Shell Programming and Scripting

maybe this helps you:

Code:

[root@sandbox tmp]# awk '{print $NF}' test
CDR3
CDR3
CDR3
CDR3

$NF = last position

This input file is worse than any of you seem to have imagined. This file uses <space> as the field separator and uses <space> as the contents of an "empty" field. Since the field separator is also field contents in some cases, there is no obvious way to specify an ERE that matches just the field separators. The following awk script converts the spaces that are data to the string "NULL" and leaves the spaces that are separators alone. It seem to do what was requested. (Note that the 2nd printf statement just shows how the input is modified along with the desired output. Remove it to just get the desired output.)

Code:

#!/bin/ksh
awk ' { gsub("  ", " NULL")
        printf("%s %s\n", $1, $4)
        printf("%s %s from %s\n", $1, $4, $0)
}' input