The Berland language consists of words having exactly two letters. Moreover, the first letter of a word is different from the second letter. Any combination of two different Berland letters (which, by the way, are the same as the lowercase letters of Latin alphabet) is a correct word in Berland language.

The Berland dictionary contains all words of this language. The words are listed in a way they are usually ordered in dictionaries. Formally, word $a$ comes earlier than word $b$ in the dictionary if one of the following conditions hold:

• the first letter of $a$ is less than the first letter of $b$;
• the first letters of $a$ and $b$ are the same, and the second letter of $a$ is less than the second letter of $b$.

So, the dictionary looks like that:

• Word $1$: ab
• Word $2$: ac
• ...
• Word $25$: az
• Word $26$: ba
• Word $27$: bc
• ...
• Word $649$: zx
• Word $650$: zy

You are given a word $s$ from the Berland language. Your task is to find its index in the dictionary.

The first line contains one integer $t$ ($1 \le t \le 650$) — the number of test cases.

Each test case consists of one line containing $s$ — a string consisting of exactly two different lowercase Latin letters (i. e. a correct word of the Berland language).

For each test case, print one integer — the index of the word $s$ in the dictionary.