Question

I have a very large data (12G) that looks like this:

foo  r1.1   abc
foo  r10.1  pqr
qux  r2.1   lmn
bar  r33.1  xpq

What I want to do is to sort 2nd field numerically yielding (in reality there are more leading fields):

foo  r1.1   abc
qux  r2.1   lmn
foo  r10.1  pqr
bar  r33.1  xpq

I tried the following but won’t work:

   sort -k1 -n

What’s the right way to do it?

Answer 1

How about sort -k1.2n if it starts with just an r