Question

2

sed replace text containing new line

rated 0 times [ 2] [ 0] / answers: 1 / hits: 21368 / 2 Years ago, mon, april 4, 2022, 10:47:27

I have a bash script for installing a web server on my Ubuntu server.

I'd like, in the /etc/apache2/apache2.conf, to replace

#ServerRoot "/etc/apache2"



#

by

#ServerRoot "/etc/apache2"

ServerName localhost



#

If I do

sed -i 's|#ServerRoot "/etc/apache2"|#ServerRoot "/etc/apache2"

ServerName localhost|' /etc/apache2/apache2.conf

it's working, but I'd like to include the blank line and the # after the #ServerRoot to avoid adding twice ServerName if the script is executed twice.

I tried

sed -i 's|#ServerRoot "/etc/apache2"



#|#ServerRoot "/etc/apache2"

ServerName localhost



#|' /etc/apache2/apache2.conf

But it's not working.

Answers

Only authorized users can answer the question. Please sign in first, or register a free account.

motivengry

Add To Favorites

Follow

Total Points: 459

Total Questions: 112

Total Answers: 108

Location: Bonaire

Member since Sat, May 1, 2021

3 Years ago

answered 2 Years ago biryanrownies · Accepted Answer

How about

search for #ServerRoot "/etc/apache2"

if found, read and append the next line into sed's pattern space

search the appended pattern space for an empty line after the newline, and substitute ServerName localhost for that

Putting that together,

sed -i '|#ServerRoot "/etc/apache2"| {N;s|

$|

ServerName localhost|}' /etc/apache2/apache2.conf

In case the 'empty' line actually contains whitespace, you could modify that to

sed -i '|#ServerRoot "/etc/apache2"| {N;s|

s*$|

ServerName localhost|}' /etc/apache2/apache2.conf