Lexicographically Smallest String by Removing One Character

Given a string \(s\) of length \(N\), remove exactly one character from it such that the resulting string is lexicographically smallest among all possible strings obtained by removing one character.

For example, if \(s = \texttt{abc}\), then by removing the character \(c\) you obtain \(\texttt{ab}\), and by removing \(b\) you obtain \(\texttt{ac}\). Among \(\texttt{ab}\) and \(\texttt{ac}\), \(\texttt{ab}\) is lexicographically smaller.

Your task is to process multiple test cases. For each test case, output the lexicographically smallest string obtained by removing exactly one character.

Note: In lexicographic order, the string \(x\) is considered smaller than \(y\) if in the first position where they differ, the character in \(x\) comes before the character in \(y\) in alphabetical order.

inputFormat

The first line contains an integer \(T\) representing the number of test cases. Each test case consists of a line containing an integer \(N\) (the length of the string) followed by a string \(s\) of length \(N\). It is guaranteed that \(N \ge 2\).

outputFormat

For each test case, output the lexicographically smallest string obtained after removing exactly one character. Each answer should be printed on a new line.

## sample

2
3 abc
5 abxyz

ab
abxy

</p>