LeetCode 题解工作台

统计没有收到请求的服务器数目

给你一个整数 n ,表示服务器的总数目,再给你一个下标从 0 开始的 二维 整数数组 logs ,其中 logs[i] = [server_id, time] 表示 id 为 server_id 的服务器在 time 时收到了一个请求。 同时给你一个整数 x 和一个下标从 0 开始的整数数组 que…

category

4

题型

code_blocks

5

代码语言

hub

3

相关题

当前训练重点

中等 · 数组·哈希·扫描

bolt

答案摘要

我们可以将所有的查询按照时间从小到大排序,然后按照时间顺序依次处理每个查询。 对于每个查询 $q = (r, i)$,其窗口左边界为 $l = r - x$,我们需要统计在窗口 $[l, r]$ 内有多少个服务器收到了请求。我们用双指针 和 分别维护窗口的左右边界,初始时 $j = k = 0$。每一次,如果 指向的日志的时间小于等于 ,我们就将其加入到窗口中,然后将 向右移动一位。如果 …

Interview AiBox logo

Interview AiBox 实时 AI 助手,陪你讲清 数组·哈希·扫描 题型思路

试试 AiBox 面试助手arrow_forward
description

题目描述

给你一个整数 n ,表示服务器的总数目,再给你一个下标从 0 开始的 二维 整数数组 logs ,其中 logs[i] = [server_id, time] 表示 id 为 server_id 的服务器在 time 时收到了一个请求。

同时给你一个整数 x 和一个下标从 0 开始的整数数组 queries  。

请你返回一个长度等于 queries.length 的数组 arr ,其中 arr[i] 表示在时间区间 [queries[i] - x, queries[i]] 内没有收到请求的服务器数目。

注意时间区间是个闭区间。

 

示例 1:

输入:n = 3, logs = [[1,3],[2,6],[1,5]], x = 5, queries = [10,11]
输出:[1,2]
解释:
对于 queries[0]:id 为 1 和 2 的服务器在区间 [5, 10] 内收到了请求,所以只有服务器 3 没有收到请求。
对于 queries[1]:id 为 2 的服务器在区间 [6,11] 内收到了请求,所以 id 为 1 和 3 的服务器在这个时间段内没有收到请求。

示例 2:

输入:n = 3, logs = [[2,4],[2,1],[1,2],[3,1]], x = 2, queries = [3,4]
输出:[0,1]
解释:
对于 queries[0]:区间 [1, 3] 内所有服务器都收到了请求。
对于 queries[1]:只有 id 为 3 的服务器在区间 [2,4] 内没有收到请求。

 

提示:

  • 1 <= n <= 105
  • 1 <= logs.length <= 105
  • 1 <= queries.length <= 105
  • logs[i].length == 2
  • 1 <= logs[i][0] <= n
  • 1 <= logs[i][1] <= 106
  • 1 <= x <= 105
  • x < queries[i] <= 106
lightbulb

解题思路

方法一:离线查询 + 排序 + 双指针

我们可以将所有的查询按照时间从小到大排序,然后按照时间顺序依次处理每个查询。

对于每个查询 q=(r,i)q = (r, i),其窗口左边界为 l=rxl = r - x,我们需要统计在窗口 [l,r][l, r] 内有多少个服务器收到了请求。我们用双指针 jjkk 分别维护窗口的左右边界,初始时 j=k=0j = k = 0。每一次,如果 kk 指向的日志的时间小于等于 rr,我们就将其加入到窗口中,然后将 kk 向右移动一位。如果 jj 指向的日志的时间小于 ll,我们就将其从窗口中移除,然后将 jj 向右移动一位。在移动的过程中,我们需要统计窗口中有多少个不同的服务器,这可以使用哈希表来实现。移动结束后,当前时间区间中没有收到请求的服务器数目就是 nn 减去哈希表中不同的服务器数目。

时间复杂度 O(l×logl+m×logm+n)O(l \times \log l + m \times \log m + n),空间复杂度 O(l+m)O(l + m)。其中 llnn 分别是数组 logs\textit{logs} 的长度和服务器的数量,而 mm 是数组 queries\textit{queries} 的长度。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
class Solution:
    def countServers(
        self, n: int, logs: List[List[int]], x: int, queries: List[int]
    ) -> List[int]:
        cnt = Counter()
        logs.sort(key=lambda x: x[1])
        ans = [0] * len(queries)
        j = k = 0
        for r, i in sorted(zip(queries, count())):
            l = r - x
            while k < len(logs) and logs[k][1] <= r:
                cnt[logs[k][0]] += 1
                k += 1
            while j < len(logs) and logs[j][1] < l:
                cnt[logs[j][0]] -= 1
                if cnt[logs[j][0]] == 0:
                    cnt.pop(logs[j][0])
                j += 1
            ans[i] = n - len(cnt)
        return ans
speed

复杂度分析

指标
时间Depends on the final approach
空间Depends on the final approach
psychology

面试官常问的追问

外企场景
  • question_mark

    The candidate demonstrates strong problem-solving ability by selecting appropriate data structures like hash tables and sliding windows.

  • question_mark

    The candidate efficiently handles large input sizes and provides optimized solutions without redundant operations.

  • question_mark

    The candidate can explain the trade-offs between brute-force and optimized solutions, discussing performance improvements and edge cases.

warning

常见陷阱

外企场景
  • error

    Overlooking time complexity when using brute force, leading to excessive computation for large inputs.

  • error

    Not correctly managing the time window during query processing, which can cause incorrect results for certain edge cases.

  • error

    Failing to optimize space complexity when storing request logs, especially for large datasets.

swap_horiz

进阶变体

外企场景
  • arrow_right_alt

    Increase the size of the server pool and logs while maintaining query complexity.

  • arrow_right_alt

    Adjust the query window size dynamically based on server request patterns.

  • arrow_right_alt

    Add constraints where logs include server downtime or maintenance periods that must be considered.

help

常见问题

外企场景

统计没有收到请求的服务器数目题解:数组·哈希·扫描 | LeetCode #2747 中等