// code by wapas
# pragma GCC optimize ("O3")
# pragma GCC optimize ("Ofast")
# pragma GCC optimize ("unroll-loops")
# pragma GCC target("sse,sse2,sse3,ssse3,sse4,avx,avx2")
#include <bits/stdc++.h>
using namespace std;
typedef long long ll;
/*
segtree code by : https://github.com/atcoder/ac-library/blob/master/atcoder/segtree.hpp
how to use : https://github.com/atcoder/ac-library/blob/master/document_en/segtree.md
*/
#if __cplusplus < 202002L
unsigned int bit_ceil(unsigned int n) {
unsigned int x = 1;
while (x < (unsigned int)(n)) x *= 2;
return x;
}
#endif
int countr_zero(unsigned int n) {
#ifdef _MSC_VER
unsigned long index;
_BitScanForward(&index, n);
return index;
#else
return __builtin_ctz(n);
#endif
}
template <class S, auto op, auto e> struct segtree {
static_assert(is_convertible_v<decltype(op), function<S(S, S)>>,
"op must work as S(S, S)");
static_assert(is_convertible_v<decltype(e), function<S()>>,
"e must work as S()");
public:
segtree() : segtree(0) {}
explicit segtree(int n) : segtree(vector<S>(n, e())) {}
explicit segtree(const vector<S>& v) : _n(int(v.size())) {
size = (int) bit_ceil((unsigned int)(_n));
log = countr_zero((unsigned int)size);
d = vector<S>(2 * size, e());
for (int i = 0; i < _n; i++) d[size + i] = v[i];
for (int i = size - 1; i >= 1; i--) {
update(i);
}
}
void set(int p, S x) {
assert(0 <= p && p < _n);
p += size;
d[p] = x;
for (int i = 1; i <= log; i++) update(p >> i);
}
S get(int p) const {
assert(0 <= p && p < _n);
return d[p + size];
}
S prod(int l, int r) const {
assert(0 <= l && l <= r && r <= _n);
S sml = e(), smr = e();
l += size;
r += size;
while (l < r) {
if (l & 1) sml = op(sml, d[l++]);
if (r & 1) smr = op(d[--r], smr);
l >>= 1;
r >>= 1;
}
return op(sml, smr);
}
S all_prod() const { return d[1]; }
template <bool (*f)(S)> int max_right(int l) const {
return max_right(l, [](S x) { return f(x); });
}
template <class F> int max_right(int l, F f) const {
assert(0 <= l && l <= _n);
assert(f(e()));
if (l == _n) return _n;
l += size;
S sm = e();
do {
while (l % 2 == 0) l >>= 1;
if (!f(op(sm, d[l]))) {
while (l < size) {
l = (2 * l);
if (f(op(sm, d[l]))) {
sm = op(sm, d[l]);
l++;
}
}
return l - size;
}
sm = op(sm, d[l]);
l++;
} while ((l & -l) != l);
return _n;
}
template <bool (*f)(S)> int min_left(int r) const {
return min_left(r, [](S x) { return f(x); });
}
template <class F> int min_left(int r, F f) const {
assert(0 <= r && r <= _n);
assert(f(e()));
if (r == 0) return 0;
r += size;
S sm = e();
do {
r--;
while (r > 1 && (r % 2)) r >>= 1;
if (!f(op(d[r], sm))) {
while (r < size) {
r = (2 * r + 1);
if (f(op(d[r], sm))) {
sm = op(d[r], sm);
r--;
}
}
return r + 1 - size;
}
sm = op(d[r], sm);
} while ((r & -r) != r);
return 0;
}
private:
int _n, size, log;
vector<S> d;
void update(int k) { d[k] = op(d[2 * k], d[2 * k + 1]); }
};
ll op(ll a, ll b) {
return a + b;
}
ll e() {
return 0;
}
// Bulldozer code by justiceHui : https://justicehui.github.io/hard-algorithm/2022/03/30/rotate-sweep-line/
struct point {
ll x, y, z;
bool operator < (const point &p) const {
return tie(x, y) < tie(p.x, p.y);
}
bool operator == (const point &p) const {
return tie(x, y) == tie(p.x, p.y);
}
};
struct line {
ll i, j, dx, dy;
line(int i, int j, const point &pi, const point &pj) : i(i), j(j), dx(pj.x - pi.x), dy(pj.y - pi.y) {}
bool operator < (const line &l) const {
ll le = dy * l.dx, ri = l.dy * dx;
return tie(le, i, j) < tie(ri, l.i, l.j);
}
bool operator == (const line &l) const {
return dy * l.dx == l.dy * dx;
}
};
point vec(point a, point b) {
return { b.x - a.x, b.y - a.y };
}
ll ccw(point a, point b, point c) {
point u = vec(a, b);
point v = vec(b, c);
return u.x * v.y - u.y * v.x;
}
void solution(int t) {
int N; cin >> N;
vector<point> prr(N);
for (int i = 0; i < N; i++) cin >> prr[i].x >> prr[i].y >> prr[i].z;
sort(prr.begin(), prr.end());
vector<segtree<ll, op, e>> st(3, segtree<ll, op, e>(N));
for (int i = 0; i < N; i++) st[prr[i].z].set(i, 1);
vector<int> pos(N);
for (int i = 0; i < N; i++) pos[i] = i;
vector<line> lrr;
for (int i = 0; i < N - 1; i++) for (int j = i + 1; j < N; j++) lrr.emplace_back(i, j, prr[i], prr[j]);
sort(lrr.begin(), lrr.end());
ll ans = 0;
for (int i = 0, j = 0; i < lrr.size(); i = j) {
while (j < lrr.size() && lrr[i] == lrr[j]) j++;
for (int k = i; k < j; k++) {
int u = lrr[k].i, v = lrr[k].j;
st[prr[pos[u]].z].set(pos[u], 0);
st[prr[pos[v]].z].set(pos[v], 0);
swap(pos[u], pos[v]);
swap(prr[pos[u]], prr[pos[v]]);
if (pos[u] > pos[v]) swap(u, v);
st[prr[pos[u]].z].set(pos[u], 1);
st[prr[pos[v]].z].set(pos[v], 1);
vector<ll> L = { 0, 0, 0 };
for (int i = 0; i < 3; i++) L[i] = st[i].prod(0, max(pos[u], 1));
ll left = 1;
for (int i = 0; i < 3; i++) if (i != prr[pos[u]].z) left *= L[i];
vector<ll> R = { 0, 0, 0 };
for (int i = 0; i < 3; i++) R[i] = st[i].prod(min(N - 1, pos[v] + 1), N);
ll right = 1;
for (int i = 0; i < 3; i++) if (i != prr[pos[v]].z) right *= R[i];
ans += left * right;
}
}
cout << ans;
}
int main() {
ios_base::sync_with_stdio(false); cin.tie(NULL); cout.tie(NULL);
// int T; cin >> T;
int T = 1;
for (int t = 0; t < T; t++) {
solution(t);
}
}
Compilation message
constellation2.cpp: In function 'void solution(int)':
constellation2.cpp:197:30: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<line>::size_type' {aka 'long unsigned int'} [-Wsign-compare]
197 | for (int i = 0, j = 0; i < lrr.size(); i = j) {
| ~~^~~~~~~~~~~~
constellation2.cpp:198:18: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<line>::size_type' {aka 'long unsigned int'} [-Wsign-compare]
198 | while (j < lrr.size() && lrr[i] == lrr[j]) j++;
| ~~^~~~~~~~~~~~
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Correct |
1 ms |
348 KB |
Output is correct |
2 |
Correct |
0 ms |
348 KB |
Output is correct |
3 |
Correct |
1 ms |
348 KB |
Output is correct |
4 |
Correct |
1 ms |
348 KB |
Output is correct |
5 |
Correct |
1 ms |
348 KB |
Output is correct |
6 |
Correct |
1 ms |
348 KB |
Output is correct |
7 |
Correct |
0 ms |
348 KB |
Output is correct |
8 |
Correct |
1 ms |
348 KB |
Output is correct |
9 |
Correct |
1 ms |
348 KB |
Output is correct |
10 |
Correct |
1 ms |
348 KB |
Output is correct |
11 |
Correct |
0 ms |
344 KB |
Output is correct |
12 |
Correct |
0 ms |
348 KB |
Output is correct |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Correct |
1 ms |
348 KB |
Output is correct |
2 |
Correct |
1 ms |
348 KB |
Output is correct |
3 |
Correct |
2 ms |
860 KB |
Output is correct |
4 |
Correct |
5 ms |
992 KB |
Output is correct |
5 |
Correct |
7 ms |
1500 KB |
Output is correct |
6 |
Correct |
21 ms |
2524 KB |
Output is correct |
7 |
Correct |
25 ms |
2524 KB |
Output is correct |
8 |
Correct |
16 ms |
2604 KB |
Output is correct |
9 |
Correct |
17 ms |
2520 KB |
Output is correct |
10 |
Correct |
12 ms |
1500 KB |
Output is correct |
11 |
Correct |
20 ms |
2660 KB |
Output is correct |
12 |
Correct |
18 ms |
2524 KB |
Output is correct |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Correct |
162 ms |
17092 KB |
Output is correct |
2 |
Correct |
196 ms |
17092 KB |
Output is correct |
3 |
Correct |
260 ms |
17092 KB |
Output is correct |
4 |
Correct |
254 ms |
17088 KB |
Output is correct |
5 |
Correct |
617 ms |
66472 KB |
Output is correct |
6 |
Correct |
1147 ms |
66420 KB |
Output is correct |
7 |
Correct |
1905 ms |
132188 KB |
Output is correct |
8 |
Runtime error |
156 ms |
262144 KB |
Execution killed with signal 9 |
9 |
Halted |
0 ms |
0 KB |
- |