服务器之家

服务器之家 > 正文

C++实现LeetCode(148.链表排序)

时间:2021-12-07 14:48     来源/作者:Grandyang

[LeetCode] 148. Sort List 链表排序

Sort a linked list in O(n log n) time using constant space complexity.

Example 1:

Input: 4->2->1->3
Output: 1->2->3->4

Example 2:

Input: -1->5->3->4->0
Output: -1->0->3->4->5

常见排序方法有很多,插入排序,选择排序,堆排序,快速排序,冒泡排序,归并排序,桶排序等等。。它们的时间复杂度不尽相同,而这里题目限定了时间必须为O(nlgn),符合要求只有快速排序,归并排序,堆排序,而根据单链表的特点,最适于用归并排序。为啥呢?这是由于链表自身的特点决定的,由于不能通过坐标来直接访问元素,所以快排什么的可能不太容易实现(但是被评论区的大神们打脸,还是可以实现的),堆排序的话,如果让新建结点的话,还是可以考虑的,若只能交换结点,最好还是不要用。而归并排序(又称混合排序)因其可以利用递归来交换数字,天然适合链表这种结构。归并排序的核心是一个 merge() 函数,其主要是合并两个有序链表,这个在 LeetCode 中也有单独的题目 Merge Two Sorted Lists。由于两个链表是要有序的才能比较容易 merge,那么对于一个无序的链表,如何才能拆分成有序的两个链表呢?我们从简单来想,什么时候两个链表一定都是有序的?就是当两个链表各只有一个结点的时候,一定是有序的。而归并排序的核心其实是分治法 Divide and Conquer,就是将链表从中间断开,分成两部分,左右两边再分别调用排序的递归函数 sortList(),得到各自有序的链表后,再进行 merge(),这样整体就是有序的了。因为子链表的递归函数中还是会再次拆成两半,当拆到链表只有一个结点时,无法继续拆分了,而这正好满足了前面所说的“一个结点的时候一定是有序的”,这样就可以进行 merge 了。然后再回溯回去,每次得到的都是有序的链表,然后进行 merge,直到还原整个长度。这里将链表从中间断开的方法,采用的就是快慢指针,大家可能对快慢指针找链表中的环比较熟悉,其实找链表中的中点同样好使,因为快指针每次走两步,慢指针每次走一步,当快指针到达链表末尾时,慢指针正好走到中间位置,参见代码如下:

C++ 解法一:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
class Solution {
public:
    ListNode* sortList(ListNode* head) {
        if (!head || !head->next) return head;
        ListNode *slow = head, *fast = head, *pre = head;
        while (fast && fast->next) {
            pre = slow;
            slow = slow->next;
            fast = fast->next->next;
        }
        pre->next = NULL;
        return merge(sortList(head), sortList(slow));
    }
    ListNode* merge(ListNode* l1, ListNode* l2) {
        ListNode *dummy = new ListNode(-1);
        ListNode *cur = dummy;
        while (l1 && l2) {
            if (l1->val < l2->val) {
                cur->next = l1;
                l1 = l1->next;
            } else {
                cur->next = l2;
                l2 = l2->next;
            }
            cur = cur->next;
        }
        if (l1) cur->next = l1;
        if (l2) cur->next = l2;
        return dummy->next;
    }
};

Java 解法一:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
public class Solution {
    public ListNode sortList(ListNode head) {
        if (head == null || head.next == null) return head;
        ListNode slow = head, fast = head, pre = head;
        while (fast != null && fast.next != null) {
            pre = slow;
            slow = slow.next;
            fast = fast.next.next;
        }
        pre.next = null;
        return merge(sortList(head), sortList(slow));
    }
    public ListNode merge(ListNode l1, ListNode l2) {
        ListNode dummy = new ListNode(-1);
        ListNode cur = dummy;
        while (l1 != null && l2 != null) {
            if (l1.val < l2.val) {
                cur.next = l1;
                l1 = l1.next;
            } else {
                cur.next = l2;
                l2 = l2.next;
            }
            cur = cur.next;
        }
        if (l1 != null) cur.next = l1;
        if (l2 != null) cur.next = l2;
        return dummy.next;
    }
}

下面这种方法也是归并排序,而且在merge函数中也使用了递归,这样使代码更加简洁啦~

C++ 解法二:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
class Solution {
public:
    ListNode* sortList(ListNode* head) {
        if (!head || !head->next) return head;
        ListNode *slow = head, *fast = head, *pre = head;
        while (fast && fast->next) {
            pre = slow;
            slow = slow->next;
            fast = fast->next->next;
        }
        pre->next = NULL;
        return merge(sortList(head), sortList(slow));
    }
    ListNode* merge(ListNode* l1, ListNode* l2) {
        if (!l1) return l2;
        if (!l2) return l1;
        if (l1->val < l2->val) {
            l1->next = merge(l1->next, l2);
            return l1;
        } else {
            l2->next = merge(l1, l2->next);
            return l2;
        }
    }
};

Java 解法二:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
public class Solution {
    public ListNode sortList(ListNode head) {
        if (head == null || head.next == null) return head;
        ListNode slow = head, fast = head, pre = head;
        while (fast != null && fast.next != null) {
            pre = slow;
            slow = slow.next;
            fast = fast.next.next;
        }
        pre.next = null;
        return merge(sortList(head), sortList(slow));
    }
    public ListNode merge(ListNode l1, ListNode l2) {
        if (l1 == null) return l2;
        if (l2 == null) return l1;
        if (l1.val < l2.val) {
            l1.next = merge(l1.next, l2);
            return l1;
        } else {
            l2.next = merge(l1, l2.next);
            return l2;
        }
    }
}

Github 同步地址:

https://github.com/grandyang/leetcode/issues/148

类似题目:

Merge Two Sorted Lists

Sort Colors

Insertion Sort List

参考资料:

https://leetcode.com/problems/sort-list/description/

https://leetcode.com/problems/sort-list/discuss/46857/clean-and-short-merge-sort-solution-in-c

https://leetcode.com/problems/sort-list/discuss/46937/56ms-c-solutions-using-quicksort-with-explanations

https://leetcode.com/problems/sort-list/discuss/46772/i-have-a-pretty-good-mergesort-method-can-anyone-speed-up-the-run-time-or-reduce-the-memory-usage

到此这篇关于C++实现LeetCode(148.链表排序)的文章就介绍到这了,更多相关C++实现链表排序内容请搜索服务器之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持服务器之家!

原文链接:https://www.cnblogs.com/grandyang/p/4249905.html

相关文章

热门资讯

yue是什么意思 网络流行语yue了是什么梗
yue是什么意思 网络流行语yue了是什么梗 2020-10-11
背刺什么意思 网络词语背刺是什么梗
背刺什么意思 网络词语背刺是什么梗 2020-05-22
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全 2019-12-26
2021年耽改剧名单 2021要播出的59部耽改剧列表
2021年耽改剧名单 2021要播出的59部耽改剧列表 2021-03-05
苹果12mini价格表官网报价 iPhone12mini全版本价格汇总
苹果12mini价格表官网报价 iPhone12mini全版本价格汇总 2020-11-13
返回顶部