[LeetCode] 42. Trapping Rain Water 接雨水

 

Given n non-negative integers representing an elevation map where the width of each bar is 1, compute how much water it can trap after raining.

Example 1:

Input: height = [0,1,0,2,1,0,1,3,2,1,2,1]
Output: 6
Explanation: The above elevation map (black section) is represented by array [0,1,0,2,1,0,1,3,2,1,2,1]. In this case, 6 units of rain water (blue section) are being trapped.

Example 2:

Input: height = [4,2,0,3,2,5]
Output: 9

Constraints:

  • n == height.length
  • 1 <= n <= 2 * 10^4
  • 0 <= height[i] <= 10^5

 

这道收集雨水的题跟之前的那道 Largest Rectangle in Histogram 有些类似,但是又不太一样,先来看一种方法,这种方法是基于动态规划 Dynamic Programming 的,维护一个一维的 dp 数组,这个 DP 算法需要遍历两遍数组,第一遍在 dp[i] 中存入i位置左边的最大值,然后开始第二遍遍历数组,第二次遍历时找右边最大值,然后和左边最大值比较取其中的较小值,然后跟当前值 A[i] 相比,如果大于当前值,则将差值存入结果,参见代码如下:

 

C++ 解法一:

class Solution {
public:
    int trap(vector<int>& height) {
        int res = 0, mx = 0, n = height.size();
        vector<int> dp(n);
        for (int i = 0; i < n; ++i) {
            dp[i] = mx;
            mx = max(mx, height[i]);
        }
        mx = 0;
        for (int i = n - 1; i >= 0; --i) {
            dp[i] = min(dp[i], mx);
            mx = max(mx, height[i]);
            if (dp[i] > height[i]) res += dp[i] - height[i];
        }
        return res;
    }
};

 

Java 解法一:

public class Solution {
    public int trap(int[] height) {
        int res = 0, mx = 0, n = height.length;
        int[] dp = new int[n];
        for (int i = 0; i < n; ++i) {
            dp[i] = mx;
            mx = Math.max(mx, height[i]);
        }
        mx = 0;
        for (int i = n - 1; i >= 0; --i) {
            dp[i] = Math.min(dp[i], mx);
            mx = Math.max(mx, height[i]);
            if (dp[i] - height[i] > 0) res += dp[i] - height[i];
        }
        return res;
    }
}

 

再看一种只需要遍历一次即可的解法,这个算法需要 left 和 right 两个指针分别指向数组的首尾位置,从两边向中间扫描,在当前两指针确定的范围内,先比较两头找出较小值,如果较小值是 left 指向的值,则从左向右扫描,如果较小值是 right 指向的值,则从右向左扫描,若遇到的值比当较小值小,则将差值存入结果,如遇到的值大,则重新确定新的窗口范围,以此类推直至 left 和 right 指针重合,参见代码如下:

 

C++ 解法二:

class Solution {
public:
    int trap(vector<int>& height) {
        int res = 0, left = 0, right = (int)height.size() - 1;
        while (left < right) {
            int mn = min(height[left], height[right]);
            if (mn == height[left]) {
                ++left;
                while (left < right && height[left] < mn) {
                    res += mn - height[left++];
                }
            } else {
                --right;
                while (left < right && height[right] < mn) {
                    res += mn - height[right--];
                }
            }
        }
        return res;
    }
};

 

Java 解法二:

public class Solution {
    public int trap(int[] height) {
        int res = 0, l = 0, r = height.length - 1;
        while (l < r) {
            int mn = Math.min(height[l], height[r]);
            if (height[l] == mn) {
                ++l;
                while (l < r && height[l] < mn) {
                    res += mn - height[l++];
                }
            } else {
                --r;
                while (l < r && height[r] < mn) {
                    res += mn - height[r--];
                }
            }
        }
        return res;
    }
}

 

我们可以对上面的解法进行进一步优化,使其更加简洁:

 

C++ 解法三:

class Solution {
public:
    int trap(vector<int>& height) {
        int left = 0, right = (int)height.size() - 1, level = 0, res = 0;
        while (left < right) {
            int lower = height[(height[left] < height[right]) ? left++ : right--];
            level = max(level, lower);
            res += level - lower;
        }
        return res;
    }
};

 

Java 解法三:

public class Solution {
    public int trap(int[] height) {
        int l = 0, r = height.length - 1, level = 0, res = 0;
        while (l < r) {
            int lower = height[(height[l] < height[r]) ? l++ : r--];
            level = Math.max(level, lower);
            res += level - lower;
        }
        return res;
    }
}

 

下面这种解法是用 stack 来做的,博主一开始都没有注意到这道题的 tag 还有 stack,所以以后在总结的时候还是要多多留意一下标签啊。其实用 stack 的方法博主感觉更容易理解,思路是,遍历高度,如果此时栈为空,或者当前高度小于等于栈顶高度,则把当前高度的坐标压入栈,注意这里不直接把高度压入栈,而是把坐标压入栈,这样方便在后来算水平距离。当遇到比栈顶高度大的时候,就说明有可能会有坑存在,可以装雨水。此时栈里至少有一个高度,如果只有一个的话,那么不能形成坑,直接跳过,如果多余一个的话,那么此时把栈顶元素取出来当作坑,新的栈顶元素就是左边界,当前高度是右边界,只要取二者较小的,减去坑的高度,长度就是右边界坐标减去左边界坐标再减1,二者相乘就是盛水量啦,参见代码如下:

 

C++ 解法四:

class Solution {
public:
    int trap(vector<int>& height) {
        stack<int> st;
        int i = 0, res = 0, n = height.size();
        while (i < n) {
            if (st.empty() || height[i] <= height[st.top()]) {
                st.push(i++);
            } else {
                int t = st.top(); st.pop();
                if (st.empty()) continue;
                res += (min(height[i], height[st.top()]) - height[t]) * (i - st.top() - 1);
            }
        }
        return res;
    }
};

 

Java 解法四:

class Solution {
    public int trap(int[] height) {
        Stack<Integer> s = new Stack<Integer>();
        int i = 0, n = height.length, res = 0;
        while (i < n) {
            if (s.isEmpty() || height[i] <= height[s.peek()]) {
                s.push(i++);
            } else {
                int t = s.pop();
                if (s.isEmpty()) continue;
                res += (Math.min(height[i], height[s.peek()]) - height[t]) * (i - s.peek() - 1);
            }
        }
        return res;
    }
}

 

Github 同步地址:

https://github.com/grandyang/leetcode/issues/42

 

类似题目:

Trapping Rain Water II 

Container With Most Water

Product of Array Except Self

Pour Water

Maximum Value of an Ordered Triplet II

 

参考资料:

https://leetcode.com/problems/trapping-rain-water/

https://leetcode.com/problems/trapping-rain-water/discuss/17364/7-lines-C-C%2B%2B

https://leetcode.com/problems/trapping-rain-water/discuss/17414/A-stack-based-solution-for-reference-inspired-by-Histogram

https://leetcode.com/problems/trapping-rain-water/discuss/17357/Sharing-my-simple-c%2B%2B-code%3A-O(n)-time-O(1)-space

 

LeetCode All in One 题目讲解汇总(持续更新中...)

posted @ 2015-04-08 14:46  Grandyang  阅读(43244)  评论(8编辑  收藏  举报
Fork me on GitHub