object) in
 273      *             no particular order.
 274      */
 275     @java.io.Serial
 276     private void writeObject(java.io.ObjectOutputStream s)
 277         throws java.io.IOException {
 278         // Write out any hidden serialization magic
 279         s.defaultWriteObject();
 280 
 281         // Write out HashMap capacity and load factor
 282         s.writeInt(map.capacity());
 283         s.writeFloat(map.loadFactor());
 284 
 285         // Write out size
 286         s.writeInt(map.size());
 287 
 288         // Write out all elements in the proper order.
 289         for (E e : map.keySet())
 290             s.writeObject(e);
 291     }
 292 
 293     /**
 294      * Reconstitute the {@code HashSet} instance from a stream (that is,
 295      * deserialize it).
 296      */
 297     @java.io.Serial
 298     private void readObject(java.io.ObjectInputStream s)
 299         throws java.io.IOException, ClassNotFoundException {
 300         // Read in any hidden serialization magic
 301         s.defaultReadObject();
 302 
 303         // Read capacity and verify non-negative.
 304         int capacity = s.readInt();
 305         if (capacity < 0) {
 306             throw new InvalidObjectException("Illegal capacity: " +
 307                                              capacity);
 308         }
 309 
 310         // Read load factor and verify positive and non NaN.
 311         float loadFactor = s.readFloat();
 312         if (loadFactor <= 0 || Float.isNaN(loadFactor)) {
 313             throw new InvalidObjectException("Illegal load factor: " +
 314                                              loadFactor);
 315         }
 316 
 317         // Read size and verify non-negative.
 318         int size = s.readInt();
 319         if (size < 0) {
 320             throw new InvalidObjectException("Illegal size: " +
 321                                              size);
 322         }
 323 
 324         // Set the capacity according to the size and load factor ensuring that
 325         // the HashMap is at least 25% full but clamping to maximum capacity.
 326         capacity = (int) Math.min(size * Math.min(1 / loadFactor, 4.0f),
 327                 HashMap.MAXIMUM_CAPACITY);
 328 
 329         // Constructing the backing map will lazily create an array when the first element is
 330         // added, so check it before construction. Call HashMap.tableSizeFor to compute the
 331         // actual allocation size. Check Map.Entry[].class since it's the nearest public type to
 332         // what is actually created.
 333         SharedSecrets.getJavaObjectInputStreamAccess()
 334                      .checkArray(s, Map.Entry[].class, HashMap.tableSizeFor(capacity));
 335 
 336         // Create backing HashMap
 337         map = (((HashSet<?>)this) instanceof LinkedHashSet ?
 338                new LinkedHashMap<>(capacity, loadFactor) :
 339                new HashMap<>(capacity, loadFactor));
 340 
 341         // Read in all elements in the proper order.
 342         for (int i=0; i<size; i++) {
 343             @SuppressWarnings("unchecked")
 344                 E e = (E) s.readObject();
 345             map.put(e, PRESENT);
 346         }
 347     }
 348 
 349     /**
 350      * Creates a <em><a href="Spliterator.html#binding">late-binding</a></em>
 351      * and <em>fail-fast</em> {@link Spliterator} over the elements in this
 352      * set.
 353      *
 354      * <p>The {@code Spliterator} reports {@link Spliterator#SIZED} and
 355      * {@link Spliterator#DISTINCT}.  Overriding implementations should document
 356      * the reporting of additional characteristic values.
 357      *
 358      * @return a {@code Spliterator} over the elements in this set
 359      * @since 1.8
 360      */
 361     public Spliterator<E> spliterator() {
 362         return new HashMap.KeySpliterator<>(map, 0, -1, 0, 0);
 363     }
 364 
 365     @Override
 366     public Object[] toArray() {
 367         return map.keysToArray(new Object[map.size()]);
 368     }
 369 
 370     @Override
 371     public <T> T[] toArray(T[] a) {
 372         return map.keysToArray(map.prepareArray(a));
 373     }
 374 }