[WIP] LeetCode 23. Merge k Sorted Lists

题目：

Merge k sorted linked lists and return it as one sorted list. Analyze and describe its complexity.

Example:

Input:
[
  1->4->5,
  1->3->4,
  2->6
]
Output: 1->1->2->3->4->4->5->6

---

设置k个pointer分别指向这k个list的head，找到几个head里的最小值插入新的list。

解法一：brute force

2024.7.10:

somehow也还是借鉴了以前写的才写出来但是大部分写的七七八八了？主要的confusing point是在内层for loop遍历每个list的时候刚开始没有用index而是直接用了一个listNode来记录最小的node，最后移动next的时候可能就出问题了。嗯……搞这些reference的时候还是尽量别搞这么复杂，还是用最基础最简单的最不容易出错。

/**
 * Definition for singly-linked list.
 * public class ListNode {
 *     int val;
 *     ListNode next;
 *     ListNode() {}
 *     ListNode(int val) { this.val = val; }
 *     ListNode(int val, ListNode next) { this.val = val; this.next = next; }
 * }
 */
class Solution {
    public ListNode mergeKLists(ListNode[] lists) {
        ListNode dummy = new ListNode();
        ListNode curr = dummy;
        while (true) {
            int smallestVal = Integer.MAX_VALUE;
            int smallestIndex = -1;
            for (int i = 0; i < lists.length; i++) {
                if (lists[i] == null) {
                    continue;
                }
                if (lists[i].val < smallestVal) {
                    smallestVal = lists[i].val;
                    smallestIndex = i;
                }
            }
            if (smallestVal != Integer.MAX_VALUE) {
                curr.next = lists[smallestIndex];
                lists[smallestIndex] = lists[smallestIndex].next;
                curr = curr.next;
            } else {
                break;
            }
        }
        return dummy.next;
    }
}

2022.9.16 Java:

debug了老半天，主要犯了以下几个错：

1. 又双叒忘记移动curr指针

2. min和minIndex放在while loop里，而且应该设为max integer和-1而不能设为第一个

3. 差点忘了lists[i] != null的条件，且最后移动lists[minIndex]是必须的，而不是跟判断next == null互斥的else

4. 再次遗漏了[[]]的edge case

/**
 * Definition for singly-linked list.
 * public class ListNode {
 *     int val;
 *     ListNode next;
 *     ListNode() {}
 *     ListNode(int val) { this.val = val; }
 *     ListNode(int val, ListNode next) { this.val = val; this.next = next; }
 * }
 */
class Solution {
    public ListNode mergeKLists(ListNode[] lists) {
        ListNode dummy = new ListNode();
        ListNode curr = dummy;
        int size = lists.length;
        
        int endCount = 0;
        while (endCount != size) {
            int min = Integer.MAX_VALUE;
            int minIndex = -1;
            for (int i = 0; i < size; i++) {
                if (lists[i] != null && lists[i].val < min) {
                    min = lists[i].val;
                    minIndex = i;
                }
            }
            // deal with case: [[]]
            if (minIndex == -1) {
                break;
            }
            curr.next = lists[minIndex];
            curr = curr.next;
            if (lists[minIndex].next == null) {
                endCount++;
            }
            lists[minIndex] = lists[minIndex].next;
        }
        return dummy.next;
        
    }
}

---

解法二：priority queue

采用prioirty_queue进行优化，本质和上面的一样，只是找最小从之前的selection sort改进成了heap sort的复杂度。时间复杂度O(Nlogk)。

Runtime: 4 ms, faster than 82.03% of Java online submissions for Merge k Sorted Lists.

Memory Usage: 40.1 MB, less than 98.23% of Java online submissions for Merge k Sorted Lists.

/**
 * Definition for singly-linked list.
 * public class ListNode {
 *     int val;
 *     ListNode next;
 *     ListNode() {}
 *     ListNode(int val) { this.val = val; }
 *     ListNode(int val, ListNode next) { this.val = val; this.next = next; }
 * }
 */
class Solution {
    public ListNode mergeKLists(ListNode[] lists) {
        ListNode result = new ListNode();
        ListNode curr = result;
        Queue<ListNode> pq = new PriorityQueue<>((a, b) -> a.val - b.val);
        for (ListNode node : lists) {
            // be sure to check if the node is null or not!!!
            if (node != null) {
                pq.add(node);
            }
        }
        
        while (!pq.isEmpty()) {
            ListNode small = pq.poll();
            curr.next = new ListNode(small.val);
            curr = curr.next;
            if (small.next != null) {
                pq.add(small.next);
            }
        }
        
        return result.next;
    }
}

2022.9.16 & 2024.7.10

几个null check又漏了，不run test case根本就想不到啊，头大。而且也完全把这个解法忘了。这里的主要思想是不把list当list看，而是当一个个单独的node看，于是pq的时间复杂度就是O(k)。

/**
 * Definition for singly-linked list.
 * public class ListNode {
 *     int val;
 *     ListNode next;
 *     ListNode() {}
 *     ListNode(int val) { this.val = val; }
 *     ListNode(int val, ListNode next) { this.val = val; this.next = next; }
 * }
 */
class Solution {
    public ListNode mergeKLists(ListNode[] lists) {
        ListNode dummy = new ListNode();
        ListNode curr = dummy;
        
        Queue<ListNode> pq = new PriorityQueue<>((a, b) -> a.val - b.val);

        for (ListNode list : lists) {
            if (list != null) {
                pq.add(list);
            }
        }

        while (!pq.isEmpty()) {
            ListNode top = pq.remove();
            curr.next = top;
            curr = curr.next;
            if (top.next != null) {
                pq.add(top.next);
            }
        }
        return dummy.next;
    }
}

---

按顺序merge两个list，并把merge好的放在最后面，相当于divide and conquer的iterative版，recursion版还没写。时间复杂度O(nk)。

2020.10.14 Java

Runtime: 180 ms, faster than 12.10% of Java online submissions for Merge k Sorted Lists.

Memory Usage: 40.8 MB, less than 15.23% of Java online submissions for Merge k Sorted Lists.

/**
 * Definition for singly-linked list.
 * public class ListNode {
 *     int val;
 *     ListNode next;
 *     ListNode() {}
 *     ListNode(int val) { this.val = val; }
 *     ListNode(int val, ListNode next) { this.val = val; this.next = next; }
 * }
 */
class Solution {
    public ListNode mergeKLists(ListNode[] lists) {
        if (lists.length == 0) {
            return null;
        }
        for (int i = 0; i < lists.length - 1; i++) {
            ListNode merged = mergeTwoLists(lists[i], lists[i + 1]);
            lists[i + 1] = merged;
        }
        return lists[lists.length - 1];
    }
    
    // copy from merge 2
    public ListNode mergeTwoLists(ListNode l1, ListNode l2) {
        ListNode dummyHead = new ListNode();
        ListNode curr = dummyHead;
        while (l1 != null || l2 != null) {
            if (l1 == null) {
                curr.next = l2;
                l2 = l2.next;
            } else if (l2 == null) {
                curr.next = l1;
                l1 = l1.next;
            } else {
                if (l1.val < l2.val) {
                    curr.next = l1;
                    l1 = l1.next;
                } else {
                    curr.next = l2;
                    l2 = l2.next;
                }
            }
            curr = curr.next;
        }
        return dummyHead.next;
    }
}

---

上面的改进版，更像是真正的divide and conquer，每次iteration挑第一个和最后一个进行merge，一次iteration后list的数目就降了一半，而前面的版本只降了1。

Runtime: 24 ms, faster than 91.12% of C++ online submissions for Merge k Sorted Lists.

Memory Usage: 11 MB, less than 73.81% of C++ online submissions for Merge k Sorted Lists.

/**
 * Definition for singly-linked list.
 * struct ListNode {
 *     int val;
 *     ListNode *next;
 *     ListNode(int x) : val(x), next(NULL) {}
 * };
 */
class Solution {
public:
    ListNode* merge2Lists(ListNode* l1, ListNode* l2) {
        ListNode* new_head = new ListNode(0);
        ListNode* curr1 = l1;
        ListNode* curr2 = l2;
        ListNode* curr_new = new_head;
        while (curr1 && curr2) {
            if (curr1->val < curr2->val) {
                curr_new->next = curr1;
                curr1 = curr1->next;
            }
            else {
                curr_new->next = curr2;
                curr2 = curr2->next;
            }
            curr_new = curr_new->next;
        }
        if (curr1) {
            curr_new->next = curr1;
        }
        else if (curr2) {
            curr_new->next = curr2;
        }
        return new_head->next;
    }
    
    ListNode* mergeKLists(vector<ListNode*>& lists) {
        if (lists.empty()) {
            return NULL;
        }
        ListNode *new_head = lists[0];
        int size = lists.size();
        while (size >= 2) {
            for (int i = 0; i < size / 2; i++) {
                lists[i] = merge2Lists(lists[i], lists[size - i - 1]);
            }
            size = (size + 1) / 2;
        }
        return lists[0];
    }
};