UOFTCA - A Research Project

no tags 

The school year has just begun, so it's time for Alice to find a suitable boyfriend! Naturally, this process will first require some careful research using a convenient online academic source known as Facebook.

Alice is considering $G$ ($1 \leq G \leq 100$) guys, and wants to estimate how well-matched she would be with each of them - in other words, how attractive each of them is. For each guy, Alice can find $N$ ($1 \leq N \leq 100$) pictures of him on Facebook, the $i$th of which has attractiveness $A_i$ ($1 \leq A_i \leq 100$). The guy might be as ugly as his least-attractive picture (the one with the smallest attractiveness value), or as hot as his most-attractive picture.

In making her important and complex decision, Alice would like to know the potential range of attractiveness of each of the $G$ potential guys!

Input

Line 1: 1 integer, $G$

For each guy:

Line 1: 1 integer, $N$

Line 2: $N$ integers, $A_{1..N}$

Output

For each guy:

2 integers, the guy's worst-case and best-case attractiveness, respectively

Example

Input:
3
4
2 5 1 3
1
98
5
16 11 11 14 21
Output:
1 5
98 98
11 21
Explanation of Sample:

The first guy's worst picture (his third) has attractiveness 1, while his best (his second) has attractiveness 5.

The second guy has only one picture, making his attractiveness definitely 98.

Finally, the third guy's worst-case attractiveness is 11 (with two of his pictures having this value), while his best is 21.



Added by:SourSpinach
Date:2014-02-18
Time limit:1s
Source limit:50000B
Memory limit:1536MB
Cluster: Cube (Intel G860)
Languages:All except: ASM64
Resource:Own problem, used in the 2013 UofT ACM Tryouts