[LeeCode]Search for a Range, 解题报告

前言

最近快回家了，虽然用印象笔记管理自己每天的计划，仍然免不了每天有贪玩或者完不成当天任务的情况，当做年前放松了，年后去了云os实习，而且都是自己没接触过的新内容，压力还挺大

这篇博客主要是复习一下二分查找算法，记得有人曾经说过90%的程序员写不对二分查找，哈哈，我们争取做那10%

题目

Given a sorted array of integers, find the starting and ending position of a given target value.

Your algorithm's runtime complexity must be in the order of O(log n).

If the target is not found in the array, return [-1, -1].

For example,
Given [5, 7, 7, 8, 8, 10] and target value 8,
return [3, 4].

思路1&&AC代码1

首先是二分查找，然后根据二分查找的坐标分别向前、向后查找即可

平均时间复杂度是O（logn），但是最坏情况下可能达到O（n）（ps：当输入数据均相等，且target也跟输入数据相同时）

AC代码

public class Solution {
    public static int[] searchRange(int[] A, int target) {
        int res[] = new int[2], bt = 0, ed = A.length - 1, mid = 0;

        while (bt <= ed) {
            mid = bt + ((ed - bt) >> 1);
            if (A[mid] == target) {
                break;
            } else if (A[mid] > target) {
                ed = mid - 1;
            } else {
                bt = mid + 1;
            }
        }

        if (bt > ed) { // not found
            res[0] = res[1] = -1;
        } else { // found
            int is, ie;

            is = mid;
            while (is - 1 >= 0 && A[is - 1] == target) {
                is -= 1;
            }

            ie = mid;
            while (ie + 1 < A.length && A[ie + 1] == target) {
                ie += 1;
            }

            res[0] = is;
            res[1] = ie;
        }

        return res;
    }
}

思路2&&AC代码

两次二分查找，第一次确定开始点，第二次确定结束点，这样省去了思路1最后时间复杂度为O(n)的朝两个方向查找的过程

AC代码

public class Solution {

    public static int[] searchRange(int[] A, int target) {
        int res[] = new int[2], bt, ed, mid;

        res[0] = res[1] = -1;

        bt = mid = 0;
        ed = A.length - 1;
        while (bt < ed) {
            mid = bt + ((ed - bt) >> 1);
            if (A[mid] < target) {
                bt = mid + 1;
            } else {
                ed = mid;
            }
        }

        if (A[bt] != target) {
            return res;
        } else {
            res[0] = bt;
        }

        bt = mid = 0;
        ed = A.length;
        while (bt < ed) {
            mid = bt + ((ed - bt) >> 1);
            if (A[mid] > target) {
                ed = mid;
            } else {
                bt = mid + 1;
            }
        }

        res[1] = ed - 1;

        return res;
    }
}

个人更推荐思路1，思路2的两个二分查找比较难想，而且容易搞错边界条件，面试的时候很容易紧张，最好写自己最拿手并且思路清晰的代码！