【数据结构】手撕双向链表

 特点：带头双向循环链表结构最复杂，一般用在单独存储数据。结构虽然结构复杂，但是使用代码实现以后会发现结构会带来多优势，实现反而简单了。

2. 链表的实现

2.1 初始化

LTNode* LTInit()
{LTNode* phead = CreateLTNode(-1);phead->next = phead;phead->prev = phead;return phead;
}

2.2 尾插

带哨兵位的链表尾插时不用判断是否有节点，两种情况的插入相同，而且还不用传递二级指针。

void LTPushBack(LTNode* phead, LTDateType x)
{assert(phead);LTNode* newnode = CreateLTNode(x);phead->prev->next = newnode;newnode->prev = phead->prev;newnode->next = phead;phead->prev = newnode;
}

2.3 尾删

在尾删时我们通过 assert(phead->next != phead);  判断链表是否有节点。同时这个代码就有普遍性，不用单独考虑剩一个节点的情况。

void LTPopBack(LTNode* phead)
{assert(phead);LTNode* tail = phead->prev;LTNode* tailprev = tail->prev;free(tail);phead->prev = tailprev;tailprev->next = phead;
}

2.4 头插

头删重要的是赋值的顺序，顺序错误会找不到后面的节点，导致内存泄漏。带哨兵位的链表不需要传递二级指针，因为改变的是结构体的变量。

void LTPushFront(LTNode* phead, LTDateType x)
{assert(phead);LTNode* newnode = CreateLTNode(x);newnode->next = phead->next;phead->next->prev = newnode;phead->next = newnode;newnode->prev = phead;
}

2.5 头删

我们可以多定义几个指针来保存后面节点的地址，这样就不会造成节点的丢失，不用考虑赋值的顺序，会更加方便。 

void LTPopFront(LTNode* phead)
{assert(phead);assert(phead->next != phead);LTNode* tail = phead->next;LTNode* next = tail->next;phead->next = next;next->prev = phead;free(tail);tail = NULL;
}

2.6 在pos位置之前插入

void LTInsert(LTNode* pos, LTDateType x)
{assert(pos);LTNode* posprev = pos->prev;LTNode* newnode = CreateLTNode(x);posprev->next = newnode;newnode->prev = posprev;newnode->next = pos;pos->prev = newnode;
}

2.7 删除pos位置

void LTErase(LTNode* pos)
{assert(pos);LTNode* posprev = pos->prev;LTNode* posnext = pos->next;posprev->next = posnext;posnext->prev = posprev;
}

3.双向链表完整源码

List.h

#include<stdio.h>
#include<stdlib.h>
#include<assert.h>typedef int LTDateType;typedef struct ListNode
{LTDateType val;struct ListNode* next;struct ListNode* prev;
}LTNode;LTNode* LTInit();void LTPrint(LTNode* phead);void LTPushBack(LTNode* phead, LTDateType x);void LTPushFront(LTNode* phead, LTDateType x);void LTPopBack(LTNode* phead);void LTPopFront(LTNode* phead);LTNode* LTFind(LTNode* phead, LTDateType x);void LTInsert(LTNode* pos, LTDateType x);void LTErase(LTNode* pos);void LTDestroy(LTNode* phead);

List.c

#include"List.h"LTNode* CreateLTNode(LTDateType x)
{LTNode* newnode = (LTNode*)malloc(sizeof(LTNode));if (newnode == NULL){perror("malloc fail");exit(-1);}newnode->val = x;newnode->next = NULL;newnode->prev = NULL;
}LTNode* LTInit()
{LTNode* phead = CreateLTNode(-1);phead->next = phead;phead->prev = phead;return phead;
}void LTPrint(LTNode* phead)
{assert(phead);LTNode* cur = phead->next;printf("<=>哨兵位<=>");while (cur != phead){printf("%d<=>", cur->val);cur = cur->next;}printf("\n");
}void LTPushBack(LTNode* phead, LTDateType x)
{assert(phead);LTNode* newnode = CreateLTNode(x);phead->prev->next = newnode;newnode->prev = phead->prev;newnode->next = phead;phead->prev = newnode;
}void LTPushFront(LTNode* phead, LTDateType x)
{assert(phead);LTNode* newnode = CreateLTNode(x);newnode->next = phead->next;phead->next->prev = newnode;phead->next = newnode;newnode->prev = phead;
}void LTPopBack(LTNode* phead)
{assert(phead);LTNode* tail = phead->prev;LTNode* tailprev = tail->prev;free(tail);phead->prev = tailprev;tailprev->next = phead;
}void LTPopFront(LTNode* phead)
{assert(phead);assert(phead->next != phead);LTNode* tail = phead->next;LTNode* next = tail->next;phead->next = next;next->prev = phead;free(tail);tail = NULL;
}LTNode* LTFind(LTNode* phead, LTDateType x)
{assert(phead);LTNode* cur = phead->next;while (cur != phead){if (cur->val == x){return cur;}cur = cur->next;}return NULL;
}void LTInsert(LTNode* pos, LTDateType x)
{assert(pos);LTNode* posprev = pos->prev;LTNode* newnode = CreateLTNode(x);posprev->next = newnode;newnode->prev = posprev;newnode->next = pos;pos->prev = newnode;
}void LTErase(LTNode* pos)
{assert(pos);LTNode* posprev = pos->prev;LTNode* posnext = pos->next;posprev->next = posnext;posnext->prev = posprev;
}void LTDestroy(LTNode* phead)
{assert(phead);LTNode* cur = phead->next;while (cur != phead){LTNode* next = cur->next;free(cur);cur = next;}free(phead);phead = NULL;
}

通过上面链表的实现，我们已经感受到了带头双向循环链表的方便和简单，它不需要去考虑链表是否有元素，还可以找到前一个元素，在我们使用中提供很大的便利。

本次的内容到这里就结束啦。希望大家阅读完可以有所收获，同时也感谢各位读者三连支持。文章有问题可以在评论区留言，博主一定认真认真修改，以后写出更好的文章。