The test case
1 4 1 2 1 2
is accepted as a valid hack for problem A, but it clearly violates the constraints (It is guaranteed that all the numbers except one in the a array are the same). Could the problemsetters please fix this and rejudge A submissions?