How to iterate over a PriorityQueue?

Question

for (Event e : pq)

doesn't iterate in the priority order.

while(!pq.isEmpty()){
  Event e = pq.poll();
}

This works but empties the queue.

Answer 1

You can make a copy of the queue and poll in a loop, in this example pq is the original priority queue:

PriorityQueue<Your_class> pqCopy = new PriorityQueue<Your_class>(pq);
while(!pqCopy.isEmpty()){
    Your_Class obj = pqCopy.poll();
    // obj is the next ordered item in the queue
    .....
}

Answer 2

Previous posters said everything but noone gave full working example (other than copying pq), so here it is:

Event[] events = pq.toArray(new Event[pq.size()]);
Arrays.sort(events, pq.comparator());
for (Event e : events) {
    System.out.println(e);
}

Answer 3

The peek() method does not remove anything from the queue, but because of this, it will continually get the top value until it IS empty. I'm guessing you checked if it was empty after your while loop, which would give you this conclusion.

The only way to do this is to sort it yourself. You can get the original comparator for it like this:

Event[] events = Arrays.sort(pq.toArray(), pq.comparator());
for (Event e : events) {
    // do stuff
}

Answer 4

for (Event event: pq.toArray(new Event[pq.size()])) {
    event.toString();
}

Answer 5

You can't traverse a Priority Queue in that order because of the underlying implementation (I think it's min-heap in Java).

It's not a sorted array, so that you can just go from one element to the one with the lesser priority.

Peeking (read the top element heap in the heap) is constant time O(1) because it looks at the smallest element.

To get the second next one you must dequeue the smallest top element, that's how it works.
Dequeing (re-heapify = O(log n) time) isn't just a matter of taking that element out, the underlying structure rearranges itself in order to bring the element with the least priority first.

Also, to go through the entire priority queue to read all items in the sorted order, it is an O(n log(n)) operation.
So you may as well just grab all the elements in the queue and sort them (also O(n log (n)) )and then you can go through them as you wish. The only disadvantage is that you're holding an extra-copy of the queue.

Nonetheless, if you need to traverse the data in this way a priority queue may not be the right data structure for your needs.

Answer 6

Recently, I had same problem. I wanted to use some particular object from the priority queue and then keep remaining elements preserved.

1) I created a newPriorityQueue. 2) Used Iterator to parse every element in oldQueue 3) used oldQueue.poll() method to retrieve the element 4) insert the element 3) to newPriorityQueue if not used.

 Queue<String> newQueue = new PriorityQueue<String>(); 
    // Assuming that oldQueue have some data in it.

    Iterator<String> itr = oldQueue.iterator();
    while(itr.hasNext()){
        String str = oldQueue.poll();
        // do some processing with str
        if(strNotUsed){
            newQueue.offer(str);
        }
    }

In the end, oldQueue will be empty. @ Others : - please suggest a better way if I can do the same thing. I can not use the iterator as it does not return elements in the correct order.