On Fri, Nov 13, 2009 at 5:07 AM, Parag Kalra <paragka...@gmail.com> wrote:
> Hey All,
>
> I have a following a LDIF file to process through Perl -
>
> ###############################################################
> dn: ou=71404558, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=People, ou=71404558, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=Groups, ou=71404558, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=40208723, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=People, ou=40208723, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=Groups, ou=40208723, ou=Company, ou=Personal, o=paragkalra.com
> dn: cn=Mygroup,ou=Groups, ou=40208723, ou=Company, ou=Personal, o=
> paragkalra.com
> dn: ou=63613012, ou=Company, ou=Personal, o=paragkalra.com
> dn: cn=Manager,ou=63613012, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=People, ou=63613012, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=Groups, ou=63613012, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=$%^&?()+-, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=[]...@#%*, ou=Company, ou=Personal, o=paragkalra.com
> dn: cn=CEO,ou=[]...@#%*, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=76884580, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=People, ou=76884580, ou=Company, ou=Personal, o=paragkalra.com
> dn: ou=Groups, ou=76884580, ou=Company, ou=Personal, o=paragkalra.com
> ################################################################
>
> My aim is to extract all those lines starting with the string - 'dn:ou' and
> having only 3 levels of OU heirarchy i.e having the string 'ou=' only 3
> times. Also he value of 'ou' can be anything i.e any character (including
> special characters) except comman (,).
>
> So to accomplish it I have written following code and is working -
>
> #################################################################
> # Author: Parag
>
> my $file = $ARGV[0];
>
> open FILE, $file;
>
> while ( <FILE> ) {
> $count = 0;
> $_ =~ s/(\s)+//g;
> $count++ while $_ =~ /(ou=)/g;
> if ( ($count == 3) && ($_ =~ /^(dn:ou)/) ) {
> print "$_\n";
> }
> }
>
> close FILE;
>
> ################################################################
>
>
>
> But I would like to accomplish it through 1 singe regex like using
> following
> code -
>
>
> ############################################################
> # Author: Parag
>
> my $file = $ARGV[0];
>
> open FILE, $file;
>
> while ( <FILE> ) {
> $_ =~ s/(\s)+//g;
> if ( $_ =~ /(dn:ou=)([\d\D]+)(ou=Company,ou=Personal,o=paragkalra.com
> )/){
> print "$_\n";
> }
> }
>
> close FILE;
>
> ############################################################
> Any pointers?
>
Nope. Just trial and error until you come up with an ugly hack. I think
this works:
use strict;
use warnings;
use 5.010;
open (my $INFILE, '<', shift @ARGV);
while (<$INFILE>) {
print if /^dn: ((ou=[^,]+\s*,)\s*){3}[^,]*$/;
}
close $INFILE;
