978. Longest Turbulent Subarray

为什么80%的码农都做不了架构师？>>>   

用二维数组 B来记录之前比较的长度

public int maxTurbulenceSize(int[] A) {
        if (A == null || A.length == 0) return 0;
        if (A.length == 1) return 1;
        int res = 0;
        int[][] B = new int[A.length][2];
        B[0][1] = B[0][0] = 1;
        for (int i = 1; i < A.length; i++) {
            if (A[i - 1] > A[i]) {
		// A[i-2] < A[i-1] > A[i]
		// 即 B[i-1][1] = A[i-1]>A[i-2] 的长度
                B[i][0] = B[i - 1][1] + 1;
                B[i][1] = 1;
            } else if (A[i - 1] < A[i]) {
                B[i][0] = 1;
		// A[i-2] > A[i-1] < A[i]
		// 即 B[i-1][0] = A[i-1]<A[i-2] 的长度
                B[i][1] = B[i - 1][0] + 1;
            }else{
		// A[i-1] = A[i]，无法延长
                B[i][0] = B[i][1] = 1;
            }
            res = Math.max(res, Math.max(B[i][0],B[i][1]));
        }
        for(int[] a:B) {
            System.out.println(Arrays.toString(a));
        }
        return res;
    }

因为B数组前面的数据都没有用到，所以可以优化成类似: https://leetcode.com/problems/longest-turbulent-subarray/discuss/221935/Java-O(N)-time-O(1)-space 的代码

转载于:https://my.oschina.net/reter/blog/3044868