Leetcode - Unique Binary Search Trees

题注

这道题本身不难，不过这次涉及到一些数学分析，很基础。LeetCode的题目本身都比较基础，但是正如《灌篮高手》里面赤木刚宪对樱木花道说的：基础最重要。没有掌握好的代码基础，剩下的一切都是扯淡。

题目

Given n, how many structurally unique BST's (binary search trees) that store values 1...n?

For example,
Given n = 3, there are a total of 5 unique BST's.

   1         3     3      2      1
    \       /     /      / \      \
     3     2     1      1   3      2
    /     /       \                 \
   2     1         2                 3

补充知识：Binary Search Trees

A binary search tree (BST), sometimes also called anordered orsorted binary tree, is a node-based binary data structure which has the following properties:

• 1. The left subtree of a node contains only nodes with keys less than the node's key.
• 2. The right subtree of a node contains only nodes with keys greater than the node's key.
• 3. The left and right subtree each must also be a binary search tree.
• 4. There must be no duplicate nodes.

分析

Binary Search Tree的最大特点是节点左子树所有节点值小于节点值本身；又子树所有节点值大于节点值本身。Binary Search Tree也有递归特性，因此自然而然地，这道题用递归特性（更确切的说，分治方法）来做最为合适。

如何递归呢？我们设函数$numTrees(n)$为存储了$[1, n]$的树中，Binary Search Tree的数量。显然地，我们有：

\[numTrees(0) = 0\]

\[numTrees(1) = 1\]

当$n \geq 2$时，我们分别将root节点的值设为$1,2, \cdots n$。当root节点值为$i \in [1, n]$时，其左子树的值必然为全部的$[1, i - 1]$（由性质1所确定），其右子树的值必然为全部的$[i+1, n]$（由性质2所确定）。同时，左右子树又必然为Binary Search Tree（由性质3所决定）。因此，左子树便成为了一个numTrees(i - 1)问题，右子树便称为了一个numTrees(n - 1 - i)问题。形式化的说，我们有，对于所有的$n \geq 2$：

\[numTrees(n) = numTrees(0) \cdot numTrees(n-1) + numTrees(1) \cdot numTrees(n-2) + \cdots + numTrees(n-1) \codt numTrees(0)\]

问题就迎刃而解了。

不过，为了简单地递归，我们这里设为$numTrees(0) = 1$，这样$numTrees(0) \cdot numTrees(n-1)$便不为$0$了，方便进行处理。

代码

public class Solution {
    public int reverse(int x) {
        boolean isNeg;
        if (x < 0){
            isNeg = true;
            x = -1 * x;
        }else{
            isNeg = false;
        }
        int reverResult = 0;
        while (x != 0){
            reverResult = reverResult * 10 + x % 10;
            x = x / 10;
        }
        if (isNeg){
            reverResult = -1 * reverResult;
        }
        return reverResult;
    }
}