You are given an array d1,d2,…,dnd1,d2,…,dn consisting of nn integer numbers.
Your task is to split this array into three parts (some of which may be empty) in such a way that each element of the array belongs to exactly one of the three parts, and each of the parts forms a consecutive contiguous subsegment (possibly, empty) of the original array.
Let the sum of elements of the first part be sum1sum1 , the sum of elements of the second part be sum2sum2 and the sum of elements of the third part be sum3sum3 . Among all possible ways to split the array you have to choose a way such that sum1=sum3sum1=sum3 and sum1sum1 is maximum possible.
More formally, if the first part of the array contains aa elements, the second part of the array contains bb elements and the third part contains cc elements, then:
sum1=∑1≤i≤adi,sum1=∑1≤i≤adi, sum2=∑a+1≤i≤a+bdi,sum2=∑a+1≤i≤a+bdi, sum3=∑a+b+1≤i≤a+b+cdi.sum3=∑a+b+1≤i≤a+b+cdi.
The sum of an empty array is 00 .
Your task is to find a way to split the array such that sum1=sum3sum1=sum3 and sum1sum1 is maximum possible.
Input
The first line of the input contains one integer nn (1≤n≤2⋅1051≤n≤2⋅105 ) — the number of elements in the array dd .
The second line of the input contains nn integers d1,d2,…,dnd1,d2,…,dn (1≤di≤1091≤di≤109 ) — the elements of the array dd .
Output
Print a single integer — the maximum possible value of sum1sum1 , considering that the condition sum1=sum3sum1=sum3 must be met.
Obviously, at least one valid way to split the array exists (use a=c=0a=c=0 and b=nb=n ).
Examples
Input
Copy
5 1 3 1 1 4
Output
Copy
5
Input
Copy
5 1 3 2 1 4
Output
Copy
4
Input
Copy
3 4 1 2
Output
Copy
0
Note
In the first example there is only one possible splitting which maximizes sum1sum1 : [1,3,1],[ ],[1,4][1,3,1],[ ],[1,4] .
In the second example the only way to have sum1=4sum1=4 is: [1,3],[2,1],[4][1,3],[2,1],[4] .
In the third example there is only one way to split the array: [ ],[4,1,2],[ ][ ],[4,1,2],[ ] .
#include<iostream>
#include<algorithm>
#include<string>
#include<map>//int dx[4]={0,0,-1,1};int dy[4]={-1,1,0,0};
#include<set>//int gcd(int a,int b){return b?gcd(b,a%b):a;}
#include<vector>
#include<cmath>
#include<stack>
#include<string.h>
#include<stdlib.h>
#include<cstdio>
#define mod 1e9+7
#define ll long long
#define maxn 200005
#define ms memset
using namespace std;
ll seq[maxn],n;
ll sum2[maxn];
/*
题目大意:给定一个n个数的序列,
分成三部分,当然任意部分可为空且分割连续。
在前部分等于后分割部分的情况下求其和的最大值。
二分思维,先记录前缀和和后缀和,
其中前缀和要用node数组记录,并附带下标,
对节点数组按值排序,然后遍历后缀和数组,
对每个后缀和在节点数组中二分查找,得到的下标和遍历的下标
如果满足和小于等于n就可以参与答案的比较。
*/
struct node
{
ll v,index;
node()
{
v=index=0;
}
};
node sum1[maxn];
bool cmp(node x,node y)
{
return x.v<y.v;
}
int main()
{
cin>>n;
for(int i=1;i<=n;i++) cin>>seq[i];
sum1[0].v=sum2[0]=0;
for(int i=1;i<=n-1;i++)
{
sum1[i].v=sum1[i-1].v+seq[i];
sum1[i].index=i;
sum2[i]=sum2[i-1]+seq[n-i+1];
}
sort(sum1+1,sum1+n,cmp);
ll op=0;
for(int i=1;i<=n-1;i++)
{
int ans=-1;
int l=1,r=n-1;
while(l<r)
{
int mid=(l+r)>>1;
// if(sum1[mid].v == sum2[i])
//{
// ans=sum1[mid].index;
// break;
//}
if(sum1[mid].v < sum2[i])
l=mid+1;
else
r=mid;
}
while(sum1[l].v==sum2[i])
{
if(l+i<=n)
op=max(op,sum2[i]);
l++;
}
}
cout<<op<<endl;
return 0;
}