mirror of
https://github.com/d0zingcat/infloop.life.git
synced 2026-05-14 15:09:52 +00:00
67 lines
2.6 KiB
Markdown
67 lines
2.6 KiB
Markdown
---
|
||
title: leetcode-7-reverse-integer.md
|
||
tags: ["leetcode", "solution"]
|
||
date: 2019-11-26 20:56:46
|
||
---
|
||
|
||
7. Reverse Integer
|
||
Given a 32-bit signed integer, reverse digits of an integer.
|
||
Example 1:
|
||
|
||
Input: 123
|
||
Output: 321
|
||
Example 2:
|
||
|
||
Input: -123
|
||
Output: -321
|
||
Example 3:
|
||
|
||
Input: 120
|
||
Output: 21
|
||
|
||
Note:
|
||
Assume we are dealing with an environment which could only store integers within the 32-bit signed integer range: [−231, 231 − 1]. For the purpose of this problem, assume that your function returns 0 when the reversed integer overflows.
|
||
|
||
<!--more-->
|
||
|
||
题意很简单,就是数字反转,例如123-\>321。解法也比较简单,就是每次取除10的模数(然后除10取整),原数乘10累加。例如123就是1+10*((0+3)*10+2)=321。需要注意有两个点,一个是可能数字是120000,那么就不能简单地通过字符串反转来做(算数方法做是没问题的);另一个就是有边界,边界是4字节的有符号整型数字的范围,也就是 $$[-2^{-31}, 2^{31}-1]$$。
|
||
|
||
我的做法比较粗暴,判断还有一位的时候用边界去减了除10,和当前数比较大小即可(其实这么做有问题,如果test case长度不是2的31次方位,比如100位的数字,那也会最后一步才会判断是否溢出,对于python这种科学计算型语言不会有问题,因为底层已经处理了大数运算。对于C、Java等传统语言,这么写就会有问题。当然,也可以上来就手动判一下数字长度,如果大于10就直接返回0,不过我没写。),也没想到居然这么写过了,代码如:
|
||
|
||
```python
|
||
class Solution:
|
||
def reverse(self, x: int) -> int:
|
||
n = 0
|
||
flag = 1
|
||
if x < 0:
|
||
flag = -1
|
||
x = -x
|
||
while x > 0:
|
||
if x < 10:
|
||
if flag == 1 and n > ((1<<31)-1-x)/10 or flag == -1 and n > ((1<<31)-1)/10:
|
||
return 0
|
||
n = n * 10 + x % 10
|
||
x = x // 10
|
||
return flag * n
|
||
```
|
||
|
||
|
||
|
||
然后就是参考题解了。发现有两个地方可以改进代码。一个其实不用一开始就把负数转换为正数,因为带符号数累加之后符号还是在的;还有一个是判断条件可以做修改为判断为当前数是不是大于MAXINT/10,因为如果大于的话那么下一次x10就必然溢出;如果等于那么判断尾数是不是大于等于8(正数时)、小于等于-9(负数时)。
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
# Refer
|
||
|
||
[7. Reverse Integer](https://leetcode.com/problems/reverse-integer/)
|
||
|
||
[7. Reverse Integer](https://leetcode.com/articles/reverse-integer/)
|
||
|
||
# Appendix |