下面这一段代码，有两行注释。如果把它们取消注释，输入最下面的那几行数字后得到的输出与直接调用下面的代码输出不一样，求原因。

#include<bits/stdc++.h>

using namespace std;

typedef long long ll;

const int N = 1e6 + 10;

inline int read(){
	int f = 0; char ch = getchar();
	while(ch < '0' || ch > '9') ch = getchar();
	while(ch >= '0' && ch <= '9'){
		f = f * 10 + ch - '0';
		ch = getchar();
	}
	return f;
}

int n;
int head[N], ver[N << 1], Next[N << 1], tot;
ll edge[N << 1];
bool e[N], vis[N];
int s;
int dota[N], cnt;
void add(int x, int y, int z){
	ver[++tot] = y; Next[tot] = head[x]; head[x] = tot; edge[tot] = z;
}
bool flag;
ll tmp, dis[N];
void Find(int x, int f, ll le){
	vis[x] = true;
	for(int i = head[x]; i && !flag; i = Next[i]){
		int y = ver[i];
		if(y == f) continue;
		if(vis[y]){
			s = y; e[x] = true;
			dota[++cnt] = x; tmp = edge[i];
			dis[cnt + 1] = le;
			flag = true; return;
		}
		Find(y, x, edge[i]);
	}
	vis[x] = false;
	if(flag){
		e[x] = true;
		dota[++cnt] = x;
		if(s != x) dis[cnt + 1] = dis[cnt] + le;
	}
	if(s == x) flag = false;
}
int d; ll len;
void dfs(int x, ll l){
	vis[x] = true;
	if(len < l) d = x, len = l;
	for(int i = head[x]; i; i = Next[i]){
		int y = ver[i];
		if(vis[y]) continue;
		dfs(y, l + edge[i]);
	}
	vis[x] = false;
}
ll half_pre[N], all_pre[N], half_sub[N], all_sub[N], f[N];
double solve(){
	ll ans1 = 0, ans2 = 0;
	for(int i = 1; i <= cnt; ++i){
		int x = dota[i];
		vis[x] = true;
		int sel;
		for(int j = head[x]; j; j = Next[j]){
			int y = ver[j];
			if(e[y]) continue;
			len = 0;
			dfs(y, edge[j]);
			if(len > f[i]){
				sel = f[i]; f[i] = len;
			}
			dfs(d, 0);
			ans1 = max(ans1, len);
		}
		ans1 = max(ans1, sel + f[i]);
		vis[x] = false;
	}
	ll maxn = 0;
	for(int i = 1; i <= cnt; ++i){
//		cerr << dota[i] << " to " << dota[1] << " : " << dis[i] << endl;
		half_pre[i] = max(half_pre[i - 1], dis[i] + f[i]);
		all_pre[i] = max(all_pre[i - 1], dis[i] + maxn + f[i]);
		maxn = max(maxn, f[i] - dis[i]);
	}
//	cerr << dota[cnt] << " to " << dota[1] << " : " << tmp << endl;
	maxn = 0;
	for(int i = cnt; i >= 1; --i){
		ll sub = dis[cnt] - dis[i];
		half_sub[i] = max(half_sub[i + 1], sub + f[i]);
		all_sub[i] = max(all_sub[i + 1], maxn + sub + f[i]);
		maxn = max(maxn, f[i] - sub);
	}
	ans2 = LLONG_MAX;
	for(int i = 1; i <= cnt; ++i) ans2 = min(ans2, max(all_pre[i], max(all_sub[i], half_pre[i] + half_sub[i+1]  + tmp)));
	return (double)max(ans1, ans2) / 2.0;
}

int main(){
	cin >> n;
	for(int i = 1, a, b, l; i <= n; ++i){
		cin >> a >> b >> l;
		add(a, b, l); add(b, a, l);
	}
	Find(1, 0, 0);
	printf("%.1f", solve());
	return 0;
}
/*
10
4 5 1
2 1 1
7 5 1
10 1 1
6 5 1
9 7 1
3 2 1
8 6 1
10 9 1
4 3 1
*/