sed command only to exectue on first instance - sed

Question

I am searching for a particular string, and appending a series of lines following it. The sed command i have now is:
sed -i "
/CLIENTVERSION/ {
n
a\define service{
a\ use generic-service
a\ host_name $var_hostname
a\ service_description NSCLient++ Version
a\ check_command check_nt!CLIENTVERSION
a\ }
}" windows.cfg;
The windows.cfg file contains service definitions for specific hosts. ( SBS and Test1 hosts are already in the file, and Test2 is the output after running my command. My output is:
define service{
use generic-service
host_name sbs
service_description NSClient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test2
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test2
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
And I want:
define service{
use generic-service
host_name sbs
service_description NSClient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test2
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
I thought that the /g option did this, but i haven't instituted it and am uncertain as to why it is adding "Test2" service definition twice.

Answer 1

You might find awk easier to work with. Using GNU awk for gensub() and multi-char RS:
$ cat tst.sh
#!/bin/env bash
infile="$1"
var_hostname="Test2"
awk -v RS='^$' -v ORS= '
NR==FNR { rec=$0; next }
{ print gensub(/CLIENTVERSION\n[^\n]+\n/,"&"rec,1) }
' - "$infile" <<!
define service{
use generic-service
host_name $var_hostname
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
!
.
$ ./tst.sh file
define service{
use generic-service
host_name sbs
service_description NSClient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test2
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}
Just change awk to awk -i inplace for inplace editing (like `sed -i) if you like. The above was run on this input file:
$ cat file
define service{
use generic-service
host_name sbs
service_description NSClient++ Version
check_command check_nt!CLIENTVERSION
}
define service{
use generic-service
host_name Test
service_description NSCLient++ Version
check_command check_nt!CLIENTVERSION
}

Answer 2

The approach borrowed from this answer works here:
sed -i "
/CLIENTVERSION/ {
n
a\define service{
a\ use generic-service
a\ host_name $var_hostname
a\ service_description NSCLient++ Version
a\ }
:a;n;ba}" windows.cfg
The difference is in the last line of the script, which puts it into a loop that silently reads the rest of the file without any processing.