合并两个LinkedList而不创建新的LinkedList

Merge two LinkedList without creating a new LinkedList

本文关键字:LinkedList 创建 两个 合并      更新时间:2023-10-16

我正在尝试获取可执行文件中的两个链表,并将它们在交替位置合并。例如ListOne 1,2,3和ListTwo 4,5新的ListOne应该是1,4,2,5,3。

LinkedList.h文件:

class LinkedList
{
private:
struct ListNode
{
    string firstName;
    string lastName;
    long int phoneNumber;
    struct ListNode *next;
};
ListNode *head;
public:
LinkedList()
{
    head = nullptr;
}
~LinkedList();
void appendNode(string f, string l, long int p);
void displayList();
};

LinkedList.cpp文件:

LinkedList::~LinkedList()
{
cout << "LinkList destructor" << endl;
}
void LinkedList::appendNode(string f, string l, long int p)
{
    ListNode *newNode;
    ListNode *nodePtr;
    newNode = new ListNode;
    newNode -> firstName = f;
    newNode -> lastName = l;
    newNode -> phoneNumber = p;
    newNode -> next = nullptr;
    if (!head)
        head = newNode;
    else
    {
        nodePtr = head;
        while (nodePtr -> next)
            //while nodePtr is pointing to another node
            nodePtr = nodePtr -> next;
            //move to that node
        nodePtr -> next = newNode;
        //inset the newNode at the end of the linked list
    }
 }
 void LinkedList::displayList()
{
    ListNode *nodePtr;
    nodePtr = head;
    while(nodePtr)
    //while nodePtr is true, meaning there is a node in the list
    {
        cout << nodePtr -> firstName << endl;
        cout << nodePtr -> lastName << endl;
        cout << nodePtr -> phoneNumber << endl;
        nodePtr = nodePtr -> next;
     }
}

可执行文件:

LinkedList ListOne;
LinkedList ListTwo;
ListOne.appendNode("Cate", "Beckem", 7704563454);
ListOne.appendNode("Cabe","Tomas", 7703451523);
ListTwo.appendNode("Mary", "Smith", 4043456543);
ListTwo.appendNode("Mark", "Carter", 4045433454);

我的程序运行得很好,包括displayList函数。我只是很困惑如何去做一个合并函数。

制作合并函数并不困难。您可以使用一个新的头来记录新列表,然后遍历这两个列表,并依次将节点从这两个名单移动到新名单,如下所示。
 LinkedList LinkedList::merge(LinkedList b)
{
    // if one list is null, return the other
    if (this->head == nullptr)
        return b;
    if (b.head == nullptr)
        return *this;
    LinkedList newlist;
    ListNode *ap = this->head, *bp = b.head, *p = nullptr;
    // if two pointer is all not null, move node from these to new list in turn
    while (ap != nullptr && bp != nullptr)
    {
        if (newlist.head == nullptr)
        {
            p = newlist.head = ap;
            ap = ap->next;
            p = p->next = bp;
            bp = bp->next;
        }
        else
        {
            p = p->next = ap;
            ap = ap->next;
            p = p->next = bp;
            bp = bp->next;
        }
    }
    // maybe one list is longer, and there is some node left.
    if (ap != nullptr)
        p->next = ap;
    if (bp != nullptr)
        p->next = bp;
    //clear original list
    this->head = b.head = nullptr;
    //if you want to merge list b to the caller list, you can change to
    //this->head = newlist->head and beginning part also need some modification.
    return newlist;
}

也许您不想更改原始列表,那么您可以复制值并为新列表创建新节点。

合并是将已从源列表复制或将从源列表被盗的节点插入到目标列表的特定位置。

我假设您希望在该操作中将源列表视为不可变的,因此源节点将被复制。

对复制和插入操作都有用的是迭代器,它指向一个节点,在++之后,op指向列表中或列表末尾后面的下一个节点(相当于nullptr):

我在单个源文件中编写这样的代码段;将实现移动到.cpp文件或内联它们

//
#include <type_traits>
using std::conditional;
#include <stdexcept>
using std::runtime_error;
class LinkedList
{

private:
    // [...]
    template <bool const_tag>
    struct NodeIterator {
        using element_type = typename conditional<
            const_tag,
            const ListNode,
            ListNode
        >::type;
        using pointer_type = element_type*;
        using reference_type = element_type&;
        static const NodeIterator end;
        NodeIterator(pointer_type p_in = nullptr) : p(p_in) {}
        NodeIterator& operator++ () {
            if (nullptr == p)
                throw runtime_error("Attempt to dereference nullptr");
            this->p = p->next;
            return *this;
        }
        bool operator== (const NodeIterator& rhs) const {
            return this->p != rhs.p;
        }
        bool operator!= (const NodeIterator& rhs) const {
            return !(*this == rhs);
        }
        pointer_type operator->() const {
            return p;
        }
        reference_type operator*() const {
            if (nullptr == p)
                throw runtime_error("Attempt to dereference nullptr");
            return *p;
        }
    private:
        pointer_type p;    
    }; // template <bool const_tag> struct NodeIterator
    static constexpr bool mutable_tag = false; 
    static constexpr bool const_tag = true; 
    using iterator_type = NodeIterator<mutable_tag>;
    using const_iterator_type = NodeIterator<const_tag>;

public:     
    LinkedList() :  head(nullptr) {}
    iterator_type begin() const { return iterator_type(head); }
    const iterator_type& end() const { return iterator_type::end; }
    const_iterator_type cbegin() const { return const_iterator_type(head); }
    const const_iterator_type& cend() const { return const_iterator_type::end; }
    // [...]    

}; // class LinkedList

// [...]    
template<bool is_const>
const LinkedList::NodeIterator<is_const>
LinkedList::NodeIterator<is_const>::end(nullptr);
// [...]

现在,在合并函数代码中,您将迭代器定位在相应列表的头部

auto it_one = listOne.begin();
auto it_two = listTwo.cbegin();

递增it_one直到它指向要插入副本的目标列表的元素,递增it_two直到它指向将要复制的源列表的节点,然后

ListNode& one = *it_one;
ListNode* ptwo = new ListNode(*it_two); // copies
ptwo->next = one.next;
one.next = ptwo;

本质上就是这样。

因此,这里真正的问题不是技术上如何合并,而是操作的复杂性。

只要这两个列表都按照要创建的结果列表的预期顺序进行排序,就可以重用迭代器,从而简单地遍历这两个清单,它就是O(N)N = max(size(listOne), size(listTwo))

如果你必须对它们进行初步排序以保持合并本身的便宜,那么这种排序会导致O(N log N)的复杂性。


顺便说一句,存储迭代器也简化了其他操作的实现;例如,显示它们只是

for (const auto& node : ListOne) {
    std::cout << node.firstName << std::endl;
    std::cout << node.lastName << std::endl;
    std::cout << node.phoneNumber << std::endl;
}