I am porting a C++ library to Java and I need a heap data structure. Is there a standard implementation or will I need to do it myself?
For Java 8, updating on an existing answer:
You can use Java Priority Queue as a Heap.
Min Heap: --> to keep the min element always on top, so you can access it in O(1).
PriorityQueue<Integer> minHeap = new PriorityQueue<Integer>();
Max Heap: --> to keep the max element always on top, the same order as above.
PriorityQueue<Integer> maxHeap = new PriorityQueue<>(Comparator.reverseOrder());
Which is the same as (Integer o1, Integer o2) -> Integer.compare(o2, o1)
or - Integer.compare(o1, o2)
as suggested from other answers.
And you can use:
add
--> to add element to the queue. O(log n)
remove
--> to get and remove the min/max. O(log n)
peek
--> to get, but not remove the min/max. O(1)
Min heap:
PriorityQueue<Integer> minHeap = new PriorityQueue<Integer>();
Max heap:
PriorityQueue<Integer> maxHeap = new PriorityQueue<Integer>(new Comparator<Integer>() {
@Override
public int compare(Integer o1, Integer o2) {
return - Integer.compare(o1, o2);
}
});
- Integer.compare(o1, o2);
shall be same as Integer.compare(o2, o1);
–
Clutter In Java PriorityQueue can be used as a Heap.
Min Heap
PriorityQueue<Integer> minHeap = new PriorityQueue<>();
Max Heap
PriorityQueue<Integer> maxHeap = new PriorityQueue<>(Comparator.reverseOrder());
PriorityQueue uses a heap. Based on the oracle documentation at https://docs.oracle.com/javase/8/docs/api/java/util/PriorityQueue.html it seems likely that it is an implementation of a binary heap. I don't think there is an official implementation of a fibonacci or pairing heap, though I'd love to see either one of the two available.
No as such there isn't but you can use Priority Queue as a Heap. Its officially told by Oracle to use Priority Queue as a Heap you can also refer to this link for further clarification.
PriorityQueue<Integer> MinHeap = new PriorityQueue<>();
PriorityQueue<Integer> MaxHeap = new PriorityQueue<>(Comparator.reverseOrder());
From Java docs PriorityQueue
which is available since 1.5 is the class to use.
This code for Min Heap
creates a PriorityQueue with the default initial capacity (11) that orders its elements according to their natural ordering in which the min is at the top.
//MIN HEAP
PriorityQueue<Integer> minHeap = new PriorityQueue<>();
//equivalent to
PriorityQueue<Integer> minHeap = new PriorityQueue<>(11);
If you want to implement a special ordering you need to override the comparator with this constructor
PriorityQueue(int initialCapacity, Comparator<? super E> comparator);
Since 1.8 we also have this version
PriorityQueue(Comparator<? super E> comparator);
which helps you create the Max Heap
in more elegant ways such as
//MAX HEAP
PriorityQueue<Integer> maxHeap =
new PriorityQueue<>((n1,n2) -> Integer.compare(n2,n1));
//equivalent to
PriorityQueue<Integer> maxHeap = new PriorityQueue<>(Comparator.reverseOrder());
For a special case check this example that shows the natural ordering for a custom object, in a scenario where we order customers based on their distance to a fictional restaurant
import java.util.List;
import java.util.PriorityQueue;
public class DeliveryHandler {
private static final Address restaurant = new Address(5.0, 5.0);
private static class Address implements Comparable<Address> {
public double x, y;
public Address(double x, double y) {
this.x = x;
this.y = y;
}
public double distanceToShop() {
return Math.pow(restaurant.x - x, 2) + Math.pow(restaurant.y - y, 2);
}
@Override
public int compareTo(Address other) {
return Double.compare(this.distanceToShop(), other.distanceToShop());
}
@Override
public String toString() {
return "Address {x=%s, y=%s}".formatted(x, y);
}
}
public static void main(String[] args) {
List<Address> customers = List.of(
new Address(13, 14),
new Address(3, 1),
new Address(9, 20),
new Address(12, 4),
new Address(4, 4));
PriorityQueue<Address> queueServingClosest = new PriorityQueue<>();
queueServingClosest.addAll(customers);
while (!queueServingClosest.isEmpty()) {
System.out.println(queueServingClosest.remove());
}
/* Prints
Address {x=4.0, y=4.0}
Address {x=3.0, y=1.0}
Address {x=12.0, y=4.0}
Address {x=13.0, y=14.0}
Address {x=9.0, y=20.0}
*/
PriorityQueue<Address> queueServingFurthest = new PriorityQueue<>(
(a1, a2) -> Double.compare(a2.distanceToShop(), a1.distanceToShop())
);
queueServingFurthest.addAll(customers);
while (!queueServingFurthest.isEmpty()) {
System.out.println(queueServingFurthest.remove());
}
/* Prints
Address {x=9.0, y=20.0}
Address {x=13.0, y=14.0}
Address {x=12.0, y=4.0}
Address {x=3.0, y=1.0}
Address {x=4.0, y=4.0}
*/
}
}
Refs
1- https://docs.oracle.com/javase/7/docs/api/java/util/PriorityQueue.html
2- https://docs.oracle.com/en/java/javase/11/docs/api/java.base/java/util/PriorityQueue.html
You can also consider TreeSet, which guarantees log(n) time for basic operations (add, remove, contains).
© 2022 - 2024 — McMap. All rights reserved.
PriorityQueue
, guava provides aMinMaxPriorityQueue
– Alexandrina