单链表的操作是面试中经常会遇到的问题,今天总结一下反转的几种方案:
1 ,两两对换
2, 放入数组,倒置数组
3, 递归实现
代码如下:
#include<stdio.h>
#include<malloc.h>
typedef struct Node
{
int data;
struct Node *pnext;
} Node,*pnode;
pnode CreateNode()
{
pnode phead=(pnode)malloc(sizeof(Node));
if(phead==NULL)
{
printf("fail to allocate memory");
return -1;
}
phead->pnext=NULL;
int n;
pnode ph=phead;
for(int i=0; i<5; i++)
{
pnode p=(pnode)malloc(sizeof(Node));
if(p==NULL)
{
printf("fail to allocate memory");
return -1;
}
p->data=(i+2)*19;
phead->pnext=p;
p->pnext=NULL;
phead=phead->pnext;
}
return ph;
}
int list(pnode head)
{
int count=0;
printf("遍历结果:\n");
while(head->pnext!=NULL)
{
printf("%d\t",head->pnext->data);
head=head->pnext;
count++;
}
printf("链表长度为:%d\n",count);
return count;
}
pnode reverse2(pnode head)//两两节点之间不断交换
{
if(head == NULL || head->next == NULL)
return head;
pnode pre = NULL;
pnode next = NULL;
while(head != NULL){
next = head->next;
head->next = pre;
pre = head;
head = next;
}
return pre;
}
void reverse1(pnode head,int count)//把链表的节点值放在数组中,倒置数组
{
int a[5]= {0};
for(int i=0; i<count,head->pnext!=NULL; i++)
{
a[i]=head->pnext->data;
head=head->pnext;
}
for(int j=0,i=count-1; j<count; j++,i--)
printf("%d\t",a[i]);
}
pnode reverse3(pnode pre,pnode cur,pnode t)//递归实现链表倒置
{
cur -> pnext = pre;
if(t == NULL)
return cur; //返回无头节点的指针,遍历的时候注意
reverse3(cur,t,t->pnext);
}
pnode new_reverse3(pnode head){ //新的递归转置
if(head == NULL || head->next == NULL)
return head;
pnode new_node = new_reverse3(head->next);
head->next->next = head;
head->next = NULL;
return new_node; //返回新链表头指针
}
int main()
{
pnode p=CreateNode();
pnode p3=CreateNode();
int n=list(p);
printf("1反转之后:\n");
reverse1(p,n);
printf("\n");
printf("2反转之后:\n");
pnode p1=reverse2(p);
list(p1);
p3 -> pnext = reverse3(NULL,p3 -> pnext,p3->pnext->pnext);
printf("3反转之后:\n");
list(p3);
free(p);
free(p1);
free(p3);
return 0;
}
毫无疑问,递归是解决的最简单方法,四行就能解决倒置问题。
思路参考:http://blog.csdn.net/feliciafay/article/details/6841115
这里注意: head ->next = pre; 以及 pre = head->next,前者把head->next 指向 pre,而后者是把head->next指向的节点赋值给pre。如果原来head->next 指向 pnext节点,前者则是head重新指向pre,与pnext节点断开,后者把pnext值赋值给pre,head与pnext并没有断开。