org.apache.hadoop.util.Sets Maven / Gradle / Ivy
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you 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 org.apache.hadoop.util;
import org.apache.hadoop.classification.InterfaceAudience;
import java.util.Arrays;
import java.util.Collection;
import java.util.Collections;
import java.util.EnumSet;
import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Set;
import java.util.SortedSet;
import java.util.TreeSet;
import java.util.concurrent.ConcurrentHashMap;
/**
* Static utility methods pertaining to {@link Set} instances.
* This class is Hadoop's internal use alternative to Guava's Sets
* utility class.
* Javadocs for majority of APIs in this class are taken from Guava's Sets
* class from Guava release version 27.0-jre.
*/
@InterfaceAudience.Private
public final class Sets {
private static final int MAX_POWER_OF_TWO = 1 << (Integer.SIZE - 2);
private Sets() {
// empty
}
/**
* Creates a mutable, initially empty {@code HashSet} instance.
*
* Note: if mutability is not required, use ImmutableSet#of()
* instead. If {@code E} is an {@link Enum} type, use {@link EnumSet#noneOf}
* instead. Otherwise, strongly consider using a {@code LinkedHashSet}
* instead, at the cost of increased memory footprint, to get
* deterministic iteration behavior.
*
* @param Generics Type E.
* @return a new, empty {@code TreeSet}
*/
public static HashSet newHashSet() {
return new HashSet();
}
/**
* Creates a mutable, empty {@code TreeSet} instance sorted by the
* natural sort ordering of its elements.
*
* Note: if mutability is not required, use ImmutableSortedSet#of()
* instead.
*
* @param Generics Type E
* @return a new, empty {@code TreeSet}
*/
public static TreeSet newTreeSet() {
return new TreeSet();
}
/**
* Creates a mutable {@code HashSet} instance initially containing
* the given elements.
*
* Note: if elements are non-null and won't be added or removed
* after this point, use ImmutableSet#of() or ImmutableSet#copyOf(Object[])
* instead. If {@code E} is an {@link Enum} type, use
* {@link EnumSet#of(Enum, Enum[])} instead. Otherwise, strongly consider
* using a {@code LinkedHashSet} instead, at the cost of increased memory
* footprint, to get deterministic iteration behavior.
*
* This method is just a small convenience, either for
* {@code newHashSet(}{@link Arrays#asList}{@code (...))}, or for creating an
* empty set then calling {@link Collections#addAll}.
*
* @param Generics Type E.
* @param elements the elements that the set should contain.
* @return a new, empty thread-safe {@code Set}
*/
@SafeVarargs
public static HashSet newHashSet(E... elements) {
HashSet set = newHashSetWithExpectedSize(elements.length);
Collections.addAll(set, elements);
return set;
}
/**
* Creates a mutable {@code HashSet} instance containing the given
* elements. A very thin convenience for creating an empty set then calling
* {@link Collection#addAll} or Iterables#addAll.
*
* Note: if mutability is not required and the elements are
* non-null, use ImmutableSet#copyOf(Iterable) instead. (Or, change
* {@code elements} to be a FluentIterable and call {@code elements.toSet()}.)
*
* Note: if {@code E} is an {@link Enum} type, use
* newEnumSet(Iterable, Class) instead.
*
* @param Generics Type E.
* @param elements the elements that the set should contain.
* @return a new, empty thread-safe {@code Set}.
*/
public static HashSet newHashSet(Iterable elements) {
return (elements instanceof Collection)
? new HashSet(cast(elements))
: newHashSet(elements.iterator());
}
/**
* Creates a mutable {@code TreeSet} instance containing the given
* elements sorted by their natural ordering.
*
* Note: if mutability is not required, use
* ImmutableSortedSet#copyOf(Iterable) instead.
*
*
Note: If {@code elements} is a {@code SortedSet} with an
* explicit comparator, this method has different behavior than
* {@link TreeSet#TreeSet(SortedSet)}, which returns a {@code TreeSet}
* with that comparator.
*
*
Note for Java 7 and later: this method is now unnecessary and
* should be treated as deprecated. Instead, use the {@code TreeSet}
* constructor directly, taking advantage of the new
* "diamond" syntax.
*
*
This method is just a small convenience for creating an empty set and
* then calling Iterables#addAll. This method is not very useful and will
* likely be deprecated in the future.
*
* @param Generics Type E.
* @param elements the elements that the set should contain
* @return a new {@code TreeSet} containing those elements (minus duplicates)
*/
public static TreeSet newTreeSet(
Iterable elements) {
TreeSet set = newTreeSet();
addAll(set, elements);
return set;
}
private static boolean addAll(TreeSet addTo,
Iterable elementsToAdd) {
if (elementsToAdd instanceof Collection) {
Collection c = cast(elementsToAdd);
return addTo.addAll(c);
}
if (elementsToAdd == null) {
throw new NullPointerException();
}
return addAll(addTo, elementsToAdd.iterator());
}
/**
* Creates a mutable {@code HashSet} instance containing the given
* elements. A very thin convenience for creating an empty set and then
* calling Iterators#addAll.
*
* Note: if mutability is not required and the elements are
* non-null, use ImmutableSet#copyOf(Iterator) instead.
*
* Note: if {@code E} is an {@link Enum} type, you should create
* an {@link EnumSet} instead.
*
* Overall, this method is not very useful and will likely be deprecated
* in the future.
*
* @param Generics Type E.
* @param elements elements.
* @return a new, empty thread-safe {@code Set}.
*/
public static HashSet newHashSet(Iterator elements) {
HashSet set = newHashSet();
addAll(set, elements);
return set;
}
/**
* Returns a new hash set using the smallest initial table size that can hold
* {@code expectedSize} elements without resizing. Note that this is not what
* {@link HashSet#HashSet(int)} does, but it is what most users want and
* expect it to do.
*
* This behavior can't be broadly guaranteed, but has been tested with
* OpenJDK 1.7 and 1.8.
*
* @param expectedSize the number of elements you expect to add to the
* returned set
* @param Generics Type E.
* @return a new, empty hash set with enough capacity to hold
* {@code expectedSize} elements without resizing
* @throws IllegalArgumentException if {@code expectedSize} is negative
*/
public static HashSet newHashSetWithExpectedSize(int expectedSize) {
return new HashSet(capacity(expectedSize));
}
private static Collection cast(Iterable iterable) {
return (Collection) iterable;
}
private static boolean addAll(Collection addTo,
Iterator iterator) {
if (addTo == null) {
throw new NullPointerException();
}
if (iterator == null) {
throw new NullPointerException();
}
boolean wasModified = false;
while (iterator.hasNext()) {
wasModified |= addTo.add(iterator.next());
}
return wasModified;
}
/**
* Returns the intersection of two sets as an unmodifiable set.
* The returned set contains all elements that are contained by both backing
* sets.
*
* Results are undefined if {@code set1} and {@code set2} are sets based
* on different equivalence relations (as {@code HashSet}, {@code TreeSet},
* and the keySet of an {@code IdentityHashMap} all are).
*
* @param set1 set1.
* @param set2 set2.
* @param Generics Type E.
* @return a new, empty thread-safe {@code Set}.
*/
public static Set intersection(final Set set1,
final Set set2) {
if (set1 == null) {
throw new NullPointerException("set1");
}
if (set2 == null) {
throw new NullPointerException("set2");
}
Set newSet = new HashSet<>(set1);
newSet.retainAll(set2);
return Collections.unmodifiableSet(newSet);
}
/**
* Returns the union of two sets as an unmodifiable set.
* The returned set contains all elements that are contained in either
* backing set.
*
* Results are undefined if {@code set1} and {@code set2} are sets
* based on different equivalence relations (as {@link HashSet},
* {@link TreeSet}, and the {@link Map#keySet} of an
* {@code IdentityHashMap} all are).
*
* @param set1 set1.
* @param set2 set2.
* @param Generics Type E.
* @return a new, empty thread-safe {@code Set}.
*/
public static Set union(
final Set set1, final Set set2) {
if (set1 == null) {
throw new NullPointerException("set1");
}
if (set2 == null) {
throw new NullPointerException("set2");
}
Set newSet = new HashSet<>(set1);
newSet.addAll(set2);
return Collections.unmodifiableSet(newSet);
}
/**
* Returns the difference of two sets as an unmodifiable set.
* The returned set contains all elements that are contained by {@code set1}
* and not contained by {@code set2}.
*
* Results are undefined if {@code set1} and {@code set2} are sets based
* on different equivalence relations (as {@code HashSet}, {@code TreeSet},
* and the keySet of an {@code IdentityHashMap} all are).
*
* This method is used to find difference for HashSets. For TreeSets with
* strict order requirement, recommended method is
* {@link #differenceInTreeSets(Set, Set)}.
*
* @param set1 set1.
* @param set2 set2.
* @param Generics Type E.
* @return a new, empty thread-safe {@code Set}.
*/
public static Set difference(
final Set set1, final Set set2) {
if (set1 == null) {
throw new NullPointerException("set1");
}
if (set2 == null) {
throw new NullPointerException("set2");
}
Set newSet = new HashSet<>(set1);
newSet.removeAll(set2);
return Collections.unmodifiableSet(newSet);
}
/**
* Returns the difference of two sets as an unmodifiable set.
* The returned set contains all elements that are contained by {@code set1}
* and not contained by {@code set2}.
*
* Results are undefined if {@code set1} and {@code set2} are sets based
* on different equivalence relations (as {@code HashSet}, {@code TreeSet},
* and the keySet of an {@code IdentityHashMap} all are).
*
* This method is used to find difference for TreeSets. For HashSets,
* recommended method is {@link #difference(Set, Set)}.
*
* @param Generics Type E.
* @param set1 set1.
* @param set2 set2.
* @return a new, empty thread-safe {@code Set}.
*/
public static Set differenceInTreeSets(
final Set set1, final Set set2) {
if (set1 == null) {
throw new NullPointerException("set1");
}
if (set2 == null) {
throw new NullPointerException("set2");
}
Set newSet = new TreeSet<>(set1);
newSet.removeAll(set2);
return Collections.unmodifiableSet(newSet);
}
/**
* Returns the symmetric difference of two sets as an unmodifiable set.
* The returned set contains all elements that are contained in either
* {@code set1} or {@code set2} but not in both. The iteration order of the
* returned set is undefined.
*
* Results are undefined if {@code set1} and {@code set2} are sets based
* on different equivalence relations (as {@code HashSet}, {@code TreeSet},
* and the keySet of an {@code IdentityHashMap} all are).
*
* @param set1 set1.
* @param set2 set2.
* @param Generics Type E.
* @return a new, empty thread-safe {@code Set}.
*/
public static Set symmetricDifference(
final Set set1, final Set set2) {
if (set1 == null) {
throw new NullPointerException("set1");
}
if (set2 == null) {
throw new NullPointerException("set2");
}
Set intersection = new HashSet<>(set1);
intersection.retainAll(set2);
Set symmetricDifference = new HashSet<>(set1);
symmetricDifference.addAll(set2);
symmetricDifference.removeAll(intersection);
return Collections.unmodifiableSet(symmetricDifference);
}
/**
* Creates a thread-safe set backed by a hash map. The set is backed by a
* {@link ConcurrentHashMap} instance, and thus carries the same concurrency
* guarantees.
*
* Unlike {@code HashSet}, this class does NOT allow {@code null} to be
* used as an element. The set is serializable.
*
* @param Generics Type.
* @return a new, empty thread-safe {@code Set}
*/
public static Set newConcurrentHashSet() {
return Collections.newSetFromMap(new ConcurrentHashMap());
}
/**
* Returns a capacity that is sufficient to keep the map from being resized
* as long as it grows no larger than expectedSize and the load factor
* is ≥ its default (0.75).
* The implementation of this method is adapted from Guava version 27.0-jre.
*/
private static int capacity(int expectedSize) {
if (expectedSize < 3) {
if (expectedSize < 0) {
throw new IllegalArgumentException(
"expectedSize cannot be negative but was: " + expectedSize);
}
return expectedSize + 1;
}
if (expectedSize < MAX_POWER_OF_TWO) {
// This is the calculation used in JDK8 to resize when a putAll
// happens; it seems to be the most conservative calculation we
// can make. 0.75 is the default load factor.
return (int) ((float) expectedSize / 0.75F + 1.0F);
}
return Integer.MAX_VALUE; // any large value
}
}