After deleting lines from a file i need to keep the file order

[blackice]

Hello All,

i have a Perl Script that deleting Zones from named.conf file and here is the script

#!/usr/bin/perl -w 
 
use strict;
 
print "please enter the domain name: ";
 
chomp (my $targetdomain = <STDIN>);
 
my $file = "/home/adam/Desktop/hello"; # copied to my home folder
 
rename $file, "$file.bak" or die "Can't rename file '$file': $!\n";
 
open my $in, '<', "$file.bak" or die "Can't read file '$file': $!\n";
open my $out, '>', $file or die "Can't write file '$file': $!\n";
 
my $comment = 0;
my $block   = 0;
 
while(<$in>) {
    if (/^zone\s+"$targetdomain"/) {
        $comment++;
        $block += () = /(\{)/g;    
    next;
    }
    if($comment) {
        $block += () = /(\{)/g;
        $block -= () = /(\})/g;
        unless ($block) {
            $comment = 0;
            next;
        }
    }
 
      print $out $_ unless $comment;
 
}

the Script simply taking the name of the zone as an input then it deletes the whole zone like that

>assume this is the sample from the file

zone "foo.com" {
type xxxxx
path xxxxx
info xxxxx
};

zone "blah.com" {
type xxxxx
path xxxxx
info xxxxx
};

zone "bar.com" {
type xxxxx
path xxxxx
info xxxxx
};

suppose we enter the blah.com as the zone that will be deleted the output will be

zone "foo.com" {
type xxxxx
path xxxxx
info xxxxx
};
#2 blank lines instead of one blank line
#
zone "bar.com" {
type xxxxx
path xxxxx
info xxxxx
};

i want to keep my file order as only one line between any 2 zones .

i have tried many things : like providing a line in my code to replacing the empty line that created instead of deleted zone like
s/\n//;
but that didn't work , also i tried to set the values like $/="" to eliminate the blank line that created So Any help will be a highly appreciated and Thanks for all .

[KevinADC]

See how this works. Uses perls inplace editor to edit the file. Will save a backup copy too.

#!/usr/bin/perl -w 
use strict;
print "please enter the domain name: ";
chomp (my $targetdomain = <STDIN>);
my $file = "/home/adam/Desktop/hello"; # copied to my home folder
 
{
    local @ARGV = ($file);
    local $^I = '.bak';
    MAINLOOP: while(<>) {
        if (/^zone\s+"$targetdomain"/) {
            while(<>) {
               next MAINLOOP if (/^\s*$/);
            }
        }
        else {
            print;
        }
    }
}

[blackice]

Many Thanks it works fine but kindly can you show me how that works ?
i mean can u illustrate the code meaning for me ?

i donot know what is the MAINLOOP and also teh $^I and i will be thankful

[KevinADC]

Many Thanks it works fine but kindly can you show me how that works ?
i mean can u illustrate the code meaning for me ?

i donot know what is the MAINLOOP and also teh $^I and i will be thankful

You can find $^I and many other perl variables on this page:

http://perldoc.perl.org/perlvar.html

MAINLOOP is a "label". It just tells perl where to go in the code otherwise the inner most enclosing loop is always assumed. Example:

OUTTER: for (1..10) {
    print $_,"\n";
    INNER: for (20..30) {
      print $_,"\n";
      next OUTTER if ($_ == 25);
   }
} 

Everytime the "INNER" loop gets to 25 it jumps back to the "OUTTER" loop. If you did not use labels, the inner loop would just repeat and continue printing up to 30.

[blackice]

so thanks for your wonderful answers so if the file was messy or something i mean if before the script running the file looks like this

zone "blah.com" {
xxxx
xxxx
xxxx
}


zone "bar.com" {
xxxx
xxxx
xxxxx
xxxxx
}

notice that the file has already 2 blank lines . so how i can make the script to edit to make the file always has only one space between each zone

i have used local $/ = "\n"; as default file separator but it didnot work and thanks again