The cut command  - usage and examples

The cut command can be used for extracting portion of text from a file by selecting columns. Here are some useful examples:

For each line extract characters 5-7:

cut -c 5-7 file.txt

First useful example - (LOCAL=NO) processes are the processes of connections using SQLnet (localhost or remote machines) and are not using MTS (Multi-Threaded server) so if we want to list all the pids of processes of connections using SQLnet:

ps -ef | grep 'LOCAL=NO' | cut -c 10-15 | wc -l

Another example, show the number of files per day in the listed directory:

$ ls -lt  | cut -c 42-47| uniq -c | sort -n
1
1 Aug 15
1 Jan 10
1 Nov 19
1 Oct 26
1 Oct 27
2 Nov  9
7 Jun 27

Explanation: ls -lt returns lines which look like this:

-rw-r--r--    1 Sagiv    UsersGrp       400 Dec 14 19:19 1.txt

The lines are sorted by date, so cut returns only the month and day. The command uniq -c counts the lines of each date, and the sort -n sorts it by number.

Parse CSV file and create file with the second,third,fourth and sixth colomns.

cut -f 3-4,6 -d ',' file.csv

Here -d can be any separator (one character).

For example, list users with “User ID Info”:

cut -f1,5 -d':' /etc/passwd  | sort | grep -P ':\w'
 adm:adm
 avahi:Avahi mDNS/DNS-SD Stack
 bin:bin
 colord:User for colord
 daemon:daemon
 dbus:System message bus
 ftp:FTP User
 games:games
 geoclue:User for geoclue
 gluster:GlusterFS daemons
 halt:halt