Linux sort and cut multiple columns

Since the result consists of two individually sorted columns of data, there need to be two calls to sort, each one with a separate set of data.

The answer of nezabudka shows how one may do that in a shell that understands process substitutions:

paste -d : \
    <( cut -d : -f 3 info | sort ) \
    <( cut -d : -f 5 info | sort ) \
    >outfile

The above assumes that you want to save the output to a file called outfile.

In a shell that does not understand process substitutions (you mentioned no specific shell in the question), one may do this in two steps using an intermediate file:

cut -d : -f 3 info | sort > outfile.tmp
cut -d : -f 5 info | sort | paste -d : outfile.tmp - >outfile
rm outfile.tmp

The first pipeline creates the initial column of the sorted names of planets. It redirects the sorted data to the file outfile.tmp. The second pipeline creates the second column in a similar manner to the first and appends it to the first using paste. We then delete the intermediate file.

Using gawk and it's builtin array sorting function, we only need a single pass through the file.

gawk '
  BEGIN { FS = OFS = ":" }
  {
    planet[NR] = $3
    name[NR] = $5
  }
  END {
    asort(planet)
    asort(name)
    for (i=1; i <= NR; i++) print planet[i], name[i]
  }
' info

step1: awk -F ":" '{print $3}' inpfile | sort -k1.1 >out.txt; awk -F ":" '{print $5}' inpfile |sort -k1.1 >out_1.txt

step2: paste -d ":" out.txt out_1.txt

output

Earth:Anna
Jupiter:Javier
Mars:Manuel
Neptune:Raymond
Venus:Sebastian