How drop specific columns¶

If you want to remove or keep specific columsn, you can cut command.

Let starts a new session and assigns it a name:

cat <<EOF > test.txt
c1 c2 c3 c4 
1 2 3 4
1 2 3 4
1 2 3 4
EOF

For an simple case, let use cut to just keep the columns 2 and drop the rest from the text file, you can use cut -d ' ' -f2 test.txt, here -d ' ' specifies the field delimiter\space character instead of the tab character, which heresis . In the -f flag you can specifies the specific list of columns that are separated by .

$ cut -d ' ' -f2 test.txt
c2
2
2
2

$ cut -d ' ' -f2,4 test.txt
c2 c4
2 4
2 4
2 4

$ cut -d ' ' -f 2-4 test.txt
c2 c3 c4
2 3 4
2 3 4
2 3 4

Now let revert that action and remove the second columns; use the previous command and add the --complement flag

$ cut -d ' ' --complement -f2 test.txt
c1 c3 c4 
1 3 4
1 3 4
1 3 4

$ cut -d ' ' --complement -f2,4 test.txt
c1 c3 
1 3
1 3
1 3

You can use -c, to selects only the characters specified. Below cut command prints characters after 3^rd.

$ cut -c  3-  test.txt
 c2 c3 c4 
2 3 4
2 3 4
2 3 4

Below cut command prints characters btween 3^rd and 6^th.

cut -c  3-6  test.txt
 c2 
2 3 
2 3 
2 3