Phoenix has a string $s$ consisting of lowercase Latin letters. He wants to distribute all the letters of his string into $k$ non-empty strings $a_1, a_2, \dots, a_k$ such that every letter of $s$ goes to exactly one of the strings $a_i$. The strings $a_i$ do not need to be substrings of $s$. Phoenix can distribute letters of $s$ and rearrange the letters within each string $a_i$ however he wants.

For example, if $s = $ baba and $k=2$, Phoenix may distribute the letters of his string in many ways, such as:

• ba and ba
• a and abb
• ab and ab
• aa and bb

But these ways are invalid:

• baa and ba
• b and ba
• baba and empty string ($a_i$ should be non-empty)

Phoenix wants to distribute the letters of his string $s$ into $k$ strings $a_1, a_2, \dots, a_k$ to minimize the lexicographically maximum string among them, i. e. minimize $max(a_1, a_2, \dots, a_k)$. Help him find the optimal distribution and print the minimal possible value of $max(a_1, a_2, \dots, a_k)$.

String $x$ is lexicographically less than string $y$ if either $x$ is a prefix of $y$ and $x \ne y$, or there exists an index $i$ ($1 \le i \le min(|x|, |y|))$ such that $x_i$ < $y_i$ and for every $j$ $(1 \le j < i)$ $x_j = y_j$. Here $|x|$ denotes the length of the string $x$.

The input consists of multiple test cases. The first line contains an integer $t$ ($1 \le t \le 1000$) — the number of test cases. Each test case consists of two lines.

The first line of each test case consists of two integers $n$ and $k$ ($1 \le k \le n \le 10^5$) — the length of string $s$ and the number of non-empty strings, into which Phoenix wants to distribute letters of $s$, respectively.

The second line of each test case contains a string $s$ of length $n$ consisting only of lowercase Latin letters.

It is guaranteed that the sum of $n$ over all test cases is $\le 10^5$.

Print $t$ answers — one per test case. The $i$-th answer should be the minimal possible value of $max(a_1, a_2, \dots, a_k)$ in the $i$-th test case.