Using sed regex to remove the top level domain from a fully qualified domain name

linux2man · 11-30-2011, 07:59 PM

Hi,
I have file contain 100 FQDN I want to remove TLD ".com, .org , .net ...." using sed regex not others, and redirect output to new file.

Thanks

kbp · 11-30-2011, 08:18 PM

Code:

echo $line | sed -e 's/\.[a-z]*$//'

.. might need a tidy up but it should get you started

Telengard · 11-30-2011, 11:53 PM

Code:

test$ cat file
www.yahoo.com
www.msn.com
www.youtube.com
www.linuxquestions.org
www.google.com
www.kubuntuforums.net
www.amazon.com
test$ sed 's/\.[^.][^.]*$//' file
www.yahoo
www.msn
www.youtube
www.linuxquestions
www.google
www.kubuntuforums
www.amazon
test$ # TLD's can include characters outside [a-z]
test$ # http://en.wikipedia.org/wiki/Top_Level_Domain

linux2man · 12-01-2011, 09:37 AM

Thanks a lot it's working like sharm
Can you please describe the command

Telengard · 12-01-2011, 11:39 AM

Quote:

Originally Posted by linux2man

Can you please describe the command

Code:

sed 's/\.[^.][^.]*$//' file

general format of command:
sed 'program' input-file

program enclosed in single quotes to prevent expansion by shell.

format of program:
s/regular-expression/replacement-string/

s = substitute command
\.[^.][^.]*$ = regular expression
replacement string is empty, so matched text is removed from pattern space

more information:

Code:

man sed

if using GNU sed then also see:

Code:

info sed

or http://www.gnu.org/software/sed/manu...ode/index.html

HTH

kbp · 12-01-2011, 03:27 PM

Code:

TLD's can include characters outside [a-z]

.. thats what I meant by cleanup

Telengard · 12-01-2011, 03:38 PM

Quote:

Originally Posted by kbp

.. thats what I meant by cleanup

TBH I was surprised too when skimming the article.