// sparse table itself

vector<vector<T>> sparse_table;

void build_sparse_table(vector<T> arr) {

// get height of the table

int height = ceil(log(arr.size()) / log(2));

// sparse table is of size height * arr.size()

sparse_table.resize(arr.size());

for (size_t i = 0; i < sparse_table.size(); ++i) {

sparse_table[i].resize(height, 0);

sparse_table[i][0] = arr[i];

}

int arr_s = arr.size();

// building the table

for (int j = 1; (1 << j) <= arr.size(); ++j) {

for (int i = 0; (i + (1 << j) - 1) < arr.size(); ++i) {

// sparse_table[k][i] = min(sparse_table[k-1][i], sparse_table[k-1][i + pow(2, k-1)]);

sparse_table[i][j] = min(sparse_table[i][j - 1],

sparse_table[i + (1 << (j - 1))][j - 1]);

}

}

}

public:

// accepts an array of type T, for which sparse table is built

SparseTable(vector<T> arr) { build_sparse_table(arr); }

int query(int l, int r) {

int k = (int)log2(r - l + 1);

return min(sparse_table[l][k], sparse_table[r - (1 << k) + 1][k]);

}

// size of the input array and ariable to read array's element in

int n, x;

cin >> n;

vector<int> v;

for (int i = 0; i < n; ++i) {

cin >> x;

v.push_back(x);

}

SparseTable<int> st(v);

cout << "Enter l, r" << endl;

int l, r;

cin >> l >> r;

cout << st.query(l, r) << endl;

return 0;

}