对希尔排序的一点理解

void  Shell_Sort_Unit(Sqlit  &L,int  d)//d为比较增量
{
    int  j,k;
    for(j=d+1;j<=L.length;j++)//length不包括哨兵
    {
         L.R[0]=L.R[j];
         for(k=j-d;k>0&&L.R[0].key<L.R[k].key;k=k-d)
         {
             L.R[k+d]=L.R[k];
         }
         L.R[k+d]=L.R[0];
    }

}


完整代码：

#include "stdafx.h"
#include "malloc.h"
#include  "stdlib.h"

#define  OK  1
#define  ERROR  -1
#define  OVERFLOW  -2
#define  MAX_SIZE   100
#define  TRUE  1
#define  FLASE  0
typedef   int   KeyType;

typedef  struct  RecType
{
    KeyType  key;/*关键字*/
}RecType;
typedef  struct  Sqlist
{
    RecType R[MAX_SIZE];
    int  length;
}Sqlist;

void  Create(Sqlist  &L)
{
    int  d = 0;
    L.length = 10;
    L.R[0].key= 0;
    for (int i = 1; i <= 10; i++)
    {
        printf("请输入第%d个数据", i);
        scanf_s("%d", &d);
        L.R[i].key = d;
    }
}

void  Travel(Sqlist  L)
{
    for (int i = 1; i <= L.length; i++)
    {
        printf("节点%d的主键:%d\r\n", i, L.R[i].key);

    }
}


/*希尔排序*/
void Shell_Sort_Unit(Sqlist  &L, int  d)
{
    int c, j;
    for (j = d + 1; j < L.length; j++)
    {
        L.R[0] = L.R[j];
        for (c = j - d; c>0 && L.R[c].key > L.R[0].key; c = c - d)
        {
            L.R[c + d] = L.R[c];
        }
        L.R[c + d] = L.R[0];
    }
}

void  Shell_Sort(Sqlist  &L, int  dk[], int t)
{
    for (int i = 0; i < t; i++)
    {
        Shell_Sort_Unit(L, dk[i]);
    }
}


int  main()
{
    Sqlist  L;
    Create(L);
    Travel(L);
    int dk[3];
    dk[0] = 5;
    dk[1] = 3;
    dk[2] = 1;
    Shell_Sort(L, dk, 3);
    Travel(L);
}