* * It uses a hash-bucket approach; that is, hash collisions are handled * by linking the new node off of the pre-existing node (or list of * nodes). In this manner, techniques such as linear probing (which * can cause primary clustering) and rehashing (which does not fit very * well with Java's method of precomputing hash codes) are avoided. In * addition, this maintains a doubly-linked list which tracks either * insertion or access order. *
*
* In insertion order, calling put adds the key to the end of
* traversal, unless the key was already in the map; changing traversal order
* requires removing and reinserting a key. On the other hand, in access
* order, all calls to put and get cause the
* accessed key to move to the end of the traversal list. Note that any
* accesses to the map's contents via its collection views and iterators do
* not affect the map's traversal order, since the collection views do not
* call put or get.
*
*
* One of the nice features of tracking insertion order is that you can
* copy a hashtable, and regardless of the implementation of the original,
* produce the same results when iterating over the copy. This is possible
* without needing the overhead of TreeMap.
*
* * When using this {@link #LinkedHashMap(int, float, boolean) constructor}, * you can build an access-order mapping. This can be used to implement LRU * caches, for example. By overriding {@link #removeEldestEntry(Map.Entry)}, * you can also control the removal of the oldest entry, and thereby do * things like keep the map at a fixed size. *
*
* Under ideal circumstances (no collisions), LinkedHashMap offers O(1)
* performance on most operations (containsValue() is,
* of course, O(n)). In the worst case (all keys map to the same
* hash code -- very unlikely), most operations are O(n). Traversal is
* faster than in HashMap (proportional to the map size, and not the space
* allocated for the map), but other operations may be slower because of the
* overhead of the maintaining the traversal order list.
*
*
* LinkedHashMap accepts the null key and null values. It is not
* synchronized, so if you need multi-threaded access, consider using:
* Map m = Collections.synchronizedMap(new LinkedHashMap(...));
*
*
* The iterators are fail-fast, meaning that any structural
* modification, except for
*
* Every element in Map m will be put into this new HashMap, in the
* order of m's iterator.
*
* @param m a Map whose key / value pairs will be put into
* the new HashMap. NOTE: key / value pairs
* are not cloned in this constructor.
* @throws NullPointerException if m is null
*/
public LinkedHashMap(Map m)
{
super(m);
accessOrder = false;
}
/**
* Construct a new insertion-ordered LinkedHashMap with a specific
* inital capacity and default load factor of 0.75.
*
* @param initialCapacity the initial capacity of this HashMap (>= 0)
* @throws IllegalArgumentException if (initialCapacity < 0)
*/
public LinkedHashMap(int initialCapacity)
{
super(initialCapacity);
accessOrder = false;
}
/**
* Construct a new insertion-orderd LinkedHashMap with a specific
* inital capacity and load factor.
*
* @param initialCapacity the initial capacity (>= 0)
* @param loadFactor the load factor (> 0, not NaN)
* @throws IllegalArgumentException if (initialCapacity < 0) ||
* ! (loadFactor > 0.0)
*/
public LinkedHashMap(int initialCapacity, float loadFactor)
{
super(initialCapacity, loadFactor);
accessOrder = false;
}
/**
* Construct a new LinkedHashMap with a specific inital capacity, load
* factor, and ordering mode.
*
* @param initialCapacity the initial capacity (>=0)
* @param loadFactor the load factor (>0, not NaN)
* @param accessOrder true for access-order, false for insertion-order
* @throws IllegalArgumentException if (initialCapacity < 0) ||
* ! (loadFactor > 0.0)
*/
public LinkedHashMap(int initialCapacity, float loadFactor,
boolean accessOrder)
{
super(initialCapacity, loadFactor);
this.accessOrder = accessOrder;
}
/**
* Clears the Map so it has no keys. This is O(1).
*/
public void clear()
{
super.clear();
root = null;
}
/**
* Returns
*
* For example, to keep the Map limited to 100 entries, override as follows:
*
*
* Typically, this method does not modify the map, but just uses the
* return value as an indication to
*
* This method is called after the eldest entry has been inserted, so
* if remove() called on the iterator
* itself, cause the iterator to throw a
* {@link ConcurrentModificationException} rather than exhibit
* non-deterministic behavior.
*
* @author Eric Blake (ebb9@email.byu.edu)
* @author Tom Tromey (tromey@redhat.com)
* @author Andrew John Hughes (gnu_andrew@member.fsf.org)
* @see Object#hashCode()
* @see Collection
* @see Map
* @see HashMap
* @see TreeMap
* @see Hashtable
* @since 1.4
* @status updated to 1.4
*/
public class LinkedHashMaptrue for
* access-order, false for insertion-order.
*
* @serial true for access order traversal
*/
final boolean accessOrder;
/**
* Class to represent an entry in the hash table. Holds a single key-value
* pair and the doubly-linked insertion order list.
*/
class LinkedHashEntrym or
* the default of 11.
* true if this HashMap contains a value
* o, such that o.equals(value).
*
* @param value the value to search for in this HashMap
* @return true if at least one key maps to the value
*/
public boolean containsValue(Object value)
{
LinkedHashEntry e = root;
while (e != null)
{
if (equals(value, e.value))
return true;
e = e.succ;
}
return false;
}
/**
* Return the value in this Map associated with the supplied key,
* or null if the key maps to nothing. If this is an
* access-ordered Map and the key is found, this performs structural
* modification, moving the key to the newest end of the list. NOTE:
* Since the value could also be null, you must use containsKey to
* see if this key actually maps to something.
*
* @param key the key for which to fetch an associated value
* @return what the key maps to, if present
* @see #put(Object, Object)
* @see #containsKey(Object)
*/
public V get(Object key)
{
int idx = hash(key);
HashEntrytrue if this map should remove the eldest entry.
* This method is invoked by all calls to put and
* putAll which place a new entry in the map, providing
* the implementer an opportunity to remove the eldest entry any time
* a new one is added. This can be used to save memory usage of the
* hashtable, as well as emulating a cache, by deleting stale entries.
*
* private static final int MAX_ENTRIES = 100;
* protected boolean removeEldestEntry(Map.Entry eldest)
* {
* return size() > MAX_ENTRIES;
* }
* put whether to proceed.
* However, if you override it to modify the map, you must return false
* (indicating that put should leave the modified map alone),
* or you face unspecified behavior. Remember that in access-order mode,
* even calling get is a structural modification, but using
* the collections views (such as keySet) is not.
* put was called on a previously empty map, the eldest
* entry is the one you just put in! The default implementation just
* returns false, so that this map always behaves like
* a normal one with unbounded growth.
*
* @param eldest the eldest element which would be removed if this
* returns true. For an access-order map, this is the least
* recently accessed; for an insertion-order map, this is the
* earliest element inserted.
* @return true if eldest should be removed
*/
protected boolean removeEldestEntry(Map.Entryput, which creates and adds a
* new Entry, followed by performing bookkeeping (like removeEldestEntry).
*
* @param key the key of the new Entry
* @param value the value
* @param idx the index in buckets where the new Entry belongs
* @param callRemove whether to call the removeEldestEntry method
* @see #put(Object, Object)
* @see #removeEldestEntry(Map.Entry)
* @see LinkedHashEntry#LinkedHashEntry(Object, Object)
*/
void addEntry(K key, V value, int idx, boolean callRemove)
{
LinkedHashEntry e = new LinkedHashEntry(key, value);
e.next = buckets[idx];
buckets[idx] = e;
if (callRemove && removeEldestEntry(root))
remove(root.key);
}
/**
* Helper method, called by clone() to reset the doubly-linked list.
*
* @param m the map to add entries from
* @see #clone()
*/
void putAllInternal(Map m)
{
root = null;
super.putAllInternal(m);
}
/**
* Generates a parameterized iterator. This allows traversal to follow
* the doubly-linked list instead of the random bin order of HashMap.
*
* @param type {@link #KEYS}, {@link #VALUES}, or {@link #ENTRIES}
* @return the appropriate iterator
*/
Iterator iterator(final int type)
{
return new Iterator()
{
/** The current Entry. */
LinkedHashEntry current = root;
/** The previous Entry returned by next(). */
LinkedHashEntry last;
/** The number of known modifications to the backing Map. */
int knownMod = modCount;
/**
* Returns true if the Iterator has more elements.
*
* @return true if there are more elements
*/
public boolean hasNext()
{
return current != null;
}
/**
* Returns the next element in the Iterator's sequential view.
*
* @return the next element
* @throws ConcurrentModificationException if the HashMap was modified
* @throws NoSuchElementException if there is none
*/
public Object next()
{
if (knownMod != modCount)
throw new ConcurrentModificationException();
if (current == null)
throw new NoSuchElementException();
last = current;
current = current.succ;
return type == VALUES ? last.value : type == KEYS ? last.key : last;
}
/**
* Removes from the backing HashMap the last element which was fetched
* with the next() method.
*
* @throws ConcurrentModificationException if the HashMap was modified
* @throws IllegalStateException if called when there is no last element
*/
public void remove()
{
if (knownMod != modCount)
throw new ConcurrentModificationException();
if (last == null)
throw new IllegalStateException();
LinkedHashMap.this.remove(last.key);
last = null;
knownMod++;
}
};
}
} // class LinkedHashMap