#include <bits/stdc++.h>
using namespace std;
class fraction_t {
public:
int64_t x, y;
// x / y, normalized to y > 0 if x != 0 and y = 1 if x = 0
void fix() {
assert(y && "Your Denominator Equal 0");
if (y < 0) {
y = -y;
x = -x;
}
int64_t d = __gcd(abs(x), y);
x /= d;
y /= d;
}
fraction_t(const int64_t& x = 0, const int64_t& y = 1) : x(x), y(y) { fix(); }
friend bool operator<(const fraction_t& e, const fraction_t& f) { return e.x * f.y < e.y * f.x; }
friend bool operator<=(const fraction_t& e, const fraction_t& f) { return e.x * f.y <= e.y * f.x; }
friend bool operator==(const fraction_t& e, const fraction_t& f) { return e.x * f.y == e.y * f.x; }
friend bool operator!=(const fraction_t& e, const fraction_t& f) { return e.x * f.y != e.y * f.x; }
friend bool operator>(const fraction_t& e, const fraction_t& f) { return e.x * f.y > e.y * f.x; }
friend bool operator>=(const fraction_t& e, const fraction_t& f) { return e.x * f.y >= e.y * f.x; }
friend ostream& operator<<(ostream& out, const fraction_t& f) {
out << setprecision(11) << fixed << ((double)f.x) / f.y;
return out;
}
};
int32_t main() {
ios_base::sync_with_stdio(0);
cin.tie(0);
int n;
cin >> n;
vector<int> a(n);
for (int i = 0; i < n; i++) cin >> a[i];
vector<int64_t> pref(n + 1, 0);
for (int i = 0; i < n; i++) pref[i + 1] = pref[i] + a[i];
fraction_t c(pref[n] - pref[0], n);
set<pair<fraction_t, int>> s;
vector<int> nxt(n, -1);
for (int i = n - 1; i > 0; i--) {
int l = 0, r = n - i, fl;
while (l <= r) {
int mid = l + r >> 1;
if (fraction_t(pref[mid + i] - pref[mid], i) <= c) {
fl = mid;
l = mid + 1;
} else {
r = mid - 1;
}
}
fraction_t f(pref[fl + i] - pref[fl], i);
if (f == c) {
nxt[i] = -1;
} else {
assert(fl + 1 + i <= n);
nxt[i] = fl + 1;
}
s.emplace(f, i);
}
s.emplace(c, 0);
auto sub = [&](fraction_t a, fraction_t b) {
int64_t ab = a.y / __gcd(a.y, b.y) * b.y;
return fraction_t(a.x * ab / a.y - b.x * ab / b.y, ab);
};
fraction_t res(1e18, 1);
while (true) {
auto [f, i] = *s.begin();
res = min(res, sub(s.rbegin()->first, s.begin()->first));
s.erase(s.begin());
if (nxt[i] == -1) break;
s.emplace(fraction_t(pref[nxt[i] + i] - pref[nxt[i]], i), i);
nxt[i] = -1;
}
cout << res;
}
Compilation message
seesaw.cpp: In function 'int32_t main()':
seesaw.cpp:50:37: warning: suggest parentheses around '+' inside '>>' [-Wparentheses]
50 | int mid = l + r >> 1;
| ~~^~~
seesaw.cpp:59:52: warning: 'fl' may be used uninitialized in this function [-Wmaybe-uninitialized]
59 | fraction_t f(pref[fl + i] - pref[fl], i);
| ^
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
0 ms |
348 KB |
Output is correct |
2 |
Correct |
0 ms |
348 KB |
Output is correct |
3 |
Correct |
0 ms |
348 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
0 ms |
348 KB |
Output is correct |
2 |
Correct |
0 ms |
348 KB |
Output is correct |
3 |
Correct |
0 ms |
348 KB |
Output is correct |
4 |
Correct |
0 ms |
348 KB |
Output is correct |
5 |
Incorrect |
0 ms |
456 KB |
Output isn't correct |
6 |
Halted |
0 ms |
0 KB |
- |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
0 ms |
348 KB |
Output is correct |
2 |
Correct |
0 ms |
348 KB |
Output is correct |
3 |
Correct |
0 ms |
348 KB |
Output is correct |
4 |
Correct |
0 ms |
348 KB |
Output is correct |
5 |
Incorrect |
0 ms |
456 KB |
Output isn't correct |
6 |
Halted |
0 ms |
0 KB |
- |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
0 ms |
348 KB |
Output is correct |
2 |
Correct |
0 ms |
348 KB |
Output is correct |
3 |
Correct |
0 ms |
348 KB |
Output is correct |
4 |
Correct |
0 ms |
348 KB |
Output is correct |
5 |
Incorrect |
0 ms |
456 KB |
Output isn't correct |
6 |
Halted |
0 ms |
0 KB |
- |