Java中出队的优先级队列问题

Question

Hi so im working on a project right now and my dequeueing of minimum values first seems to have an issue. it appears to enqueue fine. punch in 20, 15, 11, 29 it correctly corresponds with 11,20,15,29. These numbers correspond with deadlines to be reached. now, when I want to dequeue the top value, it'll grab the 11 do the operations on it that are needed discard the task with deadline 11. then when I grab the next top value it returns 20 instead of 15. I cant find out why? Any help would be greatly appreciated!

import java.util.Comparator;


public class treeCompare implements Comparator<Task> {

    public int compare(Task t1, Task t2) {

        int dead1 = t1.getDeadline();
        int dead2 = t2.getDeadline();

        return dead1 - dead2;

    }


}

    import java.util.ArrayList;
import java.util.Comparator;


public class PriorityQueue<E> implements QueueADT<E> {

    private ArrayList<E> items;
    private int max;
    private Comparator<E> comp;

    public PriorityQueue(Comparator<E> comp, int maxCapacity){
        this.comp = comp;
        this.max = maxCapacity;
        items = new ArrayList<E>();
    }



    public boolean isFull() {
        if(size() == max) return true;
        else return false;
    }


     private void siftUp() {
            int k = items.size() - 1;
            while (k > 0) {
                int p = (k-1)/2;
                E item = items.get(k);
                E parent = items.get(p);

                     if (comp.compare(items.get(k), items.get(p)) < 0) {
                        // swap
                        items.set(k, parent);
                        items.set(p, item);

                        // move up one level
                        k = p;

                } else {
                    break;
                }
            }
        }

        public void enqueue(E item)throws FullQueueException {
            if(isFull()) throw new FullQueueException();
            items.add(item);
            siftUp();
        }


    /**
     * Returns the front item of the queue without removing it.
     * @return the front item of the queue
     * @throws EmptyQueueException
     */
    public E peek() throws EmptyQueueException {
        if(isEmpty()) throw new EmptyQueueException();
        else{
            E item = items.get(0);
            return item;
        }
    }

    private void siftDown() {
        int k = 0;
        int l = 2*k+1;
        while (l < items.size()) {
            int max=l, r=l+1;
            if (r < items.size()) { // there is a right child
                if (comp.compare(items.get(r), items.get(l)) > 0) {
                    max++;
                }
            }
            if (comp.compare(items.get(k), items.get(max)) > 0) {
                    // switch
                    E temp = items.get(k);
                    items.set(k, items.get(max));
                    items.set(max, temp);
                    k = max;
                    l = 2*k+1;
            } else {
                break;
            }
        }
    }

    public E dequeue() throws EmptyQueueException {
        if (items.size() == 0) {
            throw new EmptyQueueException();
        }
        if (items.size() == 1) {
            return items.remove(0);
        }
        E hold = items.get(0);
        items.set(0, items.remove(items.size()-1));
        siftDown();
        return hold;
    }

    public int size() {
        return items.size();
    }

    public boolean isEmpty() {
        return items.isEmpty();

    }



   public String toString() {
       return items.toString();
   }


    public int capacity() {
        return max;
    }

}

Answer 1

The comparision comp.compare(items.get(r), items.get(l)) > 0 in shiftDown() seems to be inverted.

Inverting it (and renaming variable max for clarity) the algorithm seems to work (as far as I've tested).

    private void siftDown() {
        int k = 0;
        int l = 2*k+1;
        while (l < items.size()) {
            int min=l, r=l+1;
            if (r < items.size()) { // there is a right child
                if (comp.compare(items.get(r), items.get(l)) < 0) {
                    min = r;
                }
            }
            if (comp.compare(items.get(k), items.get(min)) > 0) {
                    // switch
                    E temp = items.get(k);
                    items.set(k, items.get(min));
                    items.set(min, temp);
                    k = min;
                    l = 2*k+1;
            } else {
                break;
            }
        }
    }

If I insert [20, 15, 11, 1, 29, 10, 7, 28, 3, 54, 40, 34, 58] I retrieve [1, 3, 7, 10, 11, 15, 20, 28, 29, 34, 40, 54, 58] which looks nice.
Hope it helps!