B'Yeri has an array of n + 2 non-negative integers : a_0, a_1, ..., a_n, a_{n + 1} . We know that a_0 = a_{n + 1} = 0 . She wants to make all the elements of a equal to zero in the minimum number of operations. In one operation she can do one of the following: Help her find the minimum number of operations needed to make all elements of a equal to zero. The first line contains a single integer n ( 1 <= n <= 2 cdot 10^5 ). The second line contains n integers a_1, a_2, ldots, a_n ( 0 <= a_i <= n ). Print a single integer -- the minimum number of operations needed to make all elements of a equal to zero. In the first sample, you get langle 1, underline{1}, 2, 4, 0, 2 rangle by performing the first operation and langle 1, 4, 2, underline{2}, 0, 2 rangle by performing the second operation. One way to achieve our goal is shown below. (The underlines show the last change.) langle 1, 4, 2, 4, 0, 2 rangle to langle 1, 4, 2, underline{2}, 0, 2 rangle to langle 1, underline{1}, 2, 2, 0, 2 rangle to langle 1, 1, 2, 2, 0, underline{0} rangle to langle 1, 1, 2, underline{0}, 0, 0 rangle to langle 1, 1, underline{0}, 0, 0, 0 rangle to langle underline{0}, 1, 0, 0, 0, 0 rangle to langle 0, underline{0}, 0, 0, 0, 0 rangle In the third sample each element is already equal to zero so no operations are needed. '...