Problem Description

Prof. Pang said, "gshfd1jkhaRaadfglkjerVcvuy0gf".

To understand Prof. Pang's word, we want to calculate the number of namomo subsequences in it. Prof. Pang's word is a string \( s \) with \( n \) characters, where each character is either an English letter (lower or upper case) or a digit. The \( i \)-th character of \( s \) is denoted by \( s[i] \) (\( 1 \le i \le n \)). A subsequence \( t \) of \( s \) is defined by a list of indices \( t_1, \ldots, t_6 \) such that \( 1 \le t_1 < t_2 < \ldots < t_6 \le n \). Let \( compare(c_1, c_2) \) be a function on two characters such that \( compare(c_1, c_2) = 1 \) when \( c_1 = c_2 \) and \( compare(c_1, c_2) = 0 \) otherwise. \( t \) is a namomo subsequence of \( s \) if and only if for any \( 1 \le i < j \le 6 \), \( compare(s[t_i], s[t_j]) = compare(namomo[i], namomo[j]) \), where \( namomo[x] \) represents the \( x \)-th character of the string "namomo" (\( 1 \le x \le 6 \)).

Output the number of namomo subsequences of a given string \( s \) modulo \( 998244353 \).

Input Format

The first line contains a string \( s \) with \( n \) characters (\( 6 \le n \le 1000000 \)). \( s \) contains only lower case English letters (a -- z), upper case English letters (A -- Z), and digits (0 -- 9).

Output Format

Output one integer -- the answer modulo \( 998244353 \).

Sample Input and Output

Sample Input #1

wohaha

Sample Output #1

1

Sample Input #2

momomo

Sample Output #2

0

Sample Input #3

gshfd1jkhaRaadfglkjerVcvuy0gf

Sample Output #3

73

Sample Input #4

retiredMiFaFa0v0

Sample Output #4

33

见题面 Input 部分

见题面 Output 部分