浮点数二分 AtCoder Beginner Contest 174 E - Logs

Time Limit: 2 sec / Memory Limit: 1024 MB

Score : 500500 points

Problem Statement

We have NN logs of lengths A1,A2,⋯ANA1,A2,⋯AN.

We can cut these logs at most KK times in total. When a log of length LL is cut at a point whose distance from an end of the log is tt (0<t<L)(0<t<L), it becomes two logs of lengths tt and L−tL−t.

Find the shortest possible length of the longest log after at most KK cuts, and print it after rounding up to an integer.

Constraints

• 1≤N≤2×1051≤N≤2×105
• 0≤K≤1090≤K≤109
• 1≤Ai≤1091≤Ai≤109
• All values in input are integers.

---

Input

Input is given from Standard Input in the following format:

NN KK
A1A1 A2A2 ⋯⋯ ANAN

Output

Print an integer representing the answer.

---

Sample Input 1 Copy

Copy

2 3
7 9

Sample Output 1 Copy

Copy

4

• First, we will cut the log of length 77 at a point whose distance from an end of the log is 3.53.5, resulting in two logs of length 3.53.5 each.
• Next, we will cut the log of length 99 at a point whose distance from an end of the log is 33, resulting in two logs of length 33 and 66.
• Lastly, we will cut the log of length 66 at a point whose distance from an end of the log is 3.33.3, resulting in two logs of length 3.33.3 and 2.72.7.

In this case, the longest length of a log will be 3.53.5, which is the shortest possible result. After rounding up to an integer, the output should be 44.

---

Sample Input 2 Copy

Copy

3 0
3 4 5

Sample Output 2 Copy

Copy

5

---

Sample Input 3 Copy

Copy

10 10
158260522 877914575 602436426 24979445 861648772 623690081 433933447 476190629 262703497 211047202

Sample Output 3 Copy

Copy

292638192

#include<iostream>
#include<cmath>
#define ll long long

using namespace std;

int n,k;
double num[200010];
double eps = 1e-6;

int can(double len)
{
	ll rul = 0;
	for(int i=0; i<n; i++)
	{
		ll t = num[i] / len;
		if(fabs(num[i] - t * len) > eps) t++;
		rul+=t-1;
		if(rul > k) return 0;
	}	
	return 1;
}

int main()
{
	cin>>n>>k;
	for(int i=0; i<n; i++) cin>>num[i];
	
	double l = 1e-9, r = 1e9, mid;
	
	while(l < r - eps)
	{
		mid = (l + r) / 2;
		if(can(mid)) r = mid;
		else l = mid;
	}
	printf("%.0lf",ceil(r));
	return 0;
}