B'You are given array a_1, a_2, ldots, a_n , consisting of non-negative integers. Let 's define operation of "elimination" with integer parameter k ( 1 <= q k <= q n ) as follows: Find all possible values of k , such that it 's possible to make all elements of array a equal to 0 using a finite number of elimination operations with parameter k . It can be proven that exists at least one possible k for any array a . Note that you firstly choose k and only after that perform elimination operations with value k you 've chosen initially. Each test contains multiple test cases. The first line contains the number of test cases t ( 1 <= q t <= q 10^4 ). Description of the test cases follows. The first line of each test case contains one integer n ( 1 <= q n <= q 200 ,000 ) -- the length of array a . The second line of each test case contains n integers a_1, a_2, ldots, a_n ( 0 <= q a_i < 2^{30} ) -- array a itself. It 's guaranteed that the sum of n over all test cases doesn 't exceed 200 ,000 . For each test case, print all values k , such that it 's possible to make all elements of a equal to 0 in a finite number of elimination operations with the given parameter k . Print them in increasing order. In the first test case: In the second test case, if k = 2 then we can make the following elimination operations: Formal definition of bitwise AND: Let 's define bitwise AND ( & ) as follows. Suppose we have two non-negative integers x and y , let 's look at their binary representations (possibly, with leading zeroes): x_k ... x_2 x_1 x_0 and y_k ... y_2 y_1 y_0 . Here, x_i is the i -th bit of number x , and y_i is the i -th bit of number y . Let r = x ~ & ~ y is a result of operation & on number x and y . Then binary representation of r '...