PriorityQueueMonoid

[Koert Kuipers]

In PriorityQueueMonoid there is a build method that takes an iterable:
def build(items: Iterable[K]): PriorityQueue[K]

Inside this method it creates the like this:
val q = new PriorityQueue(items.size max MINQUEUESIZE, ord.reverse);

Why base the size of the queue on the size of items if the queue can only grow to size max (or technically size max + 1 since it can first add one item before limiting it's size to max)?
If i have a very large iterable this leads to allocation a lot of memory for the queue which doesn't seem necessary to me.

Also if the queue was allocated like this
val q = new PriorityQueue(max + 1, ord.reverse);
then we could take an iterator instead of an iterable...