Counting 1`s
Let b_i(x) be the i-th least significant bit of x, i.e. the i-th least significant digit of x in base 2 (i ≥ 1). For example, since 6 = (110)_2, b_1(6) = 0, b_2(6) = 1, b_3(6) = 1, b_4(6) = 0, b_5(6) = 0, and so on.
Let A and B (1 ≤ A ≤ B ≤ 10^18) be integers, and k_i be the number of integers x such that A ≤ x ≤ B and b_i(x) = 1.
Your task is to write a program that determines A and B for a given {k_i}.
Input
The input consists of multiple datasets. The number of datasets is no more than 100000. Each dataset has the following format:
n
k_1
k_2
...
k_n
The first line of each dataset contains an integer n (1 ≤ n ≤ 64). Then n lines follow, each of which contains k_i (0 ≤ k_i ≤ 2^63 - 1). For all i > n, k_i = 0.
The input is terminated by n = 0. Your program must not produce output for it.
Output
For each dataset, print one line.
If A and B can be uniquely determined, output A and B. Separate the numbers by a single space.
If there exists more than one possible pair of A and B, output "Many" (without quotes).
Otherwise, i.e. if there exists no possible pair, output "None" (without quotes).