#include <bits/stdc++.h>

#define rep(i, j, k) for (int i = (j); i <= (k); ++i)
#define per(i, j, k) for (int i = (j); i >= (k); --i)
#define SZ(v) int((v).size())
#define ALL(v) (v).begin(),(v).end()
#define fi first
#define se second
using ll = long long;
using pii = std::pair<int, int>;
using pll = std::pair<ll, ll>;

template<class T>inline void chkmn(T &x, T y) { if (y < x) x = y; }
template<class T>inline void chkmx(T &x, T y) { if (y > x) x = y; }

using namespace std;

const int maxn = 1000010;

int n;
pii a[maxn];
double ans = 1e10;

double dis(pii x, pii y) {
  return sqrt(1ll * (x.fi - y.fi) * (x.fi - y.fi) + 1ll * (x.se - y.se) * (x.se - y.se));
}

void solve(int l, int r) {
  if (l == r) return;
  int mid = (l + r) >> 1;
  solve(l, mid), solve(mid + 1, r);
  double lim = ans / 2;
  vector<pii> pl, pr;
  rep (i, l, mid) if (fabs(a[i].fi - a[mid].fi) <= lim) pl.emplace_back(a[i]);
  rep (i, mid + 1, r) if (fabs(a[i].fi - a[mid].fi) <= lim) pr.emplace_back(a[i]);
  sort(ALL(pl), [&](pii x, pii y) { return x.se < y.se; });
  sort(ALL(pr), [&](pii x, pii y) { return x.se < y.se; });
  int L = 0, R = -1, sz = SZ(pr);
  for (pii x : pl) {
    while (L < sz && x.se - pr[L].se > lim) L++;
    while (R + 1 < sz && pr[R + 1].se - x.se <= lim) R++;
    rep (i, L, R) rep (j, i + 1, R) chkmn(ans, dis(x, pr[i]) + dis(x, pr[j]) + dis(pr[i], pr[j]));
  }
  L = 0, R = -1, sz = SZ(pl);
  for (pii x : pr) {
    while (L < sz && x.se - pl[L].se > lim) L++;
    while (R + 1 < sz && pl[R + 1].se - x.se <= lim) R++;
    rep (i, L, R) rep (j, i + 1, R) chkmn(ans, dis(x, pl[i]) + dis(x, pl[j]) + dis(pl[i], pl[j]));
  }
}

void work() {
  cin >> n, ans = 1e10;
  rep (i, 1, n) cin >> a[i].fi >> a[i].se;
  sort(a + 1, a + n + 1), solve(1, n);
  cout << fixed << setprecision(10) << ans << '\n';
}

int main() {
  cin.tie(nullptr) -> ios::sync_with_stdio(false);
  int t; cin >> t;
  rep (i, 1, t) cout << "Case #" << i << ": ", work();
}