CodeForces - 363C Fixing Typos

Many modern text editors automatically check the spelling of the user's text. Some editors even suggest how to correct typos.

In this problem your task to implement a small functionality to correct two types of typos in a word. We will assume that three identical letters together is a typo (for example, word "helllo" contains a typo). Besides, a couple of identical letters immediately followed by another couple of identical letters is a typo too (for example, words "helloo" and "wwaatt" contain typos).

Write a code that deletes the minimum number of letters from a word, correcting described typos in the word. You are allowed to delete letters from both ends and from the middle of the word.

Input

The single line of the input contains word s, its length is from 1 to 200000 characters. The given word s consists of lowercase English letters.

Output

Print such word t that it doesn't contain any typos described in the problem statement and is obtained from s by deleting the least number of letters.

If there are multiple solutions, print any of them.

Example

Input

helloo

Output

hello

Input

woooooow

Output

woow

Note

The second valid answer to the test from the statement is "heloo".

思路:

当长度小于3符合题意了

下面我们就要判断多对在一起的情况和多个字母重复的情况,为了满足最小的删除次数,对于多对在一起的情况根据奇偶对我们可以得出要优先删除后面的,对于多个在一起的情况我们只需要判断 s[i]==s[i-1]&&s[i-1]==s[i-2] 即可,如果满足并不需要输出

但是对于多对在一起的情况我们需要注意的是 我们要删除的字母不需要去删除了,只需要将它前面的

字母赋值给它,使下一个字母判断时它前面的字母是调整后的而不是我们要删除的那个字母.

而下面的代码中一些字母由于相同我就省去了更新。

#include<stdio.h>
#include<iostream>
#include<string>
#include<string.h>
#include<math.h>
using namespace std;
int main()
{
	string s,ss;
	cin>>s;
	if(s.length()<3)
	{
		cout<<s;
	}
	else
	{
		printf("%c%c",s[0],s[1]);
		for(int i=2;i<s.length();i++)
		{
			if(s[i]==s[i-1]&&s[i-1]==s[i-2])
			{
				continue;
				
			}
			else if(i>=3)
			{
				if(s[i]==s[i-1]&&s[i-2]==s[i-3])
				{
					s[i-1]=s[i-2];
				}
				else
				printf("%c",s[i]);
			}
			else
			printf("%c",s[i]);
		}
	}
	return 0;
}