Problem Statement

A string $S$ consisting of uppercase English letters is said to be CPCTF-like if the length of $S$ is 5 and $S_i = S_j$ is equivalent to $i = j \vee \lbrace i,j \rbrace = \lbrace 1,3\rbrace$. (Here, $S_i$ denotes the $i$-th character of $S$.)

In other words, it is a string of length 5 consisting only of uppercase English letters in which only the $1$-st and $3$-rd characters are equal and every other character appears only once in $S$.

For example, $\mathrm{ABACD}$ is CPCTF-like, but $\mathrm{ABABC}$ is not.

You are given a string $S$ of length $N$ consisting of uppercase English letters. How many not necessarily contiguous substrings of $S$ are CPCTF-like?

Here, substrings taken from different positions are counted as different even if they are the same as strings.

Constraints

• $1 \leq N \leq 30$
• $S$ is a string of length $N$ consisting only of uppercase English letters

Input

The input is given from standard input in the following format.

$N$
$S$

Output

Output, in one line, the number of CPCTF-like not necessarily contiguous substrings.

Sample