Leetcode No.1014 Best SightSeeing Pair

题目：https://leetcode.com/problems/best-sightseeing-pair/

简介：

1. SightSeeing Score 计算：对于SightSeeing Spots的数组，The score of a pair (i < j) of sightseeing spots is (A[i] + A[j] + i - j) : the sum of the values of the sightseeing spots, minus the distance between them.
2. 很简单的计算方法，理论上遍历列表就可以计算，但是在面对大数据量是O(n^2)的时间复杂度一般都会超时，所以需要尽可能地减少列表遍历地次数，最佳目标时复是O(n)，所以这里就需要思考如何在单次遍历的前提下，用空间资源维护另外一个数组计算。

class Solution {
    public int maxScoreSightseeingPair(int[] A) {
        /*
        // Time out of limit 
        int maxScore = 0;
        int score = 0;
        for(int i = 0; i < A.length; i++){
            for(int j = i+1; j < A.length; j++){
                score = A[i] + A[j] + i - j;
                if(score > maxScore){
                    maxScore = score;
                }
            }
        }
        
        return maxScore;
        */
        
        if(A == null || A.length <= 1){ return 0;}
        // 设计两个数组分别维护从左向右和从右向左的最佳位置值
        int[] leftFrom = new int[A.length];
        int[] rightFrom = new int[A.length];
        
        for(int i = 0; i < A.length - 1; i++){
            if(i == 0){
                leftFrom[i] = i;
            }else{
                if (A[i] >= A[leftFrom[i-1]] || A[leftFrom[i-1]] - A[i] < i - leftFrom[i-1]){
                    leftFrom[i] = i;
                } else{
                    leftFrom[i] = leftFrom[i-1];
                }
            }
        }
        
        for(int i = A.length - 1; i > 0; i--){
            if(i == A.length - 1){
                rightFrom[i] = i;
            }else{
                if (A[i] >= A[rightFrom[i+1]] || A[rightFrom[i+1]] - A[i] < rightFrom[i+1] - i){
                    rightFrom[i] = i;
                } else{
                    rightFrom[i] = rightFrom[i+1];
                }
            }
        }
        
        int result = Integer.MIN_VALUE;
        for(int i = 1; i < A.length; i++){
            int sum = A[leftFrom[i-1]] + A[rightFrom[i]] + leftFrom[i-1] - rightFrom[i];
            result = Math.max(result, sum);
        }
        
        return result;
    }
}