Class Outline

/* 
 * Copyright (c) 1997, 2017, Oracle and/or its affiliates. All rights reserved. 
 * ORACLE PROPRIETARY/CONFIDENTIAL. Use is subject to license terms. 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 * 
 */ 
 
package java.util; 
 
import java.io.IOException; 
import java.io.InvalidObjectException; 
import java.io.Serializable; 
import java.lang.reflect.ParameterizedType; 
import java.lang.reflect.Type; 
import java.util.function.BiConsumer; 
import java.util.function.BiFunction; 
import java.util.function.Consumer; 
import java.util.function.Function; 
import sun.misc.SharedSecrets; 
 
/** 
 * Hash table based implementation of the <tt>Map</tt> interface.  This 
 * implementation provides all of the optional map operations, and permits 
 * <tt>null</tt> values and the <tt>null</tt> key.  (The <tt>HashMap</tt> 
 * class is roughly equivalent to <tt>Hashtable</tt>, except that it is 
 * unsynchronized and permits nulls.)  This class makes no guarantees as to 
 * the order of the map; in particular, it does not guarantee that the order 
 * will remain constant over time. 
 * 
 * <p>This implementation provides constant-time performance for the basic 
 * operations (<tt>get</tt> and <tt>put</tt>), assuming the hash function 
 * disperses the elements properly among the buckets.  Iteration over 
 * collection views requires time proportional to the "capacity" of the 
 * <tt>HashMap</tt> instance (the number of buckets) plus its size (the number 
 * of key-value mappings).  Thus, it's very important not to set the initial 
 * capacity too high (or the load factor too low) if iteration performance is 
 * important. 
 * 
 * <p>An instance of <tt>HashMap</tt> has two parameters that affect its 
 * performance: <i>initial capacity</i> and <i>load factor</i>.  The 
 * <i>capacity</i> is the number of buckets in the hash table, and the initial 
 * capacity is simply the capacity at the time the hash table is created.  The 
 * <i>load factor</i> is a measure of how full the hash table is allowed to 
 * get before its capacity is automatically increased.  When the number of 
 * entries in the hash table exceeds the product of the load factor and the 
 * current capacity, the hash table is <i>rehashed</i> (that is, internal data 
 * structures are rebuilt) so that the hash table has approximately twice the 
 * number of buckets. 
 * 
 * <p>As a general rule, the default load factor (.75) offers a good 
 * tradeoff between time and space costs.  Higher values decrease the 
 * space overhead but increase the lookup cost (reflected in most of 
 * the operations of the <tt>HashMap</tt> class, including 
 * <tt>get</tt> and <tt>put</tt>).  The expected number of entries in 
 * the map and its load factor should be taken into account when 
 * setting its initial capacity, so as to minimize the number of 
 * rehash operations.  If the initial capacity is greater than the 
 * maximum number of entries divided by the load factor, no rehash 
 * operations will ever occur. 
 * 
 * <p>If many mappings are to be stored in a <tt>HashMap</tt> 
 * instance, creating it with a sufficiently large capacity will allow 
 * the mappings to be stored more efficiently than letting it perform 
 * automatic rehashing as needed to grow the table.  Note that using 
 * many keys with the same {@code hashCode()} is a sure way to slow 
 * down performance of any hash table. To ameliorate impact, when keys 
 * are {@link Comparable}, this class may use comparison order among 
 * keys to help break ties. 
 * 
 * <p><strong>Note that this implementation is not synchronized.</strong> 
 * If multiple threads access a hash map concurrently, and at least one of 
 * the threads modifies the map structurally, it <i>must</i> be 
 * synchronized externally.  (A structural modification is any operation 
 * that adds or deletes one or more mappings; merely changing the value 
 * associated with a key that an instance already contains is not a 
 * structural modification.)  This is typically accomplished by 
 * synchronizing on some object that naturally encapsulates the map. 
 * 
 * If no such object exists, the map should be "wrapped" using the 
 * {@link Collections#synchronizedMap Collections.synchronizedMap} 
 * method.  This is best done at creation time, to prevent accidental 
 * unsynchronized access to the map:<pre> 
 *   Map m = Collections.synchronizedMap(new HashMap(...));</pre> 
 * 
 * <p>The iterators returned by all of this class's "collection view methods" 
 * are <i>fail-fast</i>: if the map is structurally modified at any time after 
 * the iterator is created, in any way except through the iterator's own 
 * <tt>remove</tt> method, the iterator will throw a 
 * {@link ConcurrentModificationException}.  Thus, in the face of concurrent 
 * modification, the iterator fails quickly and cleanly, rather than risking 
 * arbitrary, non-deterministic behavior at an undetermined time in the 
 * future. 
 * 
 * <p>Note that the fail-fast behavior of an iterator cannot be guaranteed 
 * as it is, generally speaking, impossible to make any hard guarantees in the 
 * presence of unsynchronized concurrent modification.  Fail-fast iterators 
 * throw <tt>ConcurrentModificationException</tt> on a best-effort basis. 
 * Therefore, it would be wrong to write a program that depended on this 
 * exception for its correctness: <i>the fail-fast behavior of iterators 
 * should be used only to detect bugs.</i> 
 * 
 * <p>This class is a member of the 
 * <a href="{@docRoot}/../technotes/guides/collections/index.html"> 
 * Java Collections Framework</a>. 
 * 
 * @param <K> the type of keys maintained by this map 
 * @param <V> the type of mapped values 
 * 
 * @author  Doug Lea 
 * @author  Josh Bloch 
 * @author  Arthur van Hoff 
 * @author  Neal Gafter 
 * @see     Object#hashCode() 
 * @see     Collection 
 * @see     Map 
 * @see     TreeMap 
 * @see     Hashtable 
 * @since   1.2 
 */ 
public class HashMap<K,V> extends AbstractMap<K,V> 
    implements Map<K,V>, Cloneable, Serializable { 
 
    private static final long serialVersionUID = 362498820763181265L; 
 
    /* 
     * Implementation notes. 
     * 
     * This map usually acts as a binned (bucketed) hash table, but 
     * when bins get too large, they are transformed into bins of 
     * TreeNodes, each structured similarly to those in 
     * java.util.TreeMap. Most methods try to use normal bins, but 
     * relay to TreeNode methods when applicable (simply by checking 
     * instanceof a node).  Bins of TreeNodes may be traversed and 
     * used like any others, but additionally support faster lookup 
     * when overpopulated. However, since the vast majority of bins in 
     * normal use are not overpopulated, checking for existence of 
     * tree bins may be delayed in the course of table methods. 
     * 
     * Tree bins (i.e., bins whose elements are all TreeNodes) are 
     * ordered primarily by hashCode, but in the case of ties, if two 
     * elements are of the same "class C implements Comparable<C>", 
     * type then their compareTo method is used for ordering. (We 
     * conservatively check generic types via reflection to validate 
     * this -- see method comparableClassFor).  The added complexity 
     * of tree bins is worthwhile in providing worst-case O(log n) 
     * operations when keys either have distinct hashes or are 
     * orderable, Thus, performance degrades gracefully under 
     * accidental or malicious usages in which hashCode() methods 
     * return values that are poorly distributed, as well as those in 
     * which many keys share a hashCode, so long as they are also 
     * Comparable. (If neither of these apply, we may waste about a 
     * factor of two in time and space compared to taking no 
     * precautions. But the only known cases stem from poor user 
     * programming practices that are already so slow that this makes 
     * little difference.) 
     * 
     * Because TreeNodes are about twice the size of regular nodes, we 
     * use them only when bins contain enough nodes to warrant use 
     * (see TREEIFY_THRESHOLD). And when they become too small (due to 
     * removal or resizing) they are converted back to plain bins.  In 
     * usages with well-distributed user hashCodes, tree bins are 
     * rarely used.  Ideally, under random hashCodes, the frequency of 
     * nodes in bins follows a Poisson distribution 
     * (http://en.wikipedia.org/wiki/Poisson_distribution) with a 
     * parameter of about 0.5 on average for the default resizing 
     * threshold of 0.75, although with a large variance because of 
     * resizing granularity. Ignoring variance, the expected 
     * occurrences of list size k are (exp(-0.5) * pow(0.5, k) / 
     * factorial(k)). The first values are: 
     * 
     * 0:    0.60653066 
     * 1:    0.30326533 
     * 2:    0.07581633 
     * 3:    0.01263606 
     * 4:    0.00157952 
     * 5:    0.00015795 
     * 6:    0.00001316 
     * 7:    0.00000094 
     * 8:    0.00000006 
     * more: less than 1 in ten million 
     * 
     * The root of a tree bin is normally its first node.  However, 
     * sometimes (currently only upon Iterator.remove), the root might 
     * be elsewhere, but can be recovered following parent links 
     * (method TreeNode.root()). 
     * 
     * All applicable internal methods accept a hash code as an 
     * argument (as normally supplied from a public method), allowing 
     * them to call each other without recomputing user hashCodes. 
     * Most internal methods also accept a "tab" argument, that is 
     * normally the current table, but may be a new or old one when 
     * resizing or converting. 
     * 
     * When bin lists are treeified, split, or untreeified, we keep 
     * them in the same relative access/traversal order (i.e., field 
     * Node.next) to better preserve locality, and to slightly 
     * simplify handling of splits and traversals that invoke 
     * iterator.remove. When using comparators on insertion, to keep a 
     * total ordering (or as close as is required here) across 
     * rebalancings, we compare classes and identityHashCodes as 
     * tie-breakers. 
     * 
     * The use and transitions among plain vs tree modes is 
     * complicated by the existence of subclass LinkedHashMap. See 
     * below for hook methods defined to be invoked upon insertion, 
     * removal and access that allow LinkedHashMap internals to 
     * otherwise remain independent of these mechanics. (This also 
     * requires that a map instance be passed to some utility methods 
     * that may create new nodes.) 
     * 
     * The concurrent-programming-like SSA-based coding style helps 
     * avoid aliasing errors amid all of the twisty pointer operations. 
     */ 
 
    /** 
     * The default initial capacity - MUST be a power of two. 
     */ 
    static final int DEFAULT_INITIAL_CAPACITY = 1 << 4; // aka 16 
 
    /** 
     * The maximum capacity, used if a higher value is implicitly specified 
     * by either of the constructors with arguments. 
     * MUST be a power of two <= 1<<30. 
     */ 
    static final int MAXIMUM_CAPACITY = 1 << 30; 
 
    /** 
     * The load factor used when none specified in constructor. 
     */ 
    static final float DEFAULT_LOAD_FACTOR = 0.75f; 
 
    /** 
     * The bin count threshold for using a tree rather than list for a 
     * bin.  Bins are converted to trees when adding an element to a 
     * bin with at least this many nodes. The value must be greater 
     * than 2 and should be at least 8 to mesh with assumptions in 
     * tree removal about conversion back to plain bins upon 
     * shrinkage. 
     */ 
    static final int TREEIFY_THRESHOLD = 8; 
 
    /** 
     * The bin count threshold for untreeifying a (split) bin during a 
     * resize operation. Should be less than TREEIFY_THRESHOLD, and at 
     * most 6 to mesh with shrinkage detection under removal. 
     */ 
    static final int UNTREEIFY_THRESHOLD = 6; 
 
    /** 
     * The smallest table capacity for which bins may be treeified. 
     * (Otherwise the table is resized if too many nodes in a bin.) 
     * Should be at least 4 * TREEIFY_THRESHOLD to avoid conflicts 
     * between resizing and treeification thresholds. 
     */ 
    static final int MIN_TREEIFY_CAPACITY = 64; 
 
    /** 
     * Basic hash bin node, used for most entries.  (See below for 
     * TreeNode subclass, and in LinkedHashMap for its Entry subclass.) 
     */ 
    static class Node<K,V> implements Map.Entry<K,V> { 
        final int hash; 
        final K key; 
        V value; 
        Node<K,V> next; 
 
        Node(int hash, K key, V value, Node<K,V> next) { 
            this.hash = hash; 
            this.key = key; 
            this.value = value; 
            this.next = next; 
        } 
 
        public final K getKey()        { return key; } 
        public final V getValue()      { return value; } 
        public final String toString() { return key + "=" + value; } 
 
        public final int hashCode() { 
            return Objects.hashCode(key) ^ Objects.hashCode(value); 
        } 
 
        public final V setValue(V newValue) { 
            V oldValue = value; 
            value = newValue; 
            return oldValue; 
        } 
 
        public final boolean equals(Object o) { 
            if (o == this) 
                return true; 
            if (o instanceof Map.Entry) { 
                Map.Entry<?,?> e = (Map.Entry<?,?>)o; 
                if (Objects.equals(key, e.getKey()) && 
                    Objects.equals(value, e.getValue())) 
                    return true; 
            } 
            return false; 
        } 
    } 
 
    /* ---------------- Static utilities -------------- */ 
 
    /** 
     * Computes key.hashCode() and spreads (XORs) higher bits of hash 
     * to lower.  Because the table uses power-of-two masking, sets of 
     * hashes that vary only in bits above the current mask will 
     * always collide. (Among known examples are sets of Float keys 
     * holding consecutive whole numbers in small tables.)  So we 
     * apply a transform that spreads the impact of higher bits 
     * downward. There is a tradeoff between speed, utility, and 
     * quality of bit-spreading. Because many common sets of hashes 
     * are already reasonably distributed (so don't benefit from 
     * spreading), and because we use trees to handle large sets of 
     * collisions in bins, we just XOR some shifted bits in the 
     * cheapest possible way to reduce systematic lossage, as well as 
     * to incorporate impact of the highest bits that would otherwise 
     * never be used in index calculations because of table bounds. 
     */ 
    static final int hash(Object key) { 
        int h; 
        return (key == null) ? 0 : (h = key.hashCode()) ^ (h >>> 16); 
    } 
 
    /** 
     * Returns x's Class if it is of the form "class C implements 
     * Comparable<C>", else null. 
     */ 
    static Class<?> comparableClassFor(Object x) { 
        if (x instanceof Comparable) { 
            Class<?> c; Type[] ts, as; Type t; ParameterizedType p; 
            if ((c = x.getClass()) == String.class) // bypass checks 
                return c; 
            if ((ts = c.getGenericInterfaces()) != null) { 
                for (int i = 0; i < ts.length; ++i) { 
                    if (((t = ts[i]) instanceof ParameterizedType) && 
                        ((p = (ParameterizedType)t).getRawType() == 
                         Comparable.class) && 
                        (as = p.getActualTypeArguments()) != null && 
                        as.length == 1 && as[0] == c) // type arg is c 
                        return c; 
                } 
            } 
        } 
        return null; 
    } 
 
    /** 
     * Returns k.compareTo(x) if x matches kc (k's screened comparable 
     * class), else 0. 
     */ 
    @SuppressWarnings({"rawtypes","unchecked"}) // for cast to Comparable 
    static int compareComparables(Class<?> kc, Object k, Object x) { 
        return (x == null || x.getClass() != kc ? 0 : 
                ((Comparable)k).compareTo(x)); 
    } 
 
    /** 
     * Returns a power of two size for the given target capacity. 
     */ 
    static final int tableSizeFor(int cap) { 
        int n = cap - 1; 
        n |= n >>> 1; 
        n |= n >>> 2; 
        n |= n >>> 4; 
        n |= n >>> 8; 
        n |= n >>> 16; 
        return (n < 0) ? 1 : (n >= MAXIMUM_CAPACITY) ? MAXIMUM_CAPACITY : n + 1; 
    } 
 
    /* ---------------- Fields -------------- */ 
 
    /** 
     * The table, initialized on first use, and resized as 
     * necessary. When allocated, length is always a power of two. 
     * (We also tolerate length zero in some operations to allow 
     * bootstrapping mechanics that are currently not needed.) 
     */ 
    transient Node<K,V>[] table; 
 
    /** 
     * Holds cached entrySet(). Note that AbstractMap fields are used 
     * for keySet() and values(). 
     */ 
    transient Set<Map.Entry<K,V>> entrySet; 
 
    /** 
     * The number of key-value mappings contained in this map. 
     */ 
    transient int size; 
 
    /** 
     * The number of times this HashMap has been structurally modified 
     * Structural modifications are those that change the number of mappings in 
     * the HashMap or otherwise modify its internal structure (e.g., 
     * rehash).  This field is used to make iterators on Collection-views of 
     * the HashMap fail-fast.  (See ConcurrentModificationException). 
     */ 
    transient int modCount; 
 
    /** 
     * The next size value at which to resize (capacity * load factor). 
     * 
     * @serial 
     */ 
    // (The javadoc description is true upon serialization. 
    // Additionally, if the table array has not been allocated, this 
    // field holds the initial array capacity, or zero signifying 
    // DEFAULT_INITIAL_CAPACITY.) 
    int threshold; 
 
    /** 
     * The load factor for the hash table. 
     * 
     * @serial 
     */ 
    final float loadFactor; 
 
    /* ---------------- Public operations -------------- */ 
 
    /** 
     * Constructs an empty <tt>HashMap</tt> with the specified initial 
     * capacity and load factor. 
     * 
     * @param  initialCapacity the initial capacity 
     * @param  loadFactor      the load factor 
     * @throws IllegalArgumentException if the initial capacity is negative 
     *         or the load factor is nonpositive 
     */ 
    public HashMap(int initialCapacity, float loadFactor) { 
        if (initialCapacity < 0) 
            throw new IllegalArgumentException("Illegal initial capacity: " + 
                                               initialCapacity); 
        if (initialCapacity > MAXIMUM_CAPACITY) 
            initialCapacity = MAXIMUM_CAPACITY; 
        if (loadFactor <= 0 || Float.isNaN(loadFactor)) 
            throw new IllegalArgumentException("Illegal load factor: " + 
                                               loadFactor); 
        this.loadFactor = loadFactor; 
        this.threshold = tableSizeFor(initialCapacity); 
    } 
 
    /** 
     * Constructs an empty <tt>HashMap</tt> with the specified initial 
     * capacity and the default load factor (0.75). 
     * 
     * @param  initialCapacity the initial capacity. 
     * @throws IllegalArgumentException if the initial capacity is negative. 
     */ 
    public HashMap(int initialCapacity) { 
        this(initialCapacity, DEFAULT_LOAD_FACTOR); 
    } 
 
    /** 
     * Constructs an empty <tt>HashMap</tt> with the default initial capacity 
     * (16) and the default load factor (0.75). 
     */ 
    public HashMap() { 
        this.loadFactor = DEFAULT_LOAD_FACTOR; // all other fields defaulted 
    } 
 
    /** 
     * Constructs a new <tt>HashMap</tt> with the same mappings as the 
     * specified <tt>Map</tt>.  The <tt>HashMap</tt> is created with 
     * default load factor (0.75) and an initial capacity sufficient to 
     * hold the mappings in the specified <tt>Map</tt>. 
     * 
     * @param   m the map whose mappings are to be placed in this map 
     * @throws  NullPointerException if the specified map is null 
     */ 
    public HashMap(Map<? extends K, ? extends V> m) { 
        this.loadFactor = DEFAULT_LOAD_FACTOR; 
        putMapEntries(m, false); 
    } 
 
    /** 
     * Implements Map.putAll and Map constructor 
     * 
     * @param m the map 
     * @param evict false when initially constructing this map, else 
     * true (relayed to method afterNodeInsertion). 
     */ 
    final void putMapEntries(Map<? extends K, ? extends V> m, boolean evict) { 
        int s = m.size(); 
        if (s > 0) { 
            if (table == null) { // pre-size 
                float ft = ((float)s / loadFactor) + 1.0F; 
                int t = ((ft < (float)MAXIMUM_CAPACITY) ? 
                         (int)ft : MAXIMUM_CAPACITY); 
                if (t > threshold) 
                    threshold = tableSizeFor(t); 
            } 
            else if (s > threshold) 
                resize(); 
            for (Map.Entry<? extends K, ? extends V> e : m.entrySet()) { 
                K key = e.getKey(); 
                V value = e.getValue(); 
                putVal(hash(key), key, value, false, evict); 
            } 
        } 
    } 
 
    /** 
     * Returns the number of key-value mappings in this map. 
     * 
     * @return the number of key-value mappings in this map 
     */ 
    public int size() { 
        return size; 
    } 
 
    /** 
     * Returns <tt>true</tt> if this map contains no key-value mappings. 
     * 
     * @return <tt>true</tt> if this map contains no key-value mappings 
     */ 
    public boolean isEmpty() { 
        return size == 0; 
    } 
 
    /** 
     * Returns the value to which the specified key is mapped, 
     * or {@code null} if this map contains no mapping for the key. 
     * 
     * <p>More formally, if this map contains a mapping from a key 
     * {@code k} to a value {@code v} such that {@code (key==null ? k==null : 
     * key.equals(k))}, then this method returns {@code v}; otherwise 
     * it returns {@code null}.  (There can be at most one such mapping.) 
     * 
     * <p>A return value of {@code null} does not <i>necessarily</i> 
     * indicate that the map contains no mapping for the key; it's also 
     * possible that the map explicitly maps the key to {@code null}. 
     * The {@link #containsKey containsKey} operation may be used to 
     * distinguish these two cases. 
     * 
     * @see #put(Object, Object) 
     */ 
    public V get(Object key) { 
        Node<K,V> e; 
        return (e = getNode(hash(key), key)) == null ? null : e.value; 
    } 
 
    /** 
     * Implements Map.get and related methods 
     * 
     * @param hash hash for key 
     * @param key the key 
     * @return the node, or null if none 
     */ 
    final Node<K,V> getNode(int hash, Object key) { 
        Node<K,V>[] tab; Node<K,V> first, e; int n; K k; 
        if ((tab = table) != null && (n = tab.length) > 0 && 
            (first = tab[(n - 1) & hash]) != null) { 
            if (first.hash == hash && // always check first node 
                ((k = first.key) == key || (key != null && key.equals(k)))) 
                return first; 
            if ((e = first.next) != null) { 
                if (first instanceof TreeNode) 
                    return ((TreeNode<K,V>)first).getTreeNode(hash, key); 
                do { 
                    if (e.hash == hash && 
                        ((k = e.key) == key || (key != null && key.equals(k)))) 
                        return e; 
                } while ((e = e.next) != null); 
            } 
        } 
        return null; 
    } 
 
    /** 
     * Returns <tt>true</tt> if this map contains a mapping for the 
     * specified key. 
     * 
     * @param   key   The key whose presence in this map is to be tested 
     * @return <tt>true</tt> if this map contains a mapping for the specified 
     * key. 
     */ 
    public boolean containsKey(Object key) { 
        return getNode(hash(key), key) != null; 
    } 
 
    /** 
     * Associates the specified value with the specified key in this map. 
     * If the map previously contained a mapping for the key, the old 
     * value is replaced. 
     * 
     * @param key key with which the specified value is to be associated 
     * @param value value to be associated with the specified key 
     * @return the previous value associated with <tt>key</tt>, or 
     *         <tt>null</tt> if there was no mapping for <tt>key</tt>. 
     *         (A <tt>null</tt> return can also indicate that the map 
     *         previously associated <tt>null</tt> with <tt>key</tt>.) 
     */ 
    public V put(K key, V value) { 
        return putVal(hash(key), key, value, false, true); 
    } 
 
    /** 
     * Implements Map.put and related methods 
     * 
     * @param hash hash for key 
     * @param key the key 
     * @param value the value to put 
     * @param onlyIfAbsent if true, don't change existing value 
     * @param evict if false, the table is in creation mode. 
     * @return previous value, or null if none 
     */ 
    final V putVal(int hash, K key, V value, boolean onlyIfAbsent, 
                   boolean evict) { 
        Node<K,V>[] tab; Node<K,V> p; int n, i; 
        if ((tab = table) == null || (n = tab.length) == 0) 
            n = (tab = resize()).length; 
        if ((p = tab[i = (n - 1) & hash]) == null) 
            tab[i] = newNode(hash, key, value, null); 
        else { 
            Node<K,V> e; K k; 
            if (p.hash == hash && 
                ((k = p.key) == key || (key != null && key.equals(k)))) 
                e = p; 
            else if (p instanceof TreeNode) 
                e = ((TreeNode<K,V>)p).putTreeVal(this, tab, hash, key, value); 
            else { 
                for (int binCount = 0; ; ++binCount) { 
                    if ((e = p.next) == null) { 
                        p.next = newNode(hash, key, value, null); 
                        if (binCount >= TREEIFY_THRESHOLD - 1) // -1 for 1st 
                            treeifyBin(tab, hash); 
                        break; 
                    } 
                    if (e.hash == hash && 
                        ((k = e.key) == key || (key != null && key.equals(k)))) 
                        break; 
                    p = e; 
                } 
            } 
            if (e != null) { // existing mapping for key 
                V oldValue = e.value; 
                if (!onlyIfAbsent || oldValue == null) 
                    e.value = value; 
                afterNodeAccess(e); 
                return oldValue; 
            } 
        } 
        ++modCount; 
        if (++size > threshold) 
            resize(); 
        afterNodeInsertion(evict); 
        return null; 
    } 
 
    /** 
     * Initializes or doubles table size.  If null, allocates in 
     * accord with initial capacity target held in field threshold. 
     * Otherwise, because we are using power-of-two expansion, the 
     * elements from each bin must either stay at same index, or move 
     * with a power of two offset in the new table. 
     * 
     * @return the table 
     */ 
    final Node<K,V>[] resize() { 
        Node<K,V>[] oldTab = table; 
        int oldCap = (oldTab == null) ? 0 : oldTab.length; 
        int oldThr = threshold; 
        int newCap, newThr = 0; 
        if (oldCap > 0) { 
            if (oldCap >= MAXIMUM_CAPACITY) { 
                threshold = Integer.MAX_VALUE; 
                return oldTab; 
            } 
            else if ((newCap = oldCap << 1) < MAXIMUM_CAPACITY && 
                     oldCap >= DEFAULT_INITIAL_CAPACITY) 
                newThr = oldThr << 1; // double threshold 
        } 
        else if (oldThr > 0) // initial capacity was placed in threshold 
            newCap = oldThr; 
        else {               // zero initial threshold signifies using defaults 
            newCap = DEFAULT_INITIAL_CAPACITY; 
            newThr = (int)(DEFAULT_LOAD_FACTOR * DEFAULT_INITIAL_CAPACITY); 
        } 
        if (newThr == 0) { 
            float ft = (float)newCap * loadFactor; 
            newThr = (newCap < MAXIMUM_CAPACITY && ft < (float)MAXIMUM_CAPACITY ? 
                      (int)ft : Integer.MAX_VALUE); 
        } 
        threshold = newThr; 
        @SuppressWarnings({"rawtypes","unchecked"}) 
            Node<K,V>[] newTab = (Node<K,V>[])new Node[newCap]; 
        table = newTab; 
        if (oldTab != null) { 
            for (int j = 0; j < oldCap; ++j) { 
                Node<K,V> e; 
                if ((e = oldTab[j]) != null) { 
                    oldTab[j] = null; 
                    if (e.next == null) 
                        newTab[e.hash & (newCap - 1)] = e; 
                    else if (e instanceof TreeNode) 
                        ((TreeNode<K,V>)e).split(this, newTab, j, oldCap); 
                    else { // preserve order 
                        Node<K,V> loHead = null, loTail = null; 
                        Node<K,V> hiHead = null, hiTail = null; 
                        Node<K,V> next; 
                        do { 
                            next = e.next; 
                            if ((e.hash & oldCap) == 0) { 
                                if (loTail == null) 
                                    loHead = e; 
                                else 
                                    loTail.next = e; 
                                loTail = e; 
                            } 
                            else { 
                                if (hiTail == null) 
                                    hiHead = e; 
                                else 
                                    hiTail.next = e; 
                                hiTail = e; 
                            } 
                        } while ((e = next) != null); 
                        if (loTail != null) { 
                            loTail.next = null; 
                            newTab[j] = loHead; 
                        } 
                        if (hiTail != null) { 
                            hiTail.next = null; 
                            newTab[j + oldCap] = hiHead; 
                        } 
                    } 
                } 
            } 
        } 
        return newTab; 
    } 
 
    /** 
     * Replaces all linked nodes in bin at index for given hash unless 
     * table is too small, in which case resizes instead. 
     */ 
    final void treeifyBin(Node<K,V>[] tab, int hash) { 
        int n, index; Node<K,V> e; 
        if (tab == null || (n = tab.length) < MIN_TREEIFY_CAPACITY) 
            resize(); 
        else if ((e = tab[index = (n - 1) & hash]) != null) { 
            TreeNode<K,V> hd = null, tl = null; 
            do { 
                TreeNode<K,V> p = replacementTreeNode(e, null); 
                if (tl == null) 
                    hd = p; 
                else { 
                    p.prev = tl; 
                    tl.next = p; 
                } 
                tl = p; 
            } while ((e = e.next) != null); 
            if ((tab[index] = hd) != null) 
                hd.treeify(tab); 
        } 
    } 
 
    /** 
     * Copies all of the mappings from the specified map to this map. 
     * These mappings will replace any mappings that this map had for 
     * any of the keys currently in the specified map. 
     * 
     * @param m mappings to be stored in this map 
     * @throws NullPointerException if the specified map is null 
     */ 
    public void putAll(Map<? extends K, ? extends V> m) { 
        putMapEntries(m, true); 
    } 
 
    /** 
     * Removes the mapping for the specified key from this map if present. 
     * 
     * @param  key key whose mapping is to be removed from the map 
     * @return the previous value associated with <tt>key</tt>, or 
     *         <tt>null</tt> if there was no mapping for <tt>key</tt>. 
     *         (A <tt>null</tt> return can also indicate that the map 
     *         previously associated <tt>null</tt> with <tt>key</tt>.) 
     */ 
    public V remove(Object key) { 
        Node<K,V> e; 
        return (e = removeNode(hash(key), key, null, false, true)) == null ? 
            null : e.value; 
    } 
 
    /** 
     * Implements Map.remove and related methods 
     * 
     * @param hash hash for key 
     * @param key the key 
     * @param value the value to match if matchValue, else ignored 
     * @param matchValue if true only remove if value is equal 
     * @param movable if false do not move other nodes while removing 
     * @return the node, or null if none 
     */ 
    final Node<K,V> removeNode(int hash, Object key, Object value, 
                               boolean matchValue, boolean movable) { 
        Node<K,V>[] tab; Node<K,V> p; int n, index; 
        if ((tab = table) != null && (n = tab.length) > 0 && 
            (p = tab[index = (n - 1) & hash]) != null) { 
            Node<K,V> node = null, e; K k; V v; 
            if (p.hash == hash && 
                ((k = p.key) == key || (key != null && key.equals(k)))) 
                node = p; 
            else if ((e = p.next) != null) { 
                if (p instanceof TreeNode) 
                    node = ((TreeNode<K,V>)p).getTreeNode(hash, key); 
                else { 
                    do { 
                        if (e.hash == hash && 
                            ((k = e.key) == key || 
                             (key != null && key.equals(k)))) { 
                            node = e; 
                            break; 
                        } 
                        p = e; 
                    } while ((e = e.next) != null); 
                } 
            } 
            if (node != null && (!matchValue || (v = node.value) == value || 
                                 (value != null && value.equals(v)))) { 
                if (node instanceof TreeNode) 
                    ((TreeNode<K,V>)node).removeTreeNode(this, tab, movable); 
                else if (node == p) 
                    tab[index] = node.next; 
                else 
                    p.next = node.next; 
                ++modCount; 
                --size; 
                afterNodeRemoval(node); 
                return node; 
            } 
        } 
        return null; 
    } 
 
    /** 
     * Removes all of the mappings from this map. 
     * The map will be empty after this call returns. 
     */ 
    public void clear() { 
        Node<K,V>[] tab; 
        modCount++; 
        if ((tab = table) != null && size > 0) { 
            size = 0; 
            for (int i = 0; i < tab.length; ++i) 
                tab[i] = null; 
        } 
    } 
 
    /** 
     * Returns <tt>true</tt> if this map maps one or more keys to the 
     * specified value. 
     * 
     * @param value value whose presence in this map is to be tested 
     * @return <tt>true</tt> if this map maps one or more keys to the 
     *         specified value 
     */ 
    public boolean containsValue(Object value) { 
        Node<K,V>[] tab; V v; 
        if ((tab = table) != null && size > 0) { 
            for (int i = 0; i < tab.length; ++i) { 
                for (Node<K,V> e = tab[i]; e != null; e = e.next) { 
                    if ((v = e.value) == value || 
                        (value != null && value.equals(v))) 
                        return true; 
                } 
            } 
        } 
        return false; 
    } 
 
    /** 
     * Returns a {@link Set} view of the keys contained in this map. 
     * The set is backed by the map, so changes to the map are 
     * reflected in the set, and vice-versa.  If the map is modified 
     * while an iteration over the set is in progress (except through 
     * the iterator's own <tt>remove</tt> operation), the results of 
     * the iteration are undefined.  The set supports element removal, 
     * which removes the corresponding mapping from the map, via the 
     * <tt>Iterator.remove</tt>, <tt>Set.remove</tt>, 
     * <tt>removeAll</tt>, <tt>retainAll</tt>, and <tt>clear</tt> 
     * operations.  It does not support the <tt>add</tt> or <tt>addAll</tt> 
     * operations. 
     * 
     * @return a set view of the keys contained in this map 
     */ 
    public Set<K> keySet() { 
        Set<K> ks = keySet; 
        if (ks == null) { 
            ks = new KeySet(); 
            keySet = ks; 
        } 
        return ks; 
    } 
 
    final class KeySet extends AbstractSet<K> { 
        public final int size()                 { return size; } 
        public final void clear()               { HashMap.this.clear(); } 
        public final Iterator<K> iterator()     { return new KeyIterator(); } 
        public final boolean contains(Object o) { return containsKey(o); } 
        public final boolean remove(Object key) { 
            return removeNode(hash(key), key, null, false, true) != null; 
        } 
        public final Spliterator<K> spliterator() { 
            return new KeySpliterator<>(HashMap.this, 0, -1, 0, 0); 
        } 
        public final void forEach(Consumer<? super K> action) { 
            Node<K,V>[] tab; 
            if (action == null) 
                throw new NullPointerException(); 
            if (size > 0 && (tab = table) != null) { 
                int mc = modCount; 
                for (int i = 0; i < tab.length; ++i) { 
                    for (Node<K,V> e = tab[i]; e != null; e = e.next) 
                        action.accept(e.key); 
                } 
                if (modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
    } 
 
    /** 
     * Returns a {@link Collection} view of the values contained in this map. 
     * The collection is backed by the map, so changes to the map are 
     * reflected in the collection, and vice-versa.  If the map is 
     * modified while an iteration over the collection is in progress 
     * (except through the iterator's own <tt>remove</tt> operation), 
     * the results of the iteration are undefined.  The collection 
     * supports element removal, which removes the corresponding 
     * mapping from the map, via the <tt>Iterator.remove</tt>, 
     * <tt>Collection.remove</tt>, <tt>removeAll</tt>, 
     * <tt>retainAll</tt> and <tt>clear</tt> operations.  It does not 
     * support the <tt>add</tt> or <tt>addAll</tt> operations. 
     * 
     * @return a view of the values contained in this map 
     */ 
    public Collection<V> values() { 
        Collection<V> vs = values; 
        if (vs == null) { 
            vs = new Values(); 
            values = vs; 
        } 
        return vs; 
    } 
 
    final class Values extends AbstractCollection<V> { 
        public final int size()                 { return size; } 
        public final void clear()               { HashMap.this.clear(); } 
        public final Iterator<V> iterator()     { return new ValueIterator(); } 
        public final boolean contains(Object o) { return containsValue(o); } 
        public final Spliterator<V> spliterator() { 
            return new ValueSpliterator<>(HashMap.this, 0, -1, 0, 0); 
        } 
        public final void forEach(Consumer<? super V> action) { 
            Node<K,V>[] tab; 
            if (action == null) 
                throw new NullPointerException(); 
            if (size > 0 && (tab = table) != null) { 
                int mc = modCount; 
                for (int i = 0; i < tab.length; ++i) { 
                    for (Node<K,V> e = tab[i]; e != null; e = e.next) 
                        action.accept(e.value); 
                } 
                if (modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
    } 
 
    /** 
     * Returns a {@link Set} view of the mappings contained in this map. 
     * The set is backed by the map, so changes to the map are 
     * reflected in the set, and vice-versa.  If the map is modified 
     * while an iteration over the set is in progress (except through 
     * the iterator's own <tt>remove</tt> operation, or through the 
     * <tt>setValue</tt> operation on a map entry returned by the 
     * iterator) the results of the iteration are undefined.  The set 
     * supports element removal, which removes the corresponding 
     * mapping from the map, via the <tt>Iterator.remove</tt>, 
     * <tt>Set.remove</tt>, <tt>removeAll</tt>, <tt>retainAll</tt> and 
     * <tt>clear</tt> operations.  It does not support the 
     * <tt>add</tt> or <tt>addAll</tt> operations. 
     * 
     * @return a set view of the mappings contained in this map 
     */ 
    public Set<Map.Entry<K,V>> entrySet() { 
        Set<Map.Entry<K,V>> es; 
        return (es = entrySet) == null ? (entrySet = new EntrySet()) : es; 
    } 
 
    final class EntrySet extends AbstractSet<Map.Entry<K,V>> { 
        public final int size()                 { return size; } 
        public final void clear()               { HashMap.this.clear(); } 
        public final Iterator<Map.Entry<K,V>> iterator() { 
            return new EntryIterator(); 
        } 
        public final boolean contains(Object o) { 
            if (!(o instanceof Map.Entry)) 
                return false; 
            Map.Entry<?,?> e = (Map.Entry<?,?>) o; 
            Object key = e.getKey(); 
            Node<K,V> candidate = getNode(hash(key), key); 
            return candidate != null && candidate.equals(e); 
        } 
        public final boolean remove(Object o) { 
            if (o instanceof Map.Entry) { 
                Map.Entry<?,?> e = (Map.Entry<?,?>) o; 
                Object key = e.getKey(); 
                Object value = e.getValue(); 
                return removeNode(hash(key), key, value, true, true) != null; 
            } 
            return false; 
        } 
        public final Spliterator<Map.Entry<K,V>> spliterator() { 
            return new EntrySpliterator<>(HashMap.this, 0, -1, 0, 0); 
        } 
        public final void forEach(Consumer<? super Map.Entry<K,V>> action) { 
            Node<K,V>[] tab; 
            if (action == null) 
                throw new NullPointerException(); 
            if (size > 0 && (tab = table) != null) { 
                int mc = modCount; 
                for (int i = 0; i < tab.length; ++i) { 
                    for (Node<K,V> e = tab[i]; e != null; e = e.next) 
                        action.accept(e); 
                } 
                if (modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
    } 
 
    // Overrides of JDK8 Map extension methods 
 
    @Override 
    public V getOrDefault(Object key, V defaultValue) { 
        Node<K,V> e; 
        return (e = getNode(hash(key), key)) == null ? defaultValue : e.value; 
    } 
 
    @Override 
    public V putIfAbsent(K key, V value) { 
        return putVal(hash(key), key, value, true, true); 
    } 
 
    @Override 
    public boolean remove(Object key, Object value) { 
        return removeNode(hash(key), key, value, true, true) != null; 
    } 
 
    @Override 
    public boolean replace(K key, V oldValue, V newValue) { 
        Node<K,V> e; V v; 
        if ((e = getNode(hash(key), key)) != null && 
            ((v = e.value) == oldValue || (v != null && v.equals(oldValue)))) { 
            e.value = newValue; 
            afterNodeAccess(e); 
            return true; 
        } 
        return false; 
    } 
 
    @Override 
    public V replace(K key, V value) { 
        Node<K,V> e; 
        if ((e = getNode(hash(key), key)) != null) { 
            V oldValue = e.value; 
            e.value = value; 
            afterNodeAccess(e); 
            return oldValue; 
        } 
        return null; 
    } 
 
    @Override 
    public V computeIfAbsent(K key, 
                             Function<? super K, ? extends V> mappingFunction) { 
        if (mappingFunction == null) 
            throw new NullPointerException(); 
        int hash = hash(key); 
        Node<K,V>[] tab; Node<K,V> first; int n, i; 
        int binCount = 0; 
        TreeNode<K,V> t = null; 
        Node<K,V> old = null; 
        if (size > threshold || (tab = table) == null || 
            (n = tab.length) == 0) 
            n = (tab = resize()).length; 
        if ((first = tab[i = (n - 1) & hash]) != null) { 
            if (first instanceof TreeNode) 
                old = (t = (TreeNode<K,V>)first).getTreeNode(hash, key); 
            else { 
                Node<K,V> e = first; K k; 
                do { 
                    if (e.hash == hash && 
                        ((k = e.key) == key || (key != null && key.equals(k)))) { 
                        old = e; 
                        break; 
                    } 
                    ++binCount; 
                } while ((e = e.next) != null); 
            } 
            V oldValue; 
            if (old != null && (oldValue = old.value) != null) { 
                afterNodeAccess(old); 
                return oldValue; 
            } 
        } 
        V v = mappingFunction.apply(key); 
        if (v == null) { 
            return null; 
        } else if (old != null) { 
            old.value = v; 
            afterNodeAccess(old); 
            return v; 
        } 
        else if (t != null) 
            t.putTreeVal(this, tab, hash, key, v); 
        else { 
            tab[i] = newNode(hash, key, v, first); 
            if (binCount >= TREEIFY_THRESHOLD - 1) 
                treeifyBin(tab, hash); 
        } 
        ++modCount; 
        ++size; 
        afterNodeInsertion(true); 
        return v; 
    } 
 
    public V computeIfPresent(K key, 
                              BiFunction<? super K, ? super V, ? extends V> remappingFunction) { 
        if (remappingFunction == null) 
            throw new NullPointerException(); 
        Node<K,V> e; V oldValue; 
        int hash = hash(key); 
        if ((e = getNode(hash, key)) != null && 
            (oldValue = e.value) != null) { 
            V v = remappingFunction.apply(key, oldValue); 
            if (v != null) { 
                e.value = v; 
                afterNodeAccess(e); 
                return v; 
            } 
            else 
                removeNode(hash, key, null, false, true); 
        } 
        return null; 
    } 
 
    @Override 
    public V compute(K key, 
                     BiFunction<? super K, ? super V, ? extends V> remappingFunction) { 
        if (remappingFunction == null) 
            throw new NullPointerException(); 
        int hash = hash(key); 
        Node<K,V>[] tab; Node<K,V> first; int n, i; 
        int binCount = 0; 
        TreeNode<K,V> t = null; 
        Node<K,V> old = null; 
        if (size > threshold || (tab = table) == null || 
            (n = tab.length) == 0) 
            n = (tab = resize()).length; 
        if ((first = tab[i = (n - 1) & hash]) != null) { 
            if (first instanceof TreeNode) 
                old = (t = (TreeNode<K,V>)first).getTreeNode(hash, key); 
            else { 
                Node<K,V> e = first; K k; 
                do { 
                    if (e.hash == hash && 
                        ((k = e.key) == key || (key != null && key.equals(k)))) { 
                        old = e; 
                        break; 
                    } 
                    ++binCount; 
                } while ((e = e.next) != null); 
            } 
        } 
        V oldValue = (old == null) ? null : old.value; 
        V v = remappingFunction.apply(key, oldValue); 
        if (old != null) { 
            if (v != null) { 
                old.value = v; 
                afterNodeAccess(old); 
            } 
            else 
                removeNode(hash, key, null, false, true); 
        } 
        else if (v != null) { 
            if (t != null) 
                t.putTreeVal(this, tab, hash, key, v); 
            else { 
                tab[i] = newNode(hash, key, v, first); 
                if (binCount >= TREEIFY_THRESHOLD - 1) 
                    treeifyBin(tab, hash); 
            } 
            ++modCount; 
            ++size; 
            afterNodeInsertion(true); 
        } 
        return v; 
    } 
 
    @Override 
    public V merge(K key, V value, 
                   BiFunction<? super V, ? super V, ? extends V> remappingFunction) { 
        if (value == null) 
            throw new NullPointerException(); 
        if (remappingFunction == null) 
            throw new NullPointerException(); 
        int hash = hash(key); 
        Node<K,V>[] tab; Node<K,V> first; int n, i; 
        int binCount = 0; 
        TreeNode<K,V> t = null; 
        Node<K,V> old = null; 
        if (size > threshold || (tab = table) == null || 
            (n = tab.length) == 0) 
            n = (tab = resize()).length; 
        if ((first = tab[i = (n - 1) & hash]) != null) { 
            if (first instanceof TreeNode) 
                old = (t = (TreeNode<K,V>)first).getTreeNode(hash, key); 
            else { 
                Node<K,V> e = first; K k; 
                do { 
                    if (e.hash == hash && 
                        ((k = e.key) == key || (key != null && key.equals(k)))) { 
                        old = e; 
                        break; 
                    } 
                    ++binCount; 
                } while ((e = e.next) != null); 
            } 
        } 
        if (old != null) { 
            V v; 
            if (old.value != null) 
                v = remappingFunction.apply(old.value, value); 
            else 
                v = value; 
            if (v != null) { 
                old.value = v; 
                afterNodeAccess(old); 
            } 
            else 
                removeNode(hash, key, null, false, true); 
            return v; 
        } 
        if (value != null) { 
            if (t != null) 
                t.putTreeVal(this, tab, hash, key, value); 
            else { 
                tab[i] = newNode(hash, key, value, first); 
                if (binCount >= TREEIFY_THRESHOLD - 1) 
                    treeifyBin(tab, hash); 
            } 
            ++modCount; 
            ++size; 
            afterNodeInsertion(true); 
        } 
        return value; 
    } 
 
    @Override 
    public void forEach(BiConsumer<? super K, ? super V> action) { 
        Node<K,V>[] tab; 
        if (action == null) 
            throw new NullPointerException(); 
        if (size > 0 && (tab = table) != null) { 
            int mc = modCount; 
            for (int i = 0; i < tab.length; ++i) { 
                for (Node<K,V> e = tab[i]; e != null; e = e.next) 
                    action.accept(e.key, e.value); 
            } 
            if (modCount != mc) 
                throw new ConcurrentModificationException(); 
        } 
    } 
 
    @Override 
    public void replaceAll(BiFunction<? super K, ? super V, ? extends V> function) { 
        Node<K,V>[] tab; 
        if (function == null) 
            throw new NullPointerException(); 
        if (size > 0 && (tab = table) != null) { 
            int mc = modCount; 
            for (int i = 0; i < tab.length; ++i) { 
                for (Node<K,V> e = tab[i]; e != null; e = e.next) { 
                    e.value = function.apply(e.key, e.value); 
                } 
            } 
            if (modCount != mc) 
                throw new ConcurrentModificationException(); 
        } 
    } 
 
    /* ------------------------------------------------------------ */ 
    // Cloning and serialization 
 
    /** 
     * Returns a shallow copy of this <tt>HashMap</tt> instance: the keys and 
     * values themselves are not cloned. 
     * 
     * @return a shallow copy of this map 
     */ 
    @SuppressWarnings("unchecked") 
    @Override 
    public Object clone() { 
        HashMap<K,V> result; 
        try { 
            result = (HashMap<K,V>)super.clone(); 
        } catch (CloneNotSupportedException e) { 
            // this shouldn't happen, since we are Cloneable 
            throw new InternalError(e); 
        } 
        result.reinitialize(); 
        result.putMapEntries(this, false); 
        return result; 
    } 
 
    // These methods are also used when serializing HashSets 
    final float loadFactor() { return loadFactor; } 
    final int capacity() { 
        return (table != null) ? table.length : 
            (threshold > 0) ? threshold : 
            DEFAULT_INITIAL_CAPACITY; 
    } 
 
    /** 
     * Save the state of the <tt>HashMap</tt> instance to a stream (i.e., 
     * serialize it). 
     * 
     * @serialData The <i>capacity</i> of the HashMap (the length of the 
     *             bucket array) is emitted (int), followed by the 
     *             <i>size</i> (an int, the number of key-value 
     *             mappings), followed by the key (Object) and value (Object) 
     *             for each key-value mapping.  The key-value mappings are 
     *             emitted in no particular order. 
     */ 
    private void writeObject(java.io.ObjectOutputStream s) 
        throws IOException { 
        int buckets = capacity(); 
        // Write out the threshold, loadfactor, and any hidden stuff 
        s.defaultWriteObject(); 
        s.writeInt(buckets); 
        s.writeInt(size); 
        internalWriteEntries(s); 
    } 
 
    /** 
     * Reconstitute the {@code HashMap} instance from a stream (i.e., 
     * deserialize it). 
     */ 
    private void readObject(java.io.ObjectInputStream s) 
        throws IOException, ClassNotFoundException { 
        // Read in the threshold (ignored), loadfactor, and any hidden stuff 
        s.defaultReadObject(); 
        reinitialize(); 
        if (loadFactor <= 0 || Float.isNaN(loadFactor)) 
            throw new InvalidObjectException("Illegal load factor: " + 
                                             loadFactor); 
        s.readInt();                // Read and ignore number of buckets 
        int mappings = s.readInt(); // Read number of mappings (size) 
        if (mappings < 0) 
            throw new InvalidObjectException("Illegal mappings count: " + 
                                             mappings); 
        else if (mappings > 0) { // (if zero, use defaults) 
            // Size the table using given load factor only if within 
            // range of 0.25...4.0 
            float lf = Math.min(Math.max(0.25f, loadFactor), 4.0f); 
            float fc = (float)mappings / lf + 1.0f; 
            int cap = ((fc < DEFAULT_INITIAL_CAPACITY) ? 
                       DEFAULT_INITIAL_CAPACITY : 
                       (fc >= MAXIMUM_CAPACITY) ? 
                       MAXIMUM_CAPACITY : 
                       tableSizeFor((int)fc)); 
            float ft = (float)cap * lf; 
            threshold = ((cap < MAXIMUM_CAPACITY && ft < MAXIMUM_CAPACITY) ? 
                         (int)ft : Integer.MAX_VALUE); 
 
            // Check Map.Entry[].class since it's the nearest public type to 
            // what we're actually creating. 
            SharedSecrets.getJavaOISAccess().checkArray(s, Map.Entry[].class, cap); 
            @SuppressWarnings({"rawtypes","unchecked"}) 
            Node<K,V>[] tab = (Node<K,V>[])new Node[cap]; 
            table = tab; 
 
            // Read the keys and values, and put the mappings in the HashMap 
            for (int i = 0; i < mappings; i++) { 
                @SuppressWarnings("unchecked") 
                    K key = (K) s.readObject(); 
                @SuppressWarnings("unchecked") 
                    V value = (V) s.readObject(); 
                putVal(hash(key), key, value, false, false); 
            } 
        } 
    } 
 
    /* ------------------------------------------------------------ */ 
    // iterators 
 
    abstract class HashIterator { 
        Node<K,V> next;        // next entry to return 
        Node<K,V> current;     // current entry 
        int expectedModCount;  // for fast-fail 
        int index;             // current slot 
 
        HashIterator() { 
            expectedModCount = modCount; 
            Node<K,V>[] t = table; 
            current = next = null; 
            index = 0; 
            if (t != null && size > 0) { // advance to first entry 
                do {} while (index < t.length && (next = t[index++]) == null); 
            } 
        } 
 
        public final boolean hasNext() { 
            return next != null; 
        } 
 
        final Node<K,V> nextNode() { 
            Node<K,V>[] t; 
            Node<K,V> e = next; 
            if (modCount != expectedModCount) 
                throw new ConcurrentModificationException(); 
            if (e == null) 
                throw new NoSuchElementException(); 
            if ((next = (current = e).next) == null && (t = table) != null) { 
                do {} while (index < t.length && (next = t[index++]) == null); 
            } 
            return e; 
        } 
 
        public final void remove() { 
            Node<K,V> p = current; 
            if (p == null) 
                throw new IllegalStateException(); 
            if (modCount != expectedModCount) 
                throw new ConcurrentModificationException(); 
            current = null; 
            K key = p.key; 
            removeNode(hash(key), key, null, false, false); 
            expectedModCount = modCount; 
        } 
    } 
 
    final class KeyIterator extends HashIterator 
        implements Iterator<K> { 
        public final K next() { return nextNode().key; } 
    } 
 
    final class ValueIterator extends HashIterator 
        implements Iterator<V> { 
        public final V next() { return nextNode().value; } 
    } 
 
    final class EntryIterator extends HashIterator 
        implements Iterator<Map.Entry<K,V>> { 
        public final Map.Entry<K,V> next() { return nextNode(); } 
    } 
 
    /* ------------------------------------------------------------ */ 
    // spliterators 
 
    static class HashMapSpliterator<K,V> { 
        final HashMap<K,V> map; 
        Node<K,V> current;          // current node 
        int index;                  // current index, modified on advance/split 
        int fence;                  // one past last index 
        int est;                    // size estimate 
        int expectedModCount;       // for comodification checks 
 
        HashMapSpliterator(HashMap<K,V> m, int origin, 
                           int fence, int est, 
                           int expectedModCount) { 
            this.map = m; 
            this.index = origin; 
            this.fence = fence; 
            this.est = est; 
            this.expectedModCount = expectedModCount; 
        } 
 
        final int getFence() { // initialize fence and size on first use 
            int hi; 
            if ((hi = fence) < 0) { 
                HashMap<K,V> m = map; 
                est = m.size; 
                expectedModCount = m.modCount; 
                Node<K,V>[] tab = m.table; 
                hi = fence = (tab == null) ? 0 : tab.length; 
            } 
            return hi; 
        } 
 
        public final long estimateSize() { 
            getFence(); // force init 
            return (long) est; 
        } 
    } 
 
    static final class KeySpliterator<K,V> 
        extends HashMapSpliterator<K,V> 
        implements Spliterator<K> { 
        KeySpliterator(HashMap<K,V> m, int origin, int fence, int est, 
                       int expectedModCount) { 
            super(m, origin, fence, est, expectedModCount); 
        } 
 
        public KeySpliterator<K,V> trySplit() { 
            int hi = getFence(), lo = index, mid = (lo + hi) >>> 1; 
            return (lo >= mid || current != null) ? null : 
                new KeySpliterator<>(map, lo, index = mid, est >>>= 1, 
                                        expectedModCount); 
        } 
 
        public void forEachRemaining(Consumer<? super K> action) { 
            int i, hi, mc; 
            if (action == null) 
                throw new NullPointerException(); 
            HashMap<K,V> m = map; 
            Node<K,V>[] tab = m.table; 
            if ((hi = fence) < 0) { 
                mc = expectedModCount = m.modCount; 
                hi = fence = (tab == null) ? 0 : tab.length; 
            } 
            else 
                mc = expectedModCount; 
            if (tab != null && tab.length >= hi && 
                (i = index) >= 0 && (i < (index = hi) || current != null)) { 
                Node<K,V> p = current; 
                current = null; 
                do { 
                    if (p == null) 
                        p = tab[i++]; 
                    else { 
                        action.accept(p.key); 
                        p = p.next; 
                    } 
                } while (p != null || i < hi); 
                if (m.modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
 
        public boolean tryAdvance(Consumer<? super K> action) { 
            int hi; 
            if (action == null) 
                throw new NullPointerException(); 
            Node<K,V>[] tab = map.table; 
            if (tab != null && tab.length >= (hi = getFence()) && index >= 0) { 
                while (current != null || index < hi) { 
                    if (current == null) 
                        current = tab[index++]; 
                    else { 
                        K k = current.key; 
                        current = current.next; 
                        action.accept(k); 
                        if (map.modCount != expectedModCount) 
                            throw new ConcurrentModificationException(); 
                        return true; 
                    } 
                } 
            } 
            return false; 
        } 
 
        public int characteristics() { 
            return (fence < 0 || est == map.size ? Spliterator.SIZED : 0) | 
                Spliterator.DISTINCT; 
        } 
    } 
 
    static final class ValueSpliterator<K,V> 
        extends HashMapSpliterator<K,V> 
        implements Spliterator<V> { 
        ValueSpliterator(HashMap<K,V> m, int origin, int fence, int est, 
                         int expectedModCount) { 
            super(m, origin, fence, est, expectedModCount); 
        } 
 
        public ValueSpliterator<K,V> trySplit() { 
            int hi = getFence(), lo = index, mid = (lo + hi) >>> 1; 
            return (lo >= mid || current != null) ? null : 
                new ValueSpliterator<>(map, lo, index = mid, est >>>= 1, 
                                          expectedModCount); 
        } 
 
        public void forEachRemaining(Consumer<? super V> action) { 
            int i, hi, mc; 
            if (action == null) 
                throw new NullPointerException(); 
            HashMap<K,V> m = map; 
            Node<K,V>[] tab = m.table; 
            if ((hi = fence) < 0) { 
                mc = expectedModCount = m.modCount; 
                hi = fence = (tab == null) ? 0 : tab.length; 
            } 
            else 
                mc = expectedModCount; 
            if (tab != null && tab.length >= hi && 
                (i = index) >= 0 && (i < (index = hi) || current != null)) { 
                Node<K,V> p = current; 
                current = null; 
                do { 
                    if (p == null) 
                        p = tab[i++]; 
                    else { 
                        action.accept(p.value); 
                        p = p.next; 
                    } 
                } while (p != null || i < hi); 
                if (m.modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
 
        public boolean tryAdvance(Consumer<? super V> action) { 
            int hi; 
            if (action == null) 
                throw new NullPointerException(); 
            Node<K,V>[] tab = map.table; 
            if (tab != null && tab.length >= (hi = getFence()) && index >= 0) { 
                while (current != null || index < hi) { 
                    if (current == null) 
                        current = tab[index++]; 
                    else { 
                        V v = current.value; 
                        current = current.next; 
                        action.accept(v); 
                        if (map.modCount != expectedModCount) 
                            throw new ConcurrentModificationException(); 
                        return true; 
                    } 
                } 
            } 
            return false; 
        } 
 
        public int characteristics() { 
            return (fence < 0 || est == map.size ? Spliterator.SIZED : 0); 
        } 
    } 
 
    static final class EntrySpliterator<K,V> 
        extends HashMapSpliterator<K,V> 
        implements Spliterator<Map.Entry<K,V>> { 
        EntrySpliterator(HashMap<K,V> m, int origin, int fence, int est, 
                         int expectedModCount) { 
            super(m, origin, fence, est, expectedModCount); 
        } 
 
        public EntrySpliterator<K,V> trySplit() { 
            int hi = getFence(), lo = index, mid = (lo + hi) >>> 1; 
            return (lo >= mid || current != null) ? null : 
                new EntrySpliterator<>(map, lo, index = mid, est >>>= 1, 
                                          expectedModCount); 
        } 
 
        public void forEachRemaining(Consumer<? super Map.Entry<K,V>> action) { 
            int i, hi, mc; 
            if (action == null) 
                throw new NullPointerException(); 
            HashMap<K,V> m = map; 
            Node<K,V>[] tab = m.table; 
            if ((hi = fence) < 0) { 
                mc = expectedModCount = m.modCount; 
                hi = fence = (tab == null) ? 0 : tab.length; 
            } 
            else 
                mc = expectedModCount; 
            if (tab != null && tab.length >= hi && 
                (i = index) >= 0 && (i < (index = hi) || current != null)) { 
                Node<K,V> p = current; 
                current = null; 
                do { 
                    if (p == null) 
                        p = tab[i++]; 
                    else { 
                        action.accept(p); 
                        p = p.next; 
                    } 
                } while (p != null || i < hi); 
                if (m.modCount != mc) 
                    throw new ConcurrentModificationException(); 
            } 
        } 
 
        public boolean tryAdvance(Consumer<? super Map.Entry<K,V>> action) { 
            int hi; 
            if (action == null) 
                throw new NullPointerException(); 
            Node<K,V>[] tab = map.table; 
            if (tab != null && tab.length >= (hi = getFence()) && index >= 0) { 
                while (current != null || index < hi) { 
                    if (current == null) 
                        current = tab[index++]; 
                    else { 
                        Node<K,V> e = current; 
                        current = current.next; 
                        action.accept(e); 
                        if (map.modCount != expectedModCount) 
                            throw new ConcurrentModificationException(); 
                        return true; 
                    } 
                } 
            } 
            return false; 
        } 
 
        public int characteristics() { 
            return (fence < 0 || est == map.size ? Spliterator.SIZED : 0) | 
                Spliterator.DISTINCT; 
        } 
    } 
 
    /* ------------------------------------------------------------ */ 
    // LinkedHashMap support 
 
 
    /* 
     * The following package-protected methods are designed to be 
     * overridden by LinkedHashMap, but not by any other subclass. 
     * Nearly all other internal methods are also package-protected 
     * but are declared final, so can be used by LinkedHashMap, view 
     * classes, and HashSet. 
     */ 
 
    // Create a regular (non-tree) node 
    Node<K,V> newNode(int hash, K key, V value, Node<K,V> next) { 
        return new Node<>(hash, key, value, next); 
    } 
 
    // For conversion from TreeNodes to plain nodes 
    Node<K,V> replacementNode(Node<K,V> p, Node<K,V> next) { 
        return new Node<>(p.hash, p.key, p.value, next); 
    } 
 
    // Create a tree bin node 
    TreeNode<K,V> newTreeNode(int hash, K key, V value, Node<K,V> next) { 
        return new TreeNode<>(hash, key, value, next); 
    } 
 
    // For treeifyBin 
    TreeNode<K,V> replacementTreeNode(Node<K,V> p, Node<K,V> next) { 
        return new TreeNode<>(p.hash, p.key, p.value, next); 
    } 
 
    /** 
     * Reset to initial default state.  Called by clone and readObject. 
     */ 
    void reinitialize() { 
        table = null; 
        entrySet = null; 
        keySet = null; 
        values = null; 
        modCount = 0; 
        threshold = 0; 
        size = 0; 
    } 
 
    // Callbacks to allow LinkedHashMap post-actions 
    void afterNodeAccess(Node<K,V> p) { } 
    void afterNodeInsertion(boolean evict) { } 
    void afterNodeRemoval(Node<K,V> p) { } 
 
    // Called only from writeObject, to ensure compatible ordering. 
    void internalWriteEntries(java.io.ObjectOutputStream s) throws IOException { 
        Node<K,V>[] tab; 
        if (size > 0 && (tab = table) != null) { 
            for (int i = 0; i < tab.length; ++i) { 
                for (Node<K,V> e = tab[i]; e != null; e = e.next) { 
                    s.writeObject(e.key); 
                    s.writeObject(e.value); 
                } 
            } 
        } 
    } 
 
    /* ------------------------------------------------------------ */ 
    // Tree bins 
 
    /** 
     * Entry for Tree bins. Extends LinkedHashMap.Entry (which in turn 
     * extends Node) so can be used as extension of either regular or 
     * linked node. 
     */ 
    static final class TreeNode<K,V> extends LinkedHashMap.Entry<K,V> { 
        TreeNode<K,V> parent;  // red-black tree links 
        TreeNode<K,V> left; 
        TreeNode<K,V> right; 
        TreeNode<K,V> prev;    // needed to unlink next upon deletion 
        boolean red; 
        TreeNode(int hash, K key, V val, Node<K,V> next) { 
            super(hash, key, val, next); 
        } 
 
        /** 
         * Returns root of tree containing this node. 
         */ 
        final TreeNode<K,V> root() { 
            for (TreeNode<K,V> r = this, p;;) { 
                if ((p = r.parent) == null) 
                    return r; 
                r = p; 
            } 
        } 
 
        /** 
         * Ensures that the given root is the first node of its bin. 
         */ 
        static <K,V> void moveRootToFront(Node<K,V>[] tab, TreeNode<K,V> root) { 
            int n; 
            if (root != null && tab != null && (n = tab.length) > 0) { 
                int index = (n - 1) & root.hash; 
                TreeNode<K,V> first = (TreeNode<K,V>)tab[index]; 
                if (root != first) { 
                    Node<K,V> rn; 
                    tab[index] = root; 
                    TreeNode<K,V> rp = root.prev; 
                    if ((rn = root.next) != null) 
                        ((TreeNode<K,V>)rn).prev = rp; 
                    if (rp != null) 
                        rp.next = rn; 
                    if (first != null) 
                        first.prev = root; 
                    root.next = first; 
                    root.prev = null; 
                } 
                assert checkInvariants(root); 
            } 
        } 
 
        /** 
         * Finds the node starting at root p with the given hash and key. 
         * The kc argument caches comparableClassFor(key) upon first use 
         * comparing keys. 
         */ 
        final TreeNode<K,V> find(int h, Object k, Class<?> kc) { 
            TreeNode<K,V> p = this; 
            do { 
                int ph, dir; K pk; 
                TreeNode<K,V> pl = p.left, pr = p.right, q; 
                if ((ph = p.hash) > h) 
                    p = pl; 
                else if (ph < h) 
                    p = pr; 
                else if ((pk = p.key) == k || (k != null && k.equals(pk))) 
                    return p; 
                else if (pl == null) 
                    p = pr; 
                else if (pr == null) 
                    p = pl; 
                else if ((kc != null || 
                          (kc = comparableClassFor(k)) != null) && 
                         (dir = compareComparables(kc, k, pk)) != 0) 
                    p = (dir < 0) ? pl : pr; 
                else if ((q = pr.find(h, k, kc)) != null) 
                    return q; 
                else 
                    p = pl; 
            } while (p != null); 
            return null; 
        } 
 
        /** 
         * Calls find for root node. 
         */ 
        final TreeNode<K,V> getTreeNode(int h, Object k) { 
            return ((parent != null) ? root() : this).find(h, k, null); 
        } 
 
        /** 
         * Tie-breaking utility for ordering insertions when equal 
         * hashCodes and non-comparable. We don't require a total 
         * order, just a consistent insertion rule to maintain 
         * equivalence across rebalancings. Tie-breaking further than 
         * necessary simplifies testing a bit. 
         */ 
        static int tieBreakOrder(Object a, Object b) { 
            int d; 
            if (a == null || b == null || 
                (d = a.getClass().getName(). 
                 compareTo(b.getClass().getName())) == 0) 
                d = (System.identityHashCode(a) <= System.identityHashCode(b) ? 
                     -1 : 1); 
            return d; 
        } 
 
        /** 
         * Forms tree of the nodes linked from this node. 
         * @return root of tree 
         */ 
        final void treeify(Node<K,V>[] tab) { 
            TreeNode<K,V> root = null; 
            for (TreeNode<K,V> x = this, next; x != null; x = next) { 
                next = (TreeNode<K,V>)x.next; 
                x.left = x.right = null; 
                if (root == null) { 
                    x.parent = null; 
                    x.red = false; 
                    root = x; 
                } 
                else { 
                    K k = x.key; 
                    int h = x.hash; 
                    Class<?> kc = null; 
                    for (TreeNode<K,V> p = root;;) { 
                        int dir, ph; 
                        K pk = p.key; 
                        if ((ph = p.hash) > h) 
                            dir = -1; 
                        else if (ph < h) 
                            dir = 1; 
                        else if ((kc == null && 
                                  (kc = comparableClassFor(k)) == null) || 
                                 (dir = compareComparables(kc, k, pk)) == 0) 
                            dir = tieBreakOrder(k, pk); 
 
                        TreeNode<K,V> xp = p; 
                        if ((p = (dir <= 0) ? p.left : p.right) == null) { 
                            x.parent = xp; 
                            if (dir <= 0) 
                                xp.left = x; 
                            else 
                                xp.right = x; 
                            root = balanceInsertion(root, x); 
                            break; 
                        } 
                    } 
                } 
            } 
            moveRootToFront(tab, root); 
        } 
 
        /** 
         * Returns a list of non-TreeNodes replacing those linked from 
         * this node. 
         */ 
        final Node<K,V> untreeify(HashMap<K,V> map) { 
            Node<K,V> hd = null, tl = null; 
            for (Node<K,V> q = this; q != null; q = q.next) { 
                Node<K,V> p = map.replacementNode(q, null); 
                if (tl == null) 
                    hd = p; 
                else 
                    tl.next = p; 
                tl = p; 
            } 
            return hd; 
        } 
 
        /** 
         * Tree version of putVal. 
         */ 
        final TreeNode<K,V> putTreeVal(HashMap<K,V> map, Node<K,V>[] tab, 
                                       int h, K k, V v) { 
            Class<?> kc = null; 
            boolean searched = false; 
            TreeNode<K,V> root = (parent != null) ? root() : this; 
            for (TreeNode<K,V> p = root;;) { 
                int dir, ph; K pk; 
                if ((ph = p.hash) > h) 
                    dir = -1; 
                else if (ph < h) 
                    dir = 1; 
                else if ((pk = p.key) == k || (k != null && k.equals(pk))) 
                    return p; 
                else if ((kc == null && 
                          (kc = comparableClassFor(k)) == null) || 
                         (dir = compareComparables(kc, k, pk)) == 0) { 
                    if (!searched) { 
                        TreeNode<K,V> q, ch; 
                        searched = true; 
                        if (((ch = p.left) != null && 
                             (q = ch.find(h, k, kc)) != null) || 
                            ((ch = p.right) != null && 
                             (q = ch.find(h, k, kc)) != null)) 
                            return q; 
                    } 
                    dir = tieBreakOrder(k, pk); 
                } 
 
                TreeNode<K,V> xp = p; 
                if ((p = (dir <= 0) ? p.left : p.right) == null) { 
                    Node<K,V> xpn = xp.next; 
                    TreeNode<K,V> x = map.newTreeNode(h, k, v, xpn); 
                    if (dir <= 0) 
                        xp.left = x; 
                    else 
                        xp.right = x; 
                    xp.next = x; 
                    x.parent = x.prev = xp; 
                    if (xpn != null) 
                        ((TreeNode<K,V>)xpn).prev = x; 
                    moveRootToFront(tab, balanceInsertion(root, x)); 
                    return null; 
                } 
            } 
        } 
 
        /** 
         * Removes the given node, that must be present before this call. 
         * This is messier than typical red-black deletion code because we 
         * cannot swap the contents of an interior node with a leaf 
         * successor that is pinned by "next" pointers that are accessible 
         * independently during traversal. So instead we swap the tree 
         * linkages. If the current tree appears to have too few nodes, 
         * the bin is converted back to a plain bin. (The test triggers 
         * somewhere between 2 and 6 nodes, depending on tree structure). 
         */ 
        final void removeTreeNode(HashMap<K,V> map, Node<K,V>[] tab, 
                                  boolean movable) { 
            int n; 
            if (tab == null || (n = tab.length) == 0) 
                return; 
            int index = (n - 1) & hash; 
            TreeNode<K,V> first = (TreeNode<K,V>)tab[index], root = first, rl; 
            TreeNode<K,V> succ = (TreeNode<K,V>)next, pred = prev; 
            if (pred == null) 
                tab[index] = first = succ; 
            else 
                pred.next = succ; 
            if (succ != null) 
                succ.prev = pred; 
            if (first == null) 
                return; 
            if (root.parent != null) 
                root = root.root(); 
            if (root == null || root.right == null || 
                (rl = root.left) == null || rl.left == null) { 
                tab[index] = first.untreeify(map);  // too small 
                return; 
            } 
            TreeNode<K,V> p = this, pl = left, pr = right, replacement; 
            if (pl != null && pr != null) { 
                TreeNode<K,V> s = pr, sl; 
                while ((sl = s.left) != null) // find successor 
                    s = sl; 
                boolean c = s.red; s.red = p.red; p.red = c; // swap colors 
                TreeNode<K,V> sr = s.right; 
                TreeNode<K,V> pp = p.parent; 
                if (s == pr) { // p was s's direct parent 
                    p.parent = s; 
                    s.right = p; 
                } 
                else { 
                    TreeNode<K,V> sp = s.parent; 
                    if ((p.parent = sp) != null) { 
                        if (s == sp.left) 
                            sp.left = p; 
                        else 
                            sp.right = p; 
                    } 
                    if ((s.right = pr) != null) 
                        pr.parent = s; 
                } 
                p.left = null; 
                if ((p.right = sr) != null) 
                    sr.parent = p; 
                if ((s.left = pl) != null) 
                    pl.parent = s; 
                if ((s.parent = pp) == null) 
                    root = s; 
                else if (p == pp.left) 
                    pp.left = s; 
                else 
                    pp.right = s; 
                if (sr != null) 
                    replacement = sr; 
                else 
                    replacement = p; 
            } 
            else if (pl != null) 
                replacement = pl; 
            else if (pr != null) 
                replacement = pr; 
            else 
                replacement = p; 
            if (replacement != p) { 
                TreeNode<K,V> pp = replacement.parent = p.parent; 
                if (pp == null) 
                    root = replacement; 
                else if (p == pp.left) 
                    pp.left = replacement; 
                else 
                    pp.right = replacement; 
                p.left = p.right = p.parent = null; 
            } 
 
            TreeNode<K,V> r = p.red ? root : balanceDeletion(root, replacement); 
 
            if (replacement == p) {  // detach 
                TreeNode<K,V> pp = p.parent; 
                p.parent = null; 
                if (pp != null) { 
                    if (p == pp.left) 
                        pp.left = null; 
                    else if (p == pp.right) 
                        pp.right = null; 
                } 
            } 
            if (movable) 
                moveRootToFront(tab, r); 
        } 
 
        /** 
         * Splits nodes in a tree bin into lower and upper tree bins, 
         * or untreeifies if now too small. Called only from resize; 
         * see above discussion about split bits and indices. 
         * 
         * @param map the map 
         * @param tab the table for recording bin heads 
         * @param index the index of the table being split 
         * @param bit the bit of hash to split on 
         */ 
        final void split(HashMap<K,V> map, Node<K,V>[] tab, int index, int bit) { 
            TreeNode<K,V> b = this; 
            // Relink into lo and hi lists, preserving order 
            TreeNode<K,V> loHead = null, loTail = null; 
            TreeNode<K,V> hiHead = null, hiTail = null; 
            int lc = 0, hc = 0; 
            for (TreeNode<K,V> e = b, next; e != null; e = next) { 
                next = (TreeNode<K,V>)e.next; 
                e.next = null; 
                if ((e.hash & bit) == 0) { 
                    if ((e.prev = loTail) == null) 
                        loHead = e; 
                    else 
                        loTail.next = e; 
                    loTail = e; 
                    ++lc; 
                } 
                else { 
                    if ((e.prev = hiTail) == null) 
                        hiHead = e; 
                    else 
                        hiTail.next = e; 
                    hiTail = e; 
                    ++hc; 
                } 
            } 
 
            if (loHead != null) { 
                if (lc <= UNTREEIFY_THRESHOLD) 
                    tab[index] = loHead.untreeify(map); 
                else { 
                    tab[index] = loHead; 
                    if (hiHead != null) // (else is already treeified) 
                        loHead.treeify(tab); 
                } 
            } 
            if (hiHead != null) { 
                if (hc <= UNTREEIFY_THRESHOLD) 
                    tab[index + bit] = hiHead.untreeify(map); 
                else { 
                    tab[index + bit] = hiHead; 
                    if (loHead != null) 
                        hiHead.treeify(tab); 
                } 
            } 
        } 
 
        /* ------------------------------------------------------------ */ 
        // Red-black tree methods, all adapted from CLR 
 
        static <K,V> TreeNode<K,V> rotateLeft(TreeNode<K,V> root, 
                                              TreeNode<K,V> p) { 
            TreeNode<K,V> r, pp, rl; 
            if (p != null && (r = p.right) != null) { 
                if ((rl = p.right = r.left) != null) 
                    rl.parent = p; 
                if ((pp = r.parent = p.parent) == null) 
                    (root = r).red = false; 
                else if (pp.left == p) 
                    pp.left = r; 
                else 
                    pp.right = r; 
                r.left = p; 
                p.parent = r; 
            } 
            return root; 
        } 
 
        static <K,V> TreeNode<K,V> rotateRight(TreeNode<K,V> root, 
                                               TreeNode<K,V> p) { 
            TreeNode<K,V> l, pp, lr; 
            if (p != null && (l = p.left) != null) { 
                if ((lr = p.left = l.right) != null) 
                    lr.parent = p; 
                if ((pp = l.parent = p.parent) == null) 
                    (root = l).red = false; 
                else if (pp.right == p) 
                    pp.right = l; 
                else 
                    pp.left = l; 
                l.right = p; 
                p.parent = l; 
            } 
            return root; 
        } 
 
        static <K,V> TreeNode<K,V> balanceInsertion(TreeNode<K,V> root, 
                                                    TreeNode<K,V> x) { 
            x.red = true; 
            for (TreeNode<K,V> xp, xpp, xppl, xppr;;) { 
                if ((xp = x.parent) == null) { 
                    x.red = false; 
                    return x; 
                } 
                else if (!xp.red || (xpp = xp.parent) == null) 
                    return root; 
                if (xp == (xppl = xpp.left)) { 
                    if ((xppr = xpp.right) != null && xppr.red) { 
                        xppr.red = false; 
                        xp.red = false; 
                        xpp.red = true; 
                        x = xpp; 
                    } 
                    else { 
                        if (x == xp.right) { 
                            root = rotateLeft(root, x = xp); 
                            xpp = (xp = x.parent) == null ? null : xp.parent; 
                        } 
                        if (xp != null) { 
                            xp.red = false; 
                            if (xpp != null) { 
                                xpp.red = true; 
                                root = rotateRight(root, xpp); 
                            } 
                        } 
                    } 
                } 
                else { 
                    if (xppl != null && xppl.red) { 
                        xppl.red = false; 
                        xp.red = false; 
                        xpp.red = true; 
                        x = xpp; 
                    } 
                    else { 
                        if (x == xp.left) { 
                            root = rotateRight(root, x = xp); 
                            xpp = (xp = x.parent) == null ? null : xp.parent; 
                        } 
                        if (xp != null) { 
                            xp.red = false; 
                            if (xpp != null) { 
                                xpp.red = true; 
                                root = rotateLeft(root, xpp); 
                            } 
                        } 
                    } 
                } 
            } 
        } 
 
        static <K,V> TreeNode<K,V> balanceDeletion(TreeNode<K,V> root, 
                                                   TreeNode<K,V> x) { 
            for (TreeNode<K,V> xp, xpl, xpr;;)  { 
                if (x == null || x == root) 
                    return root; 
                else if ((xp = x.parent) == null) { 
                    x.red = false; 
                    return x; 
                } 
                else if (x.red) { 
                    x.red = false; 
                    return root; 
                } 
                else if ((xpl = xp.left) == x) { 
                    if ((xpr = xp.right) != null && xpr.red) { 
                        xpr.red = false; 
                        xp.red = true; 
                        root = rotateLeft(root, xp); 
                        xpr = (xp = x.parent) == null ? null : xp.right; 
                    } 
                    if (xpr == null) 
                        x = xp; 
                    else { 
                        TreeNode<K,V> sl = xpr.left, sr = xpr.right; 
                        if ((sr == null || !sr.red) && 
                            (sl == null || !sl.red)) { 
                            xpr.red = true; 
                            x = xp; 
                        } 
                        else { 
                            if (sr == null || !sr.red) { 
                                if (sl != null) 
                                    sl.red = false; 
                                xpr.red = true; 
                                root = rotateRight(root, xpr); 
                                xpr = (xp = x.parent) == null ? 
                                    null : xp.right; 
                            } 
                            if (xpr != null) { 
                                xpr.red = (xp == null) ? false : xp.red; 
                                if ((sr = xpr.right) != null) 
                                    sr.red = false; 
                            } 
                            if (xp != null) { 
                                xp.red = false; 
                                root = rotateLeft(root, xp); 
                            } 
                            x = root; 
                        } 
                    } 
                } 
                else { // symmetric 
                    if (xpl != null && xpl.red) { 
                        xpl.red = false; 
                        xp.red = true; 
                        root = rotateRight(root, xp); 
                        xpl = (xp = x.parent) == null ? null : xp.left; 
                    } 
                    if (xpl == null) 
                        x = xp; 
                    else { 
                        TreeNode<K,V> sl = xpl.left, sr = xpl.right; 
                        if ((sl == null || !sl.red) && 
                            (sr == null || !sr.red)) { 
                            xpl.red = true; 
                            x = xp; 
                        } 
                        else { 
                            if (sl == null || !sl.red) { 
                                if (sr != null) 
                                    sr.red = false; 
                                xpl.red = true; 
                                root = rotateLeft(root, xpl); 
                                xpl = (xp = x.parent) == null ? 
                                    null : xp.left; 
                            } 
                            if (xpl != null) { 
                                xpl.red = (xp == null) ? false : xp.red; 
                                if ((sl = xpl.left) != null) 
                                    sl.red = false; 
                            } 
                            if (xp != null) { 
                                xp.red = false; 
                                root = rotateRight(root, xp); 
                            } 
                            x = root; 
                        } 
                    } 
                } 
            } 
        } 
 
        /** 
         * Recursive invariant check 
         */ 
        static <K,V> boolean checkInvariants(TreeNode<K,V> t) { 
            TreeNode<K,V> tp = t.parent, tl = t.left, tr = t.right, 
                tb = t.prev, tn = (TreeNode<K,V>)t.next; 
            if (tb != null && tb.next != t) 
                return false; 
            if (tn != null && tn.prev != t) 
                return false; 
            if (tp != null && t != tp.left && t != tp.right) 
                return false; 
            if (tl != null && (tl.parent != t || tl.hash > t.hash)) 
                return false; 
            if (tr != null && (tr.parent != t || tr.hash < t.hash)) 
                return false; 
            if (t.red && tl != null && tl.red && tr != null && tr.red) 
                return false; 
            if (tl != null && !checkInvariants(tl)) 
                return false; 
            if (tr != null && !checkInvariants(tr)) 
                return false; 
            return true; 
        } 
    } 
 
}