堆排序函数说明需要

时间:2010-03-22 20:56:46

标签: sorting heap

有人可以清楚地解释一下堆排序的这些函数是如何工作的吗?

void heapSort(int numbers[], int array_size)
{
  int i, temp;

  for (i = (array_size / 2)-1; i >= 0; i--)
    siftDown(numbers, i, array_size);

  for (i = array_size-1; i >= 1; i--)
  {
    temp = numbers[0];
    numbers[0] = numbers[i];
    numbers[i] = temp;
    siftDown(numbers, 0, i-1);
  }
}


void siftDown(int numbers[], int root, int bottom)
{
  int done, maxChild, temp;

  done = 0;
  while ((root*2 <= bottom) && (!done))
  {
    if (root*2 == bottom)
      maxChild = root * 2;
    else if (numbers[root * 2] > numbers[root * 2 + 1])
      maxChild = root * 2;
    else
      maxChild = root * 2 + 1;

    if (numbers[root] < numbers[maxChild])
    {
      temp = numbers[root];
      numbers[root] = numbers[maxChild];
      numbers[maxChild] = temp;
      root = maxChild;
    }
    else
      done = 1;
  }
}

2 个答案:

答案 0 :(得分:1)

This page对堆排序图有充分的解释。它有助于将其视为锦标赛:首先,您插入所有玩家,以便顶级玩家成为赢家。然后你提取胜利者,宣传失败者作为新的赢家,并进行调整,以便你再次获得一个正确的锦标赛,新的获胜者是其余球员中最好的。

然后你迭代。

答案 1 :(得分:0)

public class heapsort
{


public static void buildheap(int[] a, int nextLimit)
{
    //  for parent 3 child is 3*2+1=7 and 3*2+2=8 hence parent if odd n+1/2-1 i.e (7+1)/2-1=3 for odd n/2-1 8/2-1=3
    int child = nextLimit % 2 == 1 ? nextLimit + 1 : nextLimit;
    for (int parent = child / 2 - 1; parent >= 0; parent--) {
        heapfy(a, parent, nextLimit);
    }


}

public static void heapfy(int[] a, int parentIndex, int limit)
{
    int maxChildIndex;
    //if parent have only one child  (java array index start from 0 hence left one  2i+1 and right one 2i+2)
    if ((2 * parentIndex + 2) > limit) {
        maxChildIndex = 2 * parentIndex + 1;
    } else {
        //find max value index from two child
        maxChildIndex = a[2 * parentIndex + 1] > a[2 * parentIndex + 2] ? 2 * parentIndex + 1 : 2 * parentIndex + 2;
    }


    //swap if parent less than max child   bring max value to parent
    if (a[maxChildIndex] > a[parentIndex]) {

        int maxValue = a[maxChildIndex];
        a[maxChildIndex] = a[parentIndex];
        a[parentIndex] = maxValue;

    }


}


public static void main(String[] args)
{
    int[] a = {2, 5, 4, 6, 77, 3, 1, 8};
    for (int nextArrayLength = a.length - 1; nextArrayLength >= 0; nextArrayLength--) {
        buildheap(a, nextArrayLength);

        //push to first to last
        int highest = a[0];
        a[0] = a[nextArrayLength];
        a[nextArrayLength] = highest;
    }


    for (int i = 0; i < a.length; i++) {


        System.out.println(a[i]);
    }


}

}