#include <bits/stdc++.h>
#define P(x) do {if(debug) cout << x << endl;} while(false)
#define H(x) P(#x << ": " << x)
#define FR(i, a, b) for(int i = (a); i < (b); ++i)
#define F(i, n) FR(i, 0, n)
#define DR(i, a, b) for(int i = (b); i --> (a);)
#define D(i, n) DR(i, 0, n)
#define S(s) (int)(s).size()
#define ALL(x) (x).begin(), (x).end()
#define MI(x, a) (x) = min(x, a)
#define MA(x, a) (x) = max(x, a)
#define V vector
#define pb push_back
#define mp make_pair
using namespace std;
const bool debug = 0;
const long long inf = 1e18;
struct edge
{
int u, v, w;
bool operator== (edge o){
return u == o.u && v == o.v && w == o.w;
}
bool operator!= (edge o){
return !(*this == o);
}
};
struct sliding_maximum
{
deque<pair<long long, int> > qu;
int i, l;
sliding_maximum(int _l){
i = 0;
l = _l;
}
long long getmax(){
if(qu.empty()) return -inf;
return qu.front().first;
}
void slide(long long newval){
while(!qu.empty()&&newval > qu.back().first){
qu.pop_back();
}
qu.push_back({newval,i});
if(qu.front().second<=i-l){
qu.pop_front();
}
i++;
}
};
int N;
vector<edge> edgelist; //12 MB
vector<vector<int> > graph; //8 MB (16 MB worst case)?
vector<bool> vis; //1 MB
bool flag = false;
int dfs(int id, vector<int>& cycle){
edge p = edgelist[id];
H(p.u); H(p.v); H(p.w);
//F(ii, 1e8);
if(vis[p.v]&&flag){
P("return 1");
return -1;
}
if(vis[p.v]){
flag = true;
P("return 2");
P("cyclee: "<<p.v);
H(id);
cycle.push_back(id);
return p.v;
}
vis[p.v] = true;
int ce = -1;
for(int i : graph[p.v]) if(id != (i^1)) {
int a = dfs(i, cycle);
if(a!=-1) ce = a;
}
if(ce == p.v){
//cycle.push_back(cycle[0]);
P("return 3");
return -1;
}
if(ce != -1){
P("cycle: "<<p.v);
H(id);
cycle.push_back(id);
}
P("return 4: "<<ce);
return ce;
}
int f1, f2;
long long _maxu, _maxd;
void dfs2(int id, long long d){
//F(i, 1e8);
//edge p = edgelist[id];
//H(p.u); H(p.v); H(p.w);
//H(id); H(f1); H(f2);
if(d>_maxd){
_maxu = edgelist[id].v;
_maxd = d;
P("update (maxu, maxd) to");
H(_maxu); H(_maxd);
}
//int d = 0;
for(int i : graph[edgelist[id].v]) if(i-i%2 != id-id%2 && i-i%2 != f1-f1%2 && i-i%2 != f2-f2%2){
dfs2(i, d+edgelist[i].w);
}
//return d+p.w;
}
signed main() {
ios_base::sync_with_stdio(false);
cin.tie(0);
//input
cin>>N;
edgelist.resize(2*N+1);
graph.resize(N);
F(u, N){
int v,w;
cin>>v>>w;
v--;
edgelist[2*u] = {u, v, w};
graph[u].push_back(2*u);
edgelist[2*u+1] = {v, u, w};
graph[v].push_back(2*u+1);
}
//loop to calculate everything
vis.resize(N, 0);
edgelist[2*N] = {-1, 0, 0};
long long sum = 0;
F(t, N) if(!vis[t]){
P("--------------------------------------------");
H(t);
long long maxtot = 0;
vector<int> cycle; //4MB
flag = false;
edgelist[2*N].v = t;
dfs(2*N, cycle);
int c = cycle.size();
reverse(cycle.begin(), cycle.end()); //(0,1), (1,2), ..., (c-1, 0)
P("step 1");
for(int i : cycle){
edge p = edgelist[i];
P("cycle edge "<<i);
H(p.u);
H(p.v);
H(p.w);
}
vector<long long> d(c, 0); //8MB
F(i, c){
H(i);
edge p = edgelist[cycle[i]];
f1 = cycle[(i-1+c)%c];
f2 = cycle[i];
edgelist[2*N].v = p.u;
//long long maxu = -1, maxd = -1;
P("root: "<<p.u);
_maxu = -1;
_maxd = -1;
dfs2(2*N, 0);
d[i] = _maxd;
//long long maxu2 = -1, maxd2 = -1;
edgelist[2*N].v = _maxu;
_maxu = -1; _maxd = -1;
//P("root2: "<<maxu);
dfs2(2*N, 0);
maxtot = max(maxtot, _maxd);
}
H(maxtot);
P("step 2");
//vector<long long> pref(c*2+1, 0);
long long csum = 0;
F(i, 2*c-1){
//pref[i+1] = pref[i] + edgelist[cycle[i%c]].w;
csum = csum + edgelist[cycle[i%c]].w;
}
//csum == pref(c*2-1)
P("step 3");
sliding_maximum sm(c-1);
H(c);
for(int i = 2*c-1; i>=0; i--){
H(i);
maxtot = max((long long)maxtot, sm.getmax()-csum+d[i%c]);
sm.slide(csum+d[i%c]);
csum -= edgelist[cycle[(i+c-1)%c]].w;
}
P("step 4");
H(maxtot);
sum += maxtot;
}
cout<<sum<<endl;
return 0;
}
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
2 ms |
376 KB |
Output is correct |
2 |
Correct |
2 ms |
376 KB |
Output is correct |
3 |
Correct |
2 ms |
376 KB |
Output is correct |
4 |
Correct |
2 ms |
376 KB |
Output is correct |
5 |
Correct |
2 ms |
376 KB |
Output is correct |
6 |
Correct |
2 ms |
376 KB |
Output is correct |
7 |
Correct |
2 ms |
376 KB |
Output is correct |
8 |
Correct |
2 ms |
376 KB |
Output is correct |
9 |
Correct |
2 ms |
376 KB |
Output is correct |
10 |
Correct |
2 ms |
376 KB |
Output is correct |
11 |
Correct |
2 ms |
348 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
3 ms |
504 KB |
Output is correct |
2 |
Correct |
3 ms |
504 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
4 ms |
504 KB |
Output is correct |
2 |
Correct |
4 ms |
888 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
13 ms |
1912 KB |
Output is correct |
2 |
Correct |
28 ms |
4728 KB |
Output is correct |
3 |
Correct |
17 ms |
2296 KB |
Output is correct |
4 |
Correct |
10 ms |
1272 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
39 ms |
6500 KB |
Output is correct |
2 |
Correct |
57 ms |
11024 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
118 ms |
20344 KB |
Output is correct |
2 |
Correct |
144 ms |
27480 KB |
Output is correct |
3 |
Correct |
163 ms |
37356 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
205 ms |
38520 KB |
Output is correct |
2 |
Correct |
278 ms |
60548 KB |
Output is correct |
3 |
Correct |
321 ms |
76600 KB |
Output is correct |
4 |
Correct |
389 ms |
95208 KB |
Output is correct |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Correct |
425 ms |
98256 KB |
Output is correct |
2 |
Correct |
1140 ms |
131072 KB |
Output is correct |
3 |
Correct |
407 ms |
75548 KB |
Output is correct |
4 |
Correct |
559 ms |
114408 KB |
Output is correct |
5 |
Correct |
543 ms |
114044 KB |
Output is correct |
6 |
Correct |
1748 ms |
91180 KB |
Output is correct |
7 |
Runtime error |
589 ms |
131076 KB |
Execution killed with signal 9 (could be triggered by violating memory limits) |
# |
결과 |
실행 시간 |
메모리 |
Grader output |
1 |
Runtime error |
493 ms |
131072 KB |
Execution killed with signal 9 (could be triggered by violating memory limits) |
2 |
Halted |
0 ms |
0 KB |
- |