All Downloads are FREE. Search and download functionalities are using the official Maven repository.

com.badlogic.gdx.utils.ObjectSet Maven / Gradle / Ivy

/*******************************************************************************
 * Copyright 2011 See AUTHORS file.
 *
 * 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 com.badlogic.gdx.utils;

import java.util.Arrays;
import java.util.Iterator;
import java.util.NoSuchElementException;

import com.badlogic.gdx.math.MathUtils;

/** An unordered set where the keys are objects. Null keys are not allowed. No allocation is done except when growing the table
 * size.
 * 

* This class performs fast contains and remove (typically O(1), worst case O(n) but that is rare in practice). Add may be * slightly slower, depending on hash collisions. Hashcodes are rehashed to reduce collisions and the need to resize. Load factors * greater than 0.91 greatly increase the chances to resize to the next higher POT size. *

* Unordered sets and maps are not designed to provide especially fast iteration. Iteration is faster with OrderedSet and * OrderedMap. *

* This implementation uses linear probing with the backward shift algorithm for removal. Hashcodes are rehashed using Fibonacci * hashing, instead of the more common power-of-two mask, to better distribute poor hashCodes (see Malte * Skarupke's blog post). Linear probing continues to work even when all hashCodes collide, just more slowly. * @author Nathan Sweet * @author Tommy Ettinger */ public class ObjectSet implements Iterable { public int size; T[] keyTable; float loadFactor; int threshold; /** Used by {@link #place(Object)} to bit shift the upper bits of a {@code long} into a usable range (>= 0 and <= * {@link #mask}). The shift can be negative, which is convenient to match the number of bits in mask: if mask is a 7-bit * number, a shift of -7 shifts the upper 7 bits into the lowest 7 positions. This class sets the shift > 32 and < 64, * which if used with an int will still move the upper bits of an int to the lower bits due to Java's implicit modulus on * shifts. *

* {@link #mask} can also be used to mask the low bits of a number, which may be faster for some hashcodes, if * {@link #place(Object)} is overridden. */ protected int shift; /** A bitmask used to confine hashcodes to the size of the table. Must be all 1 bits in its low positions, ie a power of two * minus 1. If {@link #place(Object)} is overriden, this can be used instead of {@link #shift} to isolate usable bits of a * hash. */ protected int mask; private ObjectSetIterator iterator1, iterator2; /** Creates a new set with an initial capacity of 51 and a load factor of 0.8. */ public ObjectSet () { this(51, 0.8f); } /** Creates a new set with a load factor of 0.8. * @param initialCapacity If not a power of two, it is increased to the next nearest power of two. */ public ObjectSet (int initialCapacity) { this(initialCapacity, 0.8f); } /** Creates a new set with the specified initial capacity and load factor. This set will hold initialCapacity items before * growing the backing table. * @param initialCapacity If not a power of two, it is increased to the next nearest power of two. */ public ObjectSet (int initialCapacity, float loadFactor) { if (loadFactor <= 0f || loadFactor >= 1f) throw new IllegalArgumentException("loadFactor must be > 0 and < 1: " + loadFactor); this.loadFactor = loadFactor; int tableSize = tableSize(initialCapacity, loadFactor); threshold = (int)(tableSize * loadFactor); mask = tableSize - 1; shift = Long.numberOfLeadingZeros(mask); keyTable = (T[])new Object[tableSize]; } /** Creates a new set identical to the specified set. */ public ObjectSet (ObjectSet set) { this((int)(set.keyTable.length * set.loadFactor), set.loadFactor); System.arraycopy(set.keyTable, 0, keyTable, 0, set.keyTable.length); size = set.size; } /** Returns an index >= 0 and <= {@link #mask} for the specified {@code item}. *

* The default implementation uses Fibonacci hashing on the item's {@link Object#hashCode()}: the hashcode is multiplied by a * long constant (2 to the 64th, divided by the golden ratio) then the uppermost bits are shifted into the lowest positions to * obtain an index in the desired range. Multiplication by a long may be slower than int (eg on GWT) but greatly improves * rehashing, allowing even very poor hashcodes, such as those that only differ in their upper bits, to be used without high * collision rates. Fibonacci hashing has increased collision rates when all or most hashcodes are multiples of larger * Fibonacci numbers (see Malte * Skarupke's blog post). *

* This method can be overriden to customizing hashing. This may be useful eg in the unlikely event that most hashcodes are * Fibonacci numbers, if keys provide poor or incorrect hashcodes, or to simplify hashing if keys provide high quality * hashcodes and don't need Fibonacci hashing: {@code return item.hashCode() & mask;} */ protected int place (T item) { return (int)(item.hashCode() * 0x9E3779B97F4A7C15L >>> shift); } /** Returns the index of the key if already present, else -(index + 1) for the next empty index. This can be overridden in this * pacakge to compare for equality differently than {@link Object#equals(Object)}. */ int locateKey (T key) { if (key == null) throw new IllegalArgumentException("key cannot be null."); T[] keyTable = this.keyTable; for (int i = place(key);; i = i + 1 & mask) { T other = keyTable[i]; if (other == null) return -(i + 1); // Empty space is available. if (other.equals(key)) return i; // Same key was found. } } /** Returns true if the key was not already in the set. If this set already contains the key, the call leaves the set unchanged * and returns false. */ public boolean add (T key) { int i = locateKey(key); if (i >= 0) return false; // Existing key was found. i = -(i + 1); // Empty space was found. keyTable[i] = key; if (++size >= threshold) resize(keyTable.length << 1); return true; } public void addAll (Array array) { addAll(array.items, 0, array.size); } public void addAll (Array array, int offset, int length) { if (offset + length > array.size) throw new IllegalArgumentException("offset + length must be <= size: " + offset + " + " + length + " <= " + array.size); addAll(array.items, offset, length); } public boolean addAll (T... array) { return addAll(array, 0, array.length); } public boolean addAll (T[] array, int offset, int length) { ensureCapacity(length); int oldSize = size; for (int i = offset, n = i + length; i < n; i++) add(array[i]); return oldSize != size; } public void addAll (ObjectSet set) { ensureCapacity(set.size); T[] keyTable = set.keyTable; for (int i = 0, n = keyTable.length; i < n; i++) { T key = keyTable[i]; if (key != null) add(key); } } /** Skips checks for existing keys, doesn't increment size. */ private void addResize (T key) { T[] keyTable = this.keyTable; for (int i = place(key);; i = (i + 1) & mask) { if (keyTable[i] == null) { keyTable[i] = key; return; } } } /** Returns true if the key was removed. */ public boolean remove (T key) { int i = locateKey(key); if (i < 0) return false; T[] keyTable = this.keyTable; int mask = this.mask, next = i + 1 & mask; while ((key = keyTable[next]) != null) { int placement = place(key); if ((next - placement & mask) > (i - placement & mask)) { keyTable[i] = key; i = next; } next = next + 1 & mask; } keyTable[i] = null; size--; return true; } /** Returns true if the set has one or more items. */ public boolean notEmpty () { return size > 0; } /** Returns true if the set is empty. */ public boolean isEmpty () { return size == 0; } /** Reduces the size of the backing arrays to be the specified capacity / loadFactor, or less. If the capacity is already less, * nothing is done. If the set contains more items than the specified capacity, the next highest power of two capacity is used * instead. */ public void shrink (int maximumCapacity) { if (maximumCapacity < 0) throw new IllegalArgumentException("maximumCapacity must be >= 0: " + maximumCapacity); int tableSize = tableSize(maximumCapacity, loadFactor); if (keyTable.length > tableSize) resize(tableSize); } /** Clears the set and reduces the size of the backing arrays to be the specified capacity / loadFactor, if they are larger. * The reduction is done by allocating new arrays, though for large arrays this can be faster than clearing the existing * array. */ public void clear (int maximumCapacity) { int tableSize = tableSize(maximumCapacity, loadFactor); if (keyTable.length <= tableSize) { clear(); return; } size = 0; resize(tableSize); } /** Clears the set, leaving the backing arrays at the current capacity. When the capacity is high and the population is low, * iteration can be unnecessarily slow. {@link #clear(int)} can be used to reduce the capacity. */ public void clear () { if (size == 0) return; size = 0; Arrays.fill(keyTable, null); } public boolean contains (T key) { return locateKey(key) >= 0; } public @Null T get (T key) { int i = locateKey(key); return i < 0 ? null : keyTable[i]; } public T first () { T[] keyTable = this.keyTable; for (int i = 0, n = keyTable.length; i < n; i++) if (keyTable[i] != null) return keyTable[i]; throw new IllegalStateException("ObjectSet is empty."); } /** Increases the size of the backing array to accommodate the specified number of additional items / loadFactor. Useful before * adding many items to avoid multiple backing array resizes. */ public void ensureCapacity (int additionalCapacity) { int tableSize = tableSize(size + additionalCapacity, loadFactor); if (keyTable.length < tableSize) resize(tableSize); } private void resize (int newSize) { int oldCapacity = keyTable.length; threshold = (int)(newSize * loadFactor); mask = newSize - 1; shift = Long.numberOfLeadingZeros(mask); T[] oldKeyTable = keyTable; keyTable = (T[])(new Object[newSize]); if (size > 0) { for (int i = 0; i < oldCapacity; i++) { T key = oldKeyTable[i]; if (key != null) addResize(key); } } } public int hashCode () { int h = size; T[] keyTable = this.keyTable; for (int i = 0, n = keyTable.length; i < n; i++) { T key = keyTable[i]; if (key != null) h += key.hashCode(); } return h; } public boolean equals (Object obj) { if (!(obj instanceof ObjectSet)) return false; ObjectSet other = (ObjectSet)obj; if (other.size != size) return false; T[] keyTable = this.keyTable; for (int i = 0, n = keyTable.length; i < n; i++) if (keyTable[i] != null && !other.contains(keyTable[i])) return false; return true; } public String toString () { return '{' + toString(", ") + '}'; } public String toString (String separator) { if (size == 0) return ""; java.lang.StringBuilder buffer = new java.lang.StringBuilder(32); T[] keyTable = this.keyTable; int i = keyTable.length; while (i-- > 0) { T key = keyTable[i]; if (key == null) continue; buffer.append(key == this ? "(this)" : key); break; } while (i-- > 0) { T key = keyTable[i]; if (key == null) continue; buffer.append(separator); buffer.append(key == this ? "(this)" : key); } return buffer.toString(); } /** Returns an iterator for the keys in the set. Remove is supported. *

* If {@link Collections#allocateIterators} is false, the same iterator instance is returned each time this method is called. * Use the {@link ObjectSetIterator} constructor for nested or multithreaded iteration. */ public ObjectSetIterator iterator () { if (Collections.allocateIterators) return new ObjectSetIterator(this); if (iterator1 == null) { iterator1 = new ObjectSetIterator(this); iterator2 = new ObjectSetIterator(this); } if (!iterator1.valid) { iterator1.reset(); iterator1.valid = true; iterator2.valid = false; return iterator1; } iterator2.reset(); iterator2.valid = true; iterator1.valid = false; return iterator2; } static public ObjectSet with (T... array) { ObjectSet set = new ObjectSet(); set.addAll(array); return set; } static int tableSize (int capacity, float loadFactor) { if (capacity < 0) throw new IllegalArgumentException("capacity must be >= 0: " + capacity); int tableSize = MathUtils.nextPowerOfTwo(Math.max(2, (int)Math.ceil(capacity / loadFactor))); if (tableSize > 1 << 30) throw new IllegalArgumentException("The required capacity is too large: " + capacity); return tableSize; } static public class ObjectSetIterator implements Iterable, Iterator { public boolean hasNext; final ObjectSet set; int nextIndex, currentIndex; boolean valid = true; public ObjectSetIterator (ObjectSet set) { this.set = set; reset(); } public void reset () { currentIndex = -1; nextIndex = -1; findNextIndex(); } private void findNextIndex () { K[] keyTable = set.keyTable; for (int n = set.keyTable.length; ++nextIndex < n;) { if (keyTable[nextIndex] != null) { hasNext = true; return; } } hasNext = false; } public void remove () { int i = currentIndex; if (i < 0) throw new IllegalStateException("next must be called before remove."); K[] keyTable = set.keyTable; int mask = set.mask, next = i + 1 & mask; K key; while ((key = keyTable[next]) != null) { int placement = set.place(key); if ((next - placement & mask) > (i - placement & mask)) { keyTable[i] = key; i = next; } next = next + 1 & mask; } keyTable[i] = null; set.size--; if (i != currentIndex) --nextIndex; currentIndex = -1; } public boolean hasNext () { if (!valid) throw new GdxRuntimeException("#iterator() cannot be used nested."); return hasNext; } public K next () { if (!hasNext) throw new NoSuchElementException(); if (!valid) throw new GdxRuntimeException("#iterator() cannot be used nested."); K key = set.keyTable[nextIndex]; currentIndex = nextIndex; findNextIndex(); return key; } public ObjectSetIterator iterator () { return this; } /** Adds the remaining values to the array. */ public Array toArray (Array array) { while (hasNext) array.add(next()); return array; } /** Returns a new array containing the remaining values. */ public Array toArray () { return toArray(new Array(true, set.size)); } } }





© 2015 - 2024 Weber Informatics LLC | Privacy Policy