题解 | #寻找最合适的生育区域#

import java.util.*;


public class Solution {
    /**
     * 代码中的类名、方法名、参数名已经指定，请勿修改，直接返回方法规定的值即可
     *
     * 
     * @param heights int整型一维数组 
     * @param k int整型 
     * @return int整型
     */
    public int findMaxRangeWithinThreshold (int[] heights, int k) {
        // write code here
	// 小根堆
        PriorityQueue<Integer> pq = new PriorityQueue<>((a, b) -> (a - b));
        // 大根堆
        PriorityQueue<Integer> pq2 = new PriorityQueue<>((a, b) -> (b - a));
        int l = 0, res = Integer.MIN_VALUE;
	// 删除数据记录
        HashMap<Integer, Integer> rec = new HashMap<>();
        for (int r = 0; r < heights.length; r++) {
            pq.offer(heights[r]);
            pq2.offer(heights[r]);
            while (pq2.peek() - pq.peek() > k) {
		// 收缩左边界
                int x = heights[l++];
                rec.put(x, rec.getOrDefault(x, 0) + 1);
                while (!pq.isEmpty() && rec.containsKey(pq.peek())) {
                    int temp = pq.poll();
                    rec.put(temp, rec.get(temp) - 1);
                    if (rec.get(temp) == 0)
                        rec.remove(temp);
                }
                while (!pq2.isEmpty() && rec.containsKey(pq2.peek())) {
                    int temp = pq2.poll();
                    rec.put(temp, rec.get(temp) - 1);
                    if (rec.get(temp) == 0)
                        rec.remove(temp);
                }
            }
            res = Math.max(res, r - l + 1);
        }
        return res;
    }
}