#pragma GCC optimize ("O3")
#pragma GCC target ("sse4")
#include <bits/stdc++.h>
using namespace std;
typedef long long ll;
typedef long double ld;
typedef vector<int> vi;
typedef pair<int, int> pi;
#define debug(x) cerr << #x << ": " << x << endl;
#define debug2(x, y) debug(x) debug(y);
#define repn(i, a, b) for(int i = (int)(a); i < (int)(b); i++)
#define rep(i, a) for(int i = 0; i < (int)(a); i++)
#define all(v) v.begin(), v.end()
#define mp make_pair
#define pb push_back
#define lb lower_bound
#define ub upper_bound
#define fi first
#define se second
#define sq(x) ((x) * (x))
template<class T> T gcd(T a, T b){ return ((b == 0) ? a : gcd(b, a % b)); }
ll n, m;
vi vis2;
vi g[100005];
ll ans1 = 0;
int vis1[100005];
int vis3[100005];
int ind[100005];
int last[100005];
ll cn[100005];
ll d[100005];
int id = 0;
ll cc = 0;
ll ans2 = 0;
int vis4[100005];
ll cnt1[2];
bool dfs(int cur, int mid, int fin, bool fnd, vi vis){
vis[cur] = 1;
if(cur == mid) fnd = 1;
if(cur == fin){
if(fnd) return 1;
return 0;
}
int ret = 0;
for(int u : g[cur]) if(!vis[u]){
ret |= dfs(u, mid, fin, fnd, vis);
if(ret) break;
}
return ret;
}
bool check(int st, int mid, int fin){
vis2.assign(n, 0);
return dfs(st, mid, fin, 0, vis2);
}
void dfs1(int cur){
//initial traversal to compute the subtree sizes
vis1[cur] = 1;
ind[cur] = id;
last[id] = cur;
cc++;
d[cur] = 1;
for(int u : g[cur]) if(!vis1[u]){
dfs1(u);
d[cur] += d[u];
}
}
void dfs2(int cur){
vis3[cur] = 1;
ll sm = 0;
for(int u : g[cur]) if(!vis3[u]){
ans1 += (ll)(d[u] * (cn[ind[cur]] - d[u] - 1));
sm += d[u];
}
ans1 += (ll)((cn[ind[cur]] - d[cur]) * sm);
for(int u : g[cur]) if(!vis3[u]) dfs2(u);
}
bool dfs4(int cur, int prev){
vis4[cur] = 1;
if(prev != -1){
cnt1[1] += cnt1[0];
cnt1[0]++;
}
int ret = 0;
for(int u : g[cur]) if(u != prev && (ind[u] == ind[cur])){
if(vis4[u]) return 1;
ret |= dfs4(u, cur);
}
return ret;
}
int main(){
ios_base::sync_with_stdio(false);
cin.tie(0);
//freopen("1.out", "r", stdin);
//freopen("output.out", "w", stdout);
cin >> n >> m;
int mx = 0;
rep(i, m){
int a, b;
cin >> a >> b;
a--, b--;
g[a].pb(b);
g[b].pb(a);
mx = max({mx, (int)g[a].size(), (int)g[b].size()});
}
if(n <= 10 && m <= 100){
//first subtask
ll ans = 0;
rep(i, n) repn(k, i + 1, n) rep(j, n) if(j != i && j != k) if(check(i, j, k)) ans += 2;
cout << ans << endl;
}
else{
//subtasks 3, 4, 5
memset(vis1, 0, sizeof(vis1));
memset(vis3, 0, sizeof(vis3));
memset(vis4, 0, sizeof(vis4));
memset(d, 0, sizeof(d));
rep(i, n) if(!vis1[i]){
cc = 0;
dfs1(i);
cn[id] = cc;
id++;
}
if(mx == 2){
vi cyc(100005, 0);
rep(i, id){
if(dfs4(last[i], -1)) cyc[i] = 1;
else{
for(ll j = 2; j < cn[i]; j++) ans2 += (ll)(2 * (j - 1) * (cn[i] - j) * 1LL);
}
}
rep(i, id) if(cyc[i]){
ans2 += ((cn[i]) * (cn[i] - 1) * (cn[i] - 2));
}
cout << ans2 << endl;
return 0;
}
rep(i, n) if(!vis3[i]) dfs2(i);
cout << ans1 << endl;
}
return 0;
}
/*
Things to look out for:
- Integer overflows
- Array bounds
- Special cases
Be careful!
*/
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Incorrect |
4 ms |
2680 KB |
Output isn't correct |
2 |
Halted |
0 ms |
0 KB |
- |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Runtime error |
8 ms |
5240 KB |
Execution killed with signal 11 (could be triggered by violating memory limits) |
2 |
Halted |
0 ms |
0 KB |
- |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Incorrect |
6 ms |
5112 KB |
Output isn't correct |
2 |
Halted |
0 ms |
0 KB |
- |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Runtime error |
8 ms |
5368 KB |
Execution killed with signal 11 (could be triggered by violating memory limits) |
2 |
Halted |
0 ms |
0 KB |
- |
# |
Verdict |
Execution time |
Memory |
Grader output |
1 |
Incorrect |
4 ms |
2680 KB |
Output isn't correct |
2 |
Halted |
0 ms |
0 KB |
- |