Sort data based on the values ​​of the 1st and 2nd columns and display only the uniq values ​​of the 1st and 2nd columns

Question

I have an input file with the following data:

AUD 100 20
USD 200 30
INR 10  20
AUD 100 30
USD 200 40
EUR 50  60
EUR 50  70
AUD 100 11
USD 200 55
INR 10  35

I run:

cat filename | sort -u -k1,2

He gives me a conclusion as follows:

AUD 100 20
EUR 50  60
INR 10  20
USD 200 30

So, I understand that it sort -u k1,2 sorts the data and gives the result based on a unique combination of values ​​in columns # 1 and # 2. Now, based on this understanding, I run:

  cat filename |  sort -u -k1 

to get the result based on the unique values ​​in column # 1. Therefore, I expected this conclusion:

AUD 100 20
EUR 50  60
INR 10  20
USD 200 30

But the actual result was:

AUD 100 11
AUD 100 20
AUD 100 30
EUR 50  60
EUR 50  70
INR 10  20
INR 10  35
USD 200 55
USD 200 30
USD 200 40

Can someone explain what happens to the option sort -u -km,n?

Answer 1

-k <start>,<end>. <end>, . :

sort -u -k 1 filename

. ( ), :

sort -u -k 1,1 filename

P.S. cat sort...