it.unimi.dsi.fastutil.objects.ObjectLinkedOpenHashSet Maven / Gradle / Ivy
Show all versions of fastutil-core Show documentation
/*
* Copyright (C) 2002-2022 Sebastiano Vigna
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package it.unimi.dsi.fastutil.objects;
import it.unimi.dsi.fastutil.Hash;
import it.unimi.dsi.fastutil.HashCommon;
import static it.unimi.dsi.fastutil.HashCommon.arraySize;
import static it.unimi.dsi.fastutil.HashCommon.maxFill;
import java.util.Arrays;
import java.util.Collection;
import java.util.Iterator;
import java.util.NoSuchElementException;
import java.util.function.Consumer;
import java.util.stream.Collector;
import java.util.Comparator;
/**
* A type-specific linked hash set with with a fast, small-footprint implementation.
*
*
* Instances of this class use a hash table to represent a set. The table is filled up to a
* specified load factor, and then doubled in size to accommodate new entries. If the table
* is emptied below one fourth of the load factor, it is halved in size; however, the table
* is never reduced to a size smaller than that at creation time: this approach makes it possible to
* create sets with a large capacity in which insertions and deletions do not cause immediately
* rehashing. Moreover, halving is not performed when deleting entries from an iterator, as it would
* interfere with the iteration process.
*
*
* Note that {@link #clear()} does not modify the hash table size. Rather, a family of
* {@linkplain #trim() trimming methods} lets you control the size of the table; this is
* particularly useful if you reuse instances of this class.
*
*
* Iterators generated by this set will enumerate elements in the same order in which they have been
* added to the set (addition of elements already present in the set does not change the iteration
* order). Note that this order has nothing in common with the natural order of the keys. The order
* is kept by means of a doubly linked list, represented via an array of longs parallel to
* the table.
*
*
* This class implements the interface of a sorted set, so to allow easy access of the iteration
* order: for instance, you can get the first element in iteration order with {@code first()}
* without having to create an iterator; however, this class partially violates the
* {@link java.util.SortedSet} contract because all subset methods throw an exception and
* {@link #comparator()} returns always {@code null}.
*
*
* Additional methods, such as {@code addAndMoveToFirst()}, make it easy to use instances of this
* class as a cache (e.g., with LRU policy).
*
*
* The iterators provided by this class are type-specific {@linkplain java.util.ListIterator list
* iterators}, and can be started at any element which is in the set (if the provided
* element is not in the set, a {@link NoSuchElementException} exception will be thrown). If,
* however, the provided element is not the first or last element in the set, the first access to
* the list index will require linear time, as in the worst case the entire set must be scanned in
* iteration order to retrieve the positional index of the starting element. If you use just the
* methods of a type-specific {@link it.unimi.dsi.fastutil.BidirectionalIterator}, however, all
* operations will be performed in constant time.
*
* @see Hash
* @see HashCommon
*/
public class ObjectLinkedOpenHashSet extends AbstractObjectSortedSet implements java.io.Serializable, Cloneable, Hash {
private static final long serialVersionUID = 0L;
private static final boolean ASSERTS = false;
/** The array of keys. */
protected transient K[] key;
/** The mask for wrapping a position counter. */
protected transient int mask;
/** Whether this set contains the null key. */
protected transient boolean containsNull;
/**
* The index of the first entry in iteration order. It is valid iff {@link #size} is nonzero;
* otherwise, it contains -1.
*/
protected transient int first = -1;
/**
* The index of the last entry in iteration order. It is valid iff {@link #size} is nonzero;
* otherwise, it contains -1.
*/
protected transient int last = -1;
/**
* For each entry, the next and the previous entry in iteration order, stored as
* {@code ((prev & 0xFFFFFFFFL) << 32) | (next & 0xFFFFFFFFL)}. The first entry contains predecessor
* -1, and the last entry contains successor -1.
*/
protected transient long[] link;
/**
* The current table size. Note that an additional element is allocated for storing the null key.
*/
protected transient int n;
/** Threshold after which we rehash. It must be the table size times {@link #f}. */
protected transient int maxFill;
/** We never resize below this threshold, which is the construction-time {#n}. */
protected final transient int minN;
/** Number of entries in the set (including the null key, if present). */
protected int size;
/** The acceptable load factor. */
protected final float f;
/**
* Creates a new hash set.
*
*
* The actual table size will be the least power of two greater than {@code expected}/{@code f}.
*
* @param expected the expected number of elements in the hash set.
* @param f the load factor.
*/
@SuppressWarnings("unchecked")
public ObjectLinkedOpenHashSet(final int expected, final float f) {
if (f <= 0 || f >= 1) throw new IllegalArgumentException("Load factor must be greater than 0 and smaller than 1");
if (expected < 0) throw new IllegalArgumentException("The expected number of elements must be nonnegative");
this.f = f;
minN = n = arraySize(expected, f);
mask = n - 1;
maxFill = maxFill(n, f);
key = (K[])new Object[n + 1];
link = new long[n + 1];
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor.
*
* @param expected the expected number of elements in the hash set.
*/
public ObjectLinkedOpenHashSet(final int expected) {
this(expected, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set with initial expected {@link Hash#DEFAULT_INITIAL_SIZE} elements and
* {@link Hash#DEFAULT_LOAD_FACTOR} as load factor.
*/
public ObjectLinkedOpenHashSet() {
this(DEFAULT_INITIAL_SIZE, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set copying a given collection.
*
* @param c a {@link Collection} to be copied into the new hash set.
* @param f the load factor.
*/
public ObjectLinkedOpenHashSet(final Collection c, final float f) {
this(c.size(), f);
addAll(c);
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor copying a given
* collection.
*
* @param c a {@link Collection} to be copied into the new hash set.
*/
public ObjectLinkedOpenHashSet(final Collection c) {
this(c, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set copying a given type-specific collection.
*
* @param c a type-specific collection to be copied into the new hash set.
* @param f the load factor.
*/
public ObjectLinkedOpenHashSet(final ObjectCollection c, final float f) {
this(c.size(), f);
addAll(c);
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor copying a given
* type-specific collection.
*
* @param c a type-specific collection to be copied into the new hash set.
*/
public ObjectLinkedOpenHashSet(final ObjectCollection c) {
this(c, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set using elements provided by a type-specific iterator.
*
* @param i a type-specific iterator whose elements will fill the set.
* @param f the load factor.
*/
public ObjectLinkedOpenHashSet(final Iterator i, final float f) {
this(DEFAULT_INITIAL_SIZE, f);
while (i.hasNext()) add(i.next());
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using elements
* provided by a type-specific iterator.
*
* @param i a type-specific iterator whose elements will fill the set.
*/
public ObjectLinkedOpenHashSet(final Iterator i) {
this(i, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set and fills it with the elements of a given array.
*
* @param a an array whose elements will be used to fill the set.
* @param offset the first element to use.
* @param length the number of elements to use.
* @param f the load factor.
*/
public ObjectLinkedOpenHashSet(final K[] a, final int offset, final int length, final float f) {
this(length < 0 ? 0 : length, f);
ObjectArrays.ensureOffsetLength(a, offset, length);
for (int i = 0; i < length; i++) add(a[offset + i]);
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor and fills it with the
* elements of a given array.
*
* @param a an array whose elements will be used to fill the set.
* @param offset the first element to use.
* @param length the number of elements to use.
*/
public ObjectLinkedOpenHashSet(final K[] a, final int offset, final int length) {
this(a, offset, length, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new hash set copying the elements of an array.
*
* @param a an array to be copied into the new hash set.
* @param f the load factor.
*/
public ObjectLinkedOpenHashSet(final K[] a, final float f) {
this(a, 0, a.length, f);
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor copying the elements
* of an array.
*
* @param a an array to be copied into the new hash set.
*/
public ObjectLinkedOpenHashSet(final K[] a) {
this(a, DEFAULT_LOAD_FACTOR);
}
/**
* Creates a new empty hash set.
*
* @return a new empty hash set.
*/
public static ObjectLinkedOpenHashSet of() {
return new ObjectLinkedOpenHashSet<>();
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using the given
* element.
*
* @param e the element that the returned set will contain.
* @return a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor containing {@code e}.
*/
public static ObjectLinkedOpenHashSet of(final K e) {
ObjectLinkedOpenHashSet result = new ObjectLinkedOpenHashSet<>(1, DEFAULT_LOAD_FACTOR);
result.add(e);
return result;
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using the elements
* given.
*
* @param e0 the first element.
* @param e1 the second element.
* @return a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor containing {@code e0}
* and {@code e1}.
* @throws IllegalArgumentException if there were duplicate entries.
*/
public static ObjectLinkedOpenHashSet of(final K e0, final K e1) {
ObjectLinkedOpenHashSet result = new ObjectLinkedOpenHashSet<>(2, DEFAULT_LOAD_FACTOR);
result.add(e0);
if (!result.add(e1)) {
throw new IllegalArgumentException("Duplicate element: " + e1);
}
return result;
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using the elements
* given.
*
* @param e0 the first element.
* @param e1 the second element.
* @param e2 the third element.
* @return a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor containing
* {@code e0}, {@code e1}, and {@code e2}.
* @throws IllegalArgumentException if there were duplicate entries.
*/
public static ObjectLinkedOpenHashSet of(final K e0, final K e1, final K e2) {
ObjectLinkedOpenHashSet result = new ObjectLinkedOpenHashSet<>(3, DEFAULT_LOAD_FACTOR);
result.add(e0);
if (!result.add(e1)) {
throw new IllegalArgumentException("Duplicate element: " + e1);
}
if (!result.add(e2)) {
throw new IllegalArgumentException("Duplicate element: " + e2);
}
return result;
}
/**
* Creates a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using a list of
* elements.
*
* @param a a list of elements that will be used to initialize the new hash set.
* @return a new hash set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor containing the
* elements of {@code a}.
* @throws IllegalArgumentException if a duplicate entry was encountered.
*/
@SafeVarargs
public static ObjectLinkedOpenHashSet of(final K... a) {
ObjectLinkedOpenHashSet result = new ObjectLinkedOpenHashSet<>(a.length, DEFAULT_LOAD_FACTOR);
for (K element : a) {
if (!result.add(element)) {
throw new IllegalArgumentException("Duplicate element " + element);
}
}
return result;
}
// Collector wants a function that returns the collection being added to.
private ObjectLinkedOpenHashSet combine(ObjectLinkedOpenHashSet toAddFrom) {
addAll(toAddFrom);
return this;
}
private static final Collector> TO_SET_COLLECTOR = Collector.of(ObjectLinkedOpenHashSet::new, ObjectLinkedOpenHashSet::add, ObjectLinkedOpenHashSet::combine);
/** Returns a {@link Collector} that collects a {@code Stream}'s elements into a new hash set. */
@SuppressWarnings({ "unchecked", "rawtypes" })
public static Collector> toSet() {
return (Collector)TO_SET_COLLECTOR;
}
/**
* Returns a {@link Collector} that collects a {@code Stream}'s elements into a new hash set,
* potentially pre-allocated to handle the given size.
*/
public static Collector> toSetWithExpectedSize(int expectedSize) {
if (expectedSize <= Hash.DEFAULT_INITIAL_SIZE) {
// Already below default capacity. Just use all default construction instead of fiddling with
// atomics in SizeDecreasingSupplier
return toSet();
}
return Collector.of(new ObjectCollections.SizeDecreasingSupplier>(expectedSize, (int size) -> size <= Hash.DEFAULT_INITIAL_SIZE ? new ObjectLinkedOpenHashSet() : new ObjectLinkedOpenHashSet(size)), ObjectLinkedOpenHashSet::add, ObjectLinkedOpenHashSet::combine);
}
private int realSize() {
return containsNull ? size - 1 : size;
}
private void ensureCapacity(final int capacity) {
final int needed = arraySize(capacity, f);
if (needed > n) rehash(needed);
}
private void tryCapacity(final long capacity) {
final int needed = (int)Math.min(1 << 30, Math.max(2, HashCommon.nextPowerOfTwo((long)Math.ceil(capacity / f))));
if (needed > n) rehash(needed);
}
@Override
public boolean addAll(Collection c) {
// The resulting collection will be at least c.size() big
if (f <= .5) ensureCapacity(c.size()); // The resulting collection will be sized for c.size() elements
else tryCapacity(size() + c.size()); // The resulting collection will be tentatively sized for size() + c.size()
// elements
return super.addAll(c);
}
@Override
public boolean add(final K k) {
int pos;
if (((k) == null)) {
if (containsNull) return false;
pos = n;
containsNull = true;
} else {
K curr;
final K[] key = this.key;
// The starting point.
if (!((curr = key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) {
if (((curr).equals(k))) return false;
while (!((curr = key[pos = (pos + 1) & mask]) == null)) if (((curr).equals(k))) return false;
}
key[pos] = k;
}
if (size == 0) {
first = last = pos;
// Special case of SET_UPPER_LOWER(link[pos], -1, -1);
link[pos] = -1L;
} else {
link[last] ^= ((link[last] ^ (pos & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[pos] = ((last & 0xFFFFFFFFL) << 32) | (-1 & 0xFFFFFFFFL);
last = pos;
}
if (size++ >= maxFill) rehash(arraySize(size + 1, f));
if (ASSERTS) checkTable();
return true;
}
/**
* Add a random element if not present, get the existing value if already present.
*
* This is equivalent to (but faster than) doing a:
*
*
* K exist = set.get(k);
* if (exist == null) {
* set.add(k);
* exist = k;
* }
*
*/
public K addOrGet(final K k) {
int pos;
if (((k) == null)) {
if (containsNull) return key[n];
pos = n;
containsNull = true;
} else {
K curr;
final K[] key = this.key;
// The starting point.
if (!((curr = key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) {
if (((curr).equals(k))) return curr;
while (!((curr = key[pos = (pos + 1) & mask]) == null)) if (((curr).equals(k))) return curr;
}
key[pos] = k;
}
if (size == 0) {
first = last = pos;
// Special case of SET_UPPER_LOWER(link[pos], -1, -1);
link[pos] = -1L;
} else {
link[last] ^= ((link[last] ^ (pos & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[pos] = ((last & 0xFFFFFFFFL) << 32) | (-1 & 0xFFFFFFFFL);
last = pos;
}
if (size++ >= maxFill) rehash(arraySize(size + 1, f));
if (ASSERTS) checkTable();
return k;
}
/**
* Shifts left entries with the specified hash code, starting at the specified position, and empties
* the resulting free entry.
*
* @param pos a starting position.
*/
protected final void shiftKeys(int pos) {
// Shift entries with the same hash.
int last, slot;
K curr;
final K[] key = this.key;
for (;;) {
pos = ((last = pos) + 1) & mask;
for (;;) {
if (((curr = key[pos]) == null)) {
key[last] = (null);
return;
}
slot = (it.unimi.dsi.fastutil.HashCommon.mix((curr).hashCode())) & mask;
if (last <= pos ? last >= slot || slot > pos : last >= slot && slot > pos) break;
pos = (pos + 1) & mask;
}
key[last] = curr;
fixPointers(pos, last);
}
}
private boolean removeEntry(final int pos) {
size--;
fixPointers(pos);
shiftKeys(pos);
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return true;
}
private boolean removeNullEntry() {
containsNull = false;
key[n] = (null);
size--;
fixPointers(n);
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return true;
}
@SuppressWarnings("unchecked")
@Override
public boolean remove(final Object k) {
if ((((K)k) == null)) {
if (containsNull) return removeNullEntry();
return false;
}
K curr;
final K[] key = this.key;
int pos;
// The starting point.
if (((curr = key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) return false;
if (((k).equals(curr))) return removeEntry(pos);
while (true) {
if (((curr = key[pos = (pos + 1) & mask]) == null)) return false;
if (((k).equals(curr))) return removeEntry(pos);
}
}
@SuppressWarnings("unchecked")
@Override
public boolean contains(final Object k) {
if ((((K)k) == null)) return containsNull;
K curr;
final K[] key = this.key;
int pos;
// The starting point.
if (((curr = key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) return false;
if (((k).equals(curr))) return true;
while (true) {
if (((curr = key[pos = (pos + 1) & mask]) == null)) return false;
if (((k).equals(curr))) return true;
}
}
/**
* Returns the element of this set that is equal to the given key, or {@code null}.
*
* @return the element of this set that is equal to the given key, or {@code null}.
*/
@SuppressWarnings("unchecked")
public K get(final Object k) {
if ((((K)k) == null)) return key[n]; // This is correct independently of the value of containsNull and of the
// set being custom
K curr;
final K[] key = this.key;
int pos;
// The starting point.
if (((curr = key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) return null;
if (((k).equals(curr))) return curr;
// There's always an unused entry.
while (true) {
if (((curr = key[pos = (pos + 1) & mask]) == null)) return null;
if (((k).equals(curr))) return curr;
}
}
/**
* Removes the first key in iteration order.
*
* @return the first key.
* @throws NoSuchElementException is this set is empty.
*/
public K removeFirst() {
if (size == 0) throw new NoSuchElementException();
final int pos = first;
// Abbreviated version of fixPointers(pos)
first = (int)link[pos];
if (0 <= first) {
// Special case of SET_PREV(link[first], -1)
link[first] |= (-1 & 0xFFFFFFFFL) << 32;
}
final K k = key[pos];
size--;
if (((k) == null)) {
containsNull = false;
key[n] = (null);
} else shiftKeys(pos);
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return k;
}
/**
* Removes the the last key in iteration order.
*
* @return the last key.
* @throws NoSuchElementException is this set is empty.
*/
public K removeLast() {
if (size == 0) throw new NoSuchElementException();
final int pos = last;
// Abbreviated version of fixPointers(pos)
last = (int)(link[pos] >>> 32);
if (0 <= last) {
// Special case of SET_NEXT(link[last], -1)
link[last] |= -1 & 0xFFFFFFFFL;
}
final K k = key[pos];
size--;
if (((k) == null)) {
containsNull = false;
key[n] = (null);
} else shiftKeys(pos);
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return k;
}
private void moveIndexToFirst(final int i) {
if (size == 1 || first == i) return;
if (last == i) {
last = (int)(link[i] >>> 32);
// Special case of SET_NEXT(link[last], -1);
link[last] |= -1 & 0xFFFFFFFFL;
} else {
final long linki = link[i];
final int prev = (int)(linki >>> 32);
final int next = (int)linki;
link[prev] ^= ((link[prev] ^ (linki & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[next] ^= ((link[next] ^ (linki & 0xFFFFFFFF00000000L)) & 0xFFFFFFFF00000000L);
}
link[first] ^= ((link[first] ^ ((i & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
link[i] = ((-1 & 0xFFFFFFFFL) << 32) | (first & 0xFFFFFFFFL);
first = i;
}
private void moveIndexToLast(final int i) {
if (size == 1 || last == i) return;
if (first == i) {
first = (int)link[i];
// Special case of SET_PREV(link[first], -1);
link[first] |= (-1 & 0xFFFFFFFFL) << 32;
} else {
final long linki = link[i];
final int prev = (int)(linki >>> 32);
final int next = (int)linki;
link[prev] ^= ((link[prev] ^ (linki & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[next] ^= ((link[next] ^ (linki & 0xFFFFFFFF00000000L)) & 0xFFFFFFFF00000000L);
}
link[last] ^= ((link[last] ^ (i & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[i] = ((last & 0xFFFFFFFFL) << 32) | (-1 & 0xFFFFFFFFL);
last = i;
}
/**
* Adds a key to the set; if the key is already present, it is moved to the first position of the
* iteration order.
*
* @param k the key.
* @return true if the key was not present.
*/
public boolean addAndMoveToFirst(final K k) {
int pos;
if (((k) == null)) {
if (containsNull) {
moveIndexToFirst(n);
return false;
}
containsNull = true;
pos = n;
} else {
// The starting point.
final K key[] = this.key;
pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask;
// There's always an unused entry. TODO
while (!((key[pos]) == null)) {
if (((k).equals(key[pos]))) {
moveIndexToFirst(pos);
return false;
}
pos = (pos + 1) & mask;
}
}
key[pos] = k;
if (size == 0) {
first = last = pos;
// Special case of SET_UPPER_LOWER(link[pos], -1, -1);
link[pos] = -1L;
} else {
link[first] ^= ((link[first] ^ ((pos & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
link[pos] = ((-1 & 0xFFFFFFFFL) << 32) | (first & 0xFFFFFFFFL);
first = pos;
}
if (size++ >= maxFill) rehash(arraySize(size, f));
if (ASSERTS) checkTable();
return true;
}
/**
* Adds a key to the set; if the key is already present, it is moved to the last position of the
* iteration order.
*
* @param k the key.
* @return true if the key was not present.
*/
public boolean addAndMoveToLast(final K k) {
int pos;
if (((k) == null)) {
if (containsNull) {
moveIndexToLast(n);
return false;
}
containsNull = true;
pos = n;
} else {
// The starting point.
final K key[] = this.key;
pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask;
// There's always an unused entry.
while (!((key[pos]) == null)) {
if (((k).equals(key[pos]))) {
moveIndexToLast(pos);
return false;
}
pos = (pos + 1) & mask;
}
}
key[pos] = k;
if (size == 0) {
first = last = pos;
// Special case of SET_UPPER_LOWER(link[pos], -1, -1);
link[pos] = -1L;
} else {
link[last] ^= ((link[last] ^ (pos & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[pos] = ((last & 0xFFFFFFFFL) << 32) | (-1 & 0xFFFFFFFFL);
last = pos;
}
if (size++ >= maxFill) rehash(arraySize(size, f));
if (ASSERTS) checkTable();
return true;
}
/* Removes all elements from this set.
*
* To increase object reuse, this method does not change the table size.
* If you want to reduce the table size, you must use {@link #trim()}.
*
*/
@Override
public void clear() {
if (size == 0) return;
size = 0;
containsNull = false;
Arrays.fill(key, (null));
first = last = -1;
}
@Override
public int size() {
return size;
}
@Override
public boolean isEmpty() {
return size == 0;
}
/**
* Modifies the {@link #link} vector so that the given entry is removed. This method will complete
* in constant time.
*
* @param i the index of an entry.
*/
protected void fixPointers(final int i) {
if (size == 0) {
first = last = -1;
return;
}
if (first == i) {
first = (int)link[i];
if (0 <= first) {
// Special case of SET_PREV(link[first], -1)
link[first] |= (-1 & 0xFFFFFFFFL) << 32;
}
return;
}
if (last == i) {
last = (int)(link[i] >>> 32);
if (0 <= last) {
// Special case of SET_NEXT(link[last], -1)
link[last] |= -1 & 0xFFFFFFFFL;
}
return;
}
final long linki = link[i];
final int prev = (int)(linki >>> 32);
final int next = (int)linki;
link[prev] ^= ((link[prev] ^ (linki & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[next] ^= ((link[next] ^ (linki & 0xFFFFFFFF00000000L)) & 0xFFFFFFFF00000000L);
}
/**
* Modifies the {@link #link} vector for a shift from s to d. This method will complete in constant
* time.
*
* @param s the source position.
* @param d the destination position.
*/
protected void fixPointers(int s, int d) {
if (size == 1) {
first = last = d;
// Special case of SET(link[d], -1, -1)
link[d] = -1L;
return;
}
if (first == s) {
first = d;
link[(int)link[s]] ^= ((link[(int)link[s]] ^ ((d & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
link[d] = link[s];
return;
}
if (last == s) {
last = d;
link[(int)(link[s] >>> 32)] ^= ((link[(int)(link[s] >>> 32)] ^ (d & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[d] = link[s];
return;
}
final long links = link[s];
final int prev = (int)(links >>> 32);
final int next = (int)links;
link[prev] ^= ((link[prev] ^ (d & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[next] ^= ((link[next] ^ ((d & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
link[d] = links;
}
/**
* Returns the first element of this set in iteration order.
*
* @return the first element in iteration order.
*/
@Override
public K first() {
if (size == 0) throw new NoSuchElementException();
return key[first];
}
/**
* Returns the last element of this set in iteration order.
*
* @return the last element in iteration order.
*/
@Override
public K last() {
if (size == 0) throw new NoSuchElementException();
return key[last];
}
/**
* {@inheritDoc}
*
* @implSpec This implementation just throws an {@link UnsupportedOperationException}.
*/
@Override
public ObjectSortedSet tailSet(K from) {
throw new UnsupportedOperationException();
}
/**
* {@inheritDoc}
*
* @implSpec This implementation just throws an {@link UnsupportedOperationException}.
*/
@Override
public ObjectSortedSet headSet(K to) {
throw new UnsupportedOperationException();
}
/**
* {@inheritDoc}
*
* @implSpec This implementation just throws an {@link UnsupportedOperationException}.
*/
@Override
public ObjectSortedSet subSet(K from, K to) {
throw new UnsupportedOperationException();
}
/**
* {@inheritDoc}
*
* @implSpec This implementation just returns {@code null}.
*/
@Override
public Comparator comparator() {
return null;
}
/**
* A list iterator over a linked set.
*
*
* This class provides a list iterator over a linked hash set. The constructor runs in constant
* time.
*/
private final class SetIterator implements ObjectListIterator {
/**
* The entry that will be returned by the next call to {@link java.util.ListIterator#previous()} (or
* {@code null} if no previous entry exists).
*/
int prev = -1;
/**
* The entry that will be returned by the next call to {@link java.util.ListIterator#next()} (or
* {@code null} if no next entry exists).
*/
int next = -1;
/** The last entry that was returned (or -1 if we did not iterate or used {@link #remove()}). */
int curr = -1;
/**
* The current index (in the sense of a {@link java.util.ListIterator}). When -1, we do not know the
* current index.
*/
int index = -1;
SetIterator() {
next = first;
index = 0;
}
SetIterator(K from) {
if (((from) == null)) {
if (ObjectLinkedOpenHashSet.this.containsNull) {
next = (int)link[n];
prev = n;
return;
} else throw new NoSuchElementException("The key " + from + " does not belong to this set.");
}
if (java.util.Objects.equals(key[last], from)) {
prev = last;
index = size;
return;
}
// The starting point.
final K key[] = ObjectLinkedOpenHashSet.this.key;
int pos = (it.unimi.dsi.fastutil.HashCommon.mix((from).hashCode())) & mask;
// There's always an unused entry.
while (!((key[pos]) == null)) {
if (((key[pos]).equals(from))) {
// Note: no valid index known.
next = (int)link[pos];
prev = pos;
return;
}
pos = (pos + 1) & mask;
}
throw new NoSuchElementException("The key " + from + " does not belong to this set.");
}
@Override
public boolean hasNext() {
return next != -1;
}
@Override
public boolean hasPrevious() {
return prev != -1;
}
@Override
public K next() {
if (!hasNext()) throw new NoSuchElementException();
curr = next;
next = (int)link[curr];
prev = curr;
if (index >= 0) index++;
if (ASSERTS) assert curr == n || !((key[curr]) == null) : "Position " + curr + " is not used";
return key[curr];
}
@Override
public K previous() {
if (!hasPrevious()) throw new NoSuchElementException();
curr = prev;
prev = (int)(link[curr] >>> 32);
next = curr;
if (index >= 0) index--;
return key[curr];
}
@Override
public void forEachRemaining(final Consumer action) {
final K key[] = ObjectLinkedOpenHashSet.this.key;
final long link[] = ObjectLinkedOpenHashSet.this.link;
while (next != -1) {
curr = next;
next = (int)link[curr];
prev = curr;
if (index >= 0) index++;
if (ASSERTS) assert curr == n || !((key[curr]) == null) : "Position " + curr + " is not used";
action.accept(key[curr]);
}
}
private final void ensureIndexKnown() {
if (index >= 0) return;
if (prev == -1) {
index = 0;
return;
}
if (next == -1) {
index = size;
return;
}
int pos = first;
index = 1;
while (pos != prev) {
pos = (int)link[pos];
index++;
}
}
@Override
public int nextIndex() {
ensureIndexKnown();
return index;
}
@Override
public int previousIndex() {
ensureIndexKnown();
return index - 1;
}
@Override
public void remove() {
ensureIndexKnown();
if (curr == -1) throw new IllegalStateException();
if (curr == prev) {
/* If the last operation was a next(), we are removing an entry that preceeds
* the current index, and thus we must decrement it. */
index--;
prev = (int)(link[curr] >>> 32);
} else next = (int)link[curr];
size--;
/* Now we manually fix the pointers. Because of our knowledge of next
* and prev, this is going to be faster than calling fixPointers(). */
if (prev == -1) first = next;
else link[prev] ^= ((link[prev] ^ (next & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
if (next == -1) last = prev;
else link[next] ^= ((link[next] ^ ((prev & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
int last, slot, pos = curr;
curr = -1;
if (pos == n) {
ObjectLinkedOpenHashSet.this.containsNull = false;
ObjectLinkedOpenHashSet.this.key[n] = (null);
} else {
K curr;
final K[] key = ObjectLinkedOpenHashSet.this.key;
// We have to horribly duplicate the shiftKeys() code because we need to update next/prev.
for (;;) {
pos = ((last = pos) + 1) & mask;
for (;;) {
if (((curr = key[pos]) == null)) {
key[last] = (null);
return;
}
slot = (it.unimi.dsi.fastutil.HashCommon.mix((curr).hashCode())) & mask;
if (last <= pos ? last >= slot || slot > pos : last >= slot && slot > pos) break;
pos = (pos + 1) & mask;
}
key[last] = curr;
if (next == pos) next = last;
if (prev == pos) prev = last;
fixPointers(pos, last);
}
}
}
}
/**
* Returns a type-specific list iterator on the elements in this set, starting from a given element
* of the set. Please see the class documentation for implementation details.
*
* @param from an element to start from.
* @return a type-specific list iterator starting at the given element.
* @throws IllegalArgumentException if {@code from} does not belong to the set.
*/
@Override
public ObjectListIterator iterator(K from) {
return new SetIterator(from);
}
/**
* Returns a type-specific list iterator on the elements in this set, starting from the first
* element. Please see the class documentation for implementation details.
*
* @return a type-specific list iterator starting at the first element.
*/
@Override
public ObjectListIterator iterator() {
return new SetIterator();
}
private static final int SPLITERATOR_CHARACTERISTICS = ObjectSpliterators.SET_SPLITERATOR_CHARACTERISTICS | java.util.Spliterator.ORDERED;
/**
* {@inheritDoc}
*
*
* There isn't a way to split efficiently while still preserving order for a linked data structure,
* so this implementation is just backed by the iterator. Thus, this spliterator is not well
* optimized for parallel streams.
*
*
* Note, contrary to the specification of {@link java.util.SortedSet}, this spliterator does not,
* report {@link java.util.Spliterator#SORTED}. This is because iteration order is based on
* insertion order, not natural ordering.
*/
@Override
public ObjectSpliterator spliterator() {
return ObjectSpliterators.asSpliterator(iterator(), it.unimi.dsi.fastutil.Size64.sizeOf(this), SPLITERATOR_CHARACTERISTICS);
}
@Override
public void forEach(final Consumer action) {
int curr;
int next = first;
while (next != -1) {
curr = next;
next = (int)link[curr];
if (ASSERTS) assert curr == n || !((key[curr]) == null) : "Position " + curr + " is not used";
action.accept(key[curr]);
}
}
/**
* Rehashes this set, making the table as small as possible.
*
*
* This method rehashes the table to the smallest size satisfying the load factor. It can be used
* when the set will not be changed anymore, so to optimize access speed and size.
*
*
* If the table size is already the minimum possible, this method does nothing.
*
* @return true if there was enough memory to trim the set.
* @see #trim(int)
*/
public boolean trim() {
return trim(size);
}
/**
* Rehashes this set if the table is too large.
*
*
* Let N be the smallest table size that can hold max(n,{@link #size()})
* entries, still satisfying the load factor. If the current table size is smaller than or equal to
* N, this method does nothing. Otherwise, it rehashes this set in a table of size
* N.
*
*
* This method is useful when reusing sets. {@linkplain #clear() Clearing a set} leaves the table
* size untouched. If you are reusing a set many times, you can call this method with a typical size
* to avoid keeping around a very large table just because of a few large transient sets.
*
* @param n the threshold for the trimming.
* @return true if there was enough memory to trim the set.
* @see #trim()
*/
public boolean trim(final int n) {
final int l = HashCommon.nextPowerOfTwo((int)Math.ceil(n / f));
if (l >= this.n || size > maxFill(l, f)) return true;
try {
rehash(l);
} catch (OutOfMemoryError cantDoIt) {
return false;
}
return true;
}
/**
* Rehashes the set.
*
*
* This method implements the basic rehashing strategy, and may be overriden by subclasses
* implementing different rehashing strategies (e.g., disk-based rehashing). However, you should not
* override this method unless you understand the internal workings of this class.
*
* @param newN the new size
*/
@SuppressWarnings("unchecked")
protected void rehash(final int newN) {
final K key[] = this.key;
final int mask = newN - 1; // Note that this is used by the hashing macro
final K newKey[] = (K[])new Object[newN + 1];
int i = first, prev = -1, newPrev = -1, t, pos;
final long link[] = this.link;
final long newLink[] = new long[newN + 1];
first = -1;
for (int j = size; j-- != 0;) {
if (((key[i]) == null)) pos = newN;
else {
pos = (it.unimi.dsi.fastutil.HashCommon.mix((key[i]).hashCode())) & mask;
while (!((newKey[pos]) == null)) pos = (pos + 1) & mask;
}
newKey[pos] = key[i];
if (prev != -1) {
newLink[newPrev] ^= ((newLink[newPrev] ^ (pos & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
newLink[pos] ^= ((newLink[pos] ^ ((newPrev & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
newPrev = pos;
} else {
newPrev = first = pos;
// Special case of SET(newLink[pos], -1, -1);
newLink[pos] = -1L;
}
t = i;
i = (int)link[i];
prev = t;
}
this.link = newLink;
this.last = newPrev;
if (newPrev != -1)
// Special case of SET_NEXT(newLink[newPrev], -1);
newLink[newPrev] |= -1 & 0xFFFFFFFFL;
n = newN;
this.mask = mask;
maxFill = maxFill(n, f);
this.key = newKey;
}
/**
* Returns a deep copy of this set.
*
*
* This method performs a deep copy of this hash set; the data stored in the set, however, is not
* cloned. Note that this makes a difference only for object keys.
*
* @return a deep copy of this set.
*/
@Override
@SuppressWarnings("unchecked")
public ObjectLinkedOpenHashSet clone() {
ObjectLinkedOpenHashSet c;
try {
c = (ObjectLinkedOpenHashSet)super.clone();
} catch (CloneNotSupportedException cantHappen) {
throw new InternalError();
}
c.key = key.clone();
c.containsNull = containsNull;
c.link = link.clone();
return c;
}
/**
* Returns a hash code for this set.
*
* This method overrides the generic method provided by the superclass. Since {@code equals()} is
* not overriden, it is important that the value returned by this method is the same value as the
* one returned by the overriden method.
*
* @return a hash code for this set.
*/
@Override
public int hashCode() {
int h = 0;
for (int j = realSize(), i = 0; j-- != 0;) {
while (((key[i]) == null)) i++;
if (this != key[i]) h += ((key[i]).hashCode());
i++;
}
// Zero / null have hash zero.
return h;
}
private void writeObject(java.io.ObjectOutputStream s) throws java.io.IOException {
final ObjectIterator i = iterator();
s.defaultWriteObject();
for (int j = size; j-- != 0;) s.writeObject(i.next());
}
@SuppressWarnings("unchecked")
private void readObject(java.io.ObjectInputStream s) throws java.io.IOException, ClassNotFoundException {
s.defaultReadObject();
n = arraySize(size, f);
maxFill = maxFill(n, f);
mask = n - 1;
final K key[] = this.key = (K[])new Object[n + 1];
final long link[] = this.link = new long[n + 1];
int prev = -1;
first = last = -1;
K k;
for (int i = size, pos; i-- != 0;) {
k = (K)s.readObject();
if (((k) == null)) {
pos = n;
containsNull = true;
} else {
if (!((key[pos = (it.unimi.dsi.fastutil.HashCommon.mix((k).hashCode())) & mask]) == null)) while (!((key[pos = (pos + 1) & mask]) == null));
}
key[pos] = k;
if (first != -1) {
link[prev] ^= ((link[prev] ^ (pos & 0xFFFFFFFFL)) & 0xFFFFFFFFL);
link[pos] ^= ((link[pos] ^ ((prev & 0xFFFFFFFFL) << 32)) & 0xFFFFFFFF00000000L);
prev = pos;
} else {
prev = first = pos;
// Special case of SET_PREV(newLink[pos], -1);
link[pos] |= (-1L & 0xFFFFFFFFL) << 32;
}
}
last = prev;
if (prev != -1)
// Special case of SET_NEXT(link[prev], -1);
link[prev] |= -1 & 0xFFFFFFFFL;
if (ASSERTS) checkTable();
}
private void checkTable() {
}
}