10. Heaps

In this chapter, we discuss two implementations of the extremely useful priority $\mathtt{Queue}$ data structure. Both of these structures are a special kind of binary tree called a heap, which means ``a disorganized pile.'' This is in contrast to binary search trees that can thought of as a highly organized pile.

The first heap implementation uses an array to simulate a complete binary tree. It is very fast and is the basis of one of the fastest known sorting algorithms, namely heapsort (see Section 11.1.3). The second implementation is based on more flexible binary trees. It supports a $\mathtt{meld(h)}$ operation that allows the priority queue to absorb the elements of a second priority queue $\mathtt{h}$.