Question

Good Day Everyone,

I’m currently trying to write a routine that will examine a file for a particular unique string, then delete the entire encapsulated text as follows:

define service{
    use                     generic-service,srv-pnp
    host_name               MEVIE.COM
    service_description     NSClient++ Version
    check_command           check_nt!CLIENTVERSION
    }

There’s many instances of define service { ... } in the file. So I can’t just search on define service, I have to search on the unique string then delete the entire define service { ... } encapsulation.

For example:

• Find the string “NSClient++ Version” in the file…
• Delete EVERYTHING within define service{ .... }
• Delete define service { .... } tags as well.

I really hope my goal is clear.

I’ve tried: sed -n '/NSClient+++ Version/{H;g;p};H' MEVIE.bak|tail -n5 > mevie.fil .. it returns everything except the last ‘}’

Answer 1

Here’s one way using sed. Run like:

sed -f script.sed file

Contents of script.sed:

:t
/define service{/ , /}/ {
    /}/! {
        $! {
            N;
            bt
        }
    }
    /NSClient++ Version/d;
}

Alternatively, here’s the one-liner:

sed ':t /define service{/ , /}/ { /}/! { $! { N; bt } }; /NSClient++ Version/d }' file