Dilu is a young boy who loves mathematics. He recently learned how to code. He was interested in solving different mathematical problems, but was irritated with the fact of double number precision.
He wanted to write a high precision arithmetic library for himself which will have precision up to 100 digits after decimal point. Since Dilu is new in programming, he is facing many problems doing it.
So, he asked for your help. He said only division operation would be okay for now. You'll be given two integers A and B by Dilu. You have to print A/B up to 100 digits after decimal point. If the output exceeds 100 placed after decimal point, you need to round the quotient.
Input begins with a integer T (0 ≤ T ≤ 1000) denoting the number of test cases to be followed. Each of next T line contains two integers A and B (0 ≤ A,B ≤ 10^7 and B ≠ 0).
For each case print the quotient of A/B up to 100. If possible exclude trailing zeros.
3 2 4 5 5 0 3
0.5 1 0