Shell Programming and Scripting

Hello,

How do I use sed to search for a pattern in the middle of a line and delete the whole line.

The command I have is as below.

Code:

sed -i '/soft nofile/ d' /etc/security/limits.conf

I have some *'s and spaces in front of the string. With the above command I'm not able to delete it. Any help is appreciated.

try:

Code:

sed -i '/soft[ \t][ \t]*nofile/ d' /etc/security/limits.conf

Thanks rdrtx1 for your reply. But it didn't work. The issue is sometimes the number of spaces differs like below.

*<>soft<><><> nofile<> 12000 (or) *<><>soft<>nofile<><><><>12000 (or) *soft<><><><> nofile<><> 12000. So i've to ignore the spaces somehow and look for the pattern "soft nofile" and then remove the whole line.

Note : <> refers to a space.

[ \t] refers to space or tab ( [<>\t] as per note)

for single space your solution should work unless the file is not writeable.