sed: Remove All Digits / Input From Input

Author: Vivek Gite Last updated: December 22, 2008 6 comments

Q. How do I remove all digits or number from my text file or input using sed command?

A. You need to use regular expression to remove digits or number from input. For example, ‘s/[0-9]*//g’ will remove all numbers from the input.

Remove all digits

Type the following command:
$ echo 'This is a test 12335 and 669384 535xy4' | sed 's/[0-9]*//g'
Sample output:

This is a test  and  xy

You can remove all digits from a text file:
$ sed 's/[0-9]*//g' input.txt $ sed 's/[0-9]*//g' input.txt > output.txt

🐧 Get the latest tutorials on SysAdmin, Linux/Unix, Open Source/DevOps topics:

RSS feed or Weekly email newsletter
Share to Twitter • Facebook • 6 comments... add one ↓

UP NEXT

Category	List of Unix and Linux commands
File Management	cat
Firewall	CentOS 8 • OpenSUSE • RHEL 8 • Ubuntu 16.04 • Ubuntu 18.04 • Ubuntu 20.04
Network Utilities	dig • host • ip • nmap
OpenVPN	CentOS 7 • CentOS 8 • Debian 10 • Debian 8/9 • Ubuntu 18.04 • Ubuntu 20.04
Package Manager	apk • apt
Processes Management	bg • chroot • cron • disown • fg • jobs • killall • kill • pidof • pstree • pwdx • time
Searching	grep • whereis • which
User Information	groups • id • lastcomm • last • lid/libuser-lid • logname • members • users • whoami • who • w
WireGuard VPN	CentOS 8 • Debian 10 • Firewall • Ubuntu 20.04

ADVERTISEMENTS

6 comments… add one

dimka Aug 19, 2009 @ 9:46
But how can i remove all, but digits?!
Reply Link
🐧 nixCraft Aug 19, 2009 @ 12:04
Try:
echo 'This is a test 12335 and 669384 535xy4' | sed 's/[^0-9]*//g'
Reply Link
dimka Aug 21, 2009 @ 6:04
Oh, thanks! That’s what I was looking for!
Reply Link

np Sep 5, 2012 @ 14:56

Hi!
I’m doing this:

echo 'something=1234
something=12345
andsomething=1234&somethingelse&something=1234
something=2345' | sed 's/something=1234[^0-9]/something=9999/g'

The output is :

something=1234
something=12345
andsomething=9999&somethingelse&something=1234
something=2345

And what I’m trying to get is :

something=9999
something=12345
andsomething=9999&somethingelse&something=9999
something=2345

Why isn’t it working when the pattern is at the end of the line ? What a pain….
Help, anyone ?

Reply Link

wudeng Dec 11, 2013 @ 9:17
you can use \b instead of [^0-9], \b means word boundary in regexp
Reply Link
Vlad May 23, 2016 @ 14:15
sed -i “s/…/…/g” file
replaces the contents of file
Reply Link