Making an LDAP filter more readable

bronto has asked for the wisdom of the Perl Monks concerning the following question:

Hello everybody

A colleague of mine asked me if I knew how to automatically indent an LDAP filter to make it more readable. E.g.: he wantend something like this:

(&(&(&(& (mailnickname=*) (| (&(objectCategory=person)(objectClass=use
+r)(!(homeMDB=*))(!(msExchHomeServerName=*)))(&(objectCategory=person)
+(objectClass=user)(|(homeMDB=*)(msExchHomeServerName=*)))(&(objectCat
+egory=person)(objectClass=contact))(objectCategory=group)(objectCateg
+ory=publicFolder)(objectCategory=msExchDynamicDistributionList) )))(o
+bjectCategory=contact)(proxyAddresses=smtp:*example.com)))
[download]

to become something more similar to this

(&
    (&
        (&
            (&
                (mailnickname=*)
                (|
                    (&
                        (objectCategory=person)
                        (objectClass=user)
                        (!
                            (homeMDB=*)
                        )
                        (!
                            (msExchHomeServerName=*)
                        )
                    )
                    (&
                        (objectCategory=person)
                        (objectClass=user)
                        (|
                            (homeMDB=*)
                            (msExchHomeServerName=*)
                        )
                    )
                    (&
                        (objectCategory=person)
                        (objectClass=contact)
                    )
                    (objectCategory=group)
                    (objectCategory=publicFolder)
                    (objectCategory=msExchDynamicDistributionList)
                )
            )
        )
        (objectCategory=contact)
        (proxyAddresses=smtp:*example.com)
    )
)
[download]

I didn't want to spend a long time over this, so after a few attempts with the trial-and-error technique and with the help of Devel::ptkdb I came out with this quick and dirty script based on Text::Balanced

#!/usr/bin/perl

use strict ;
use warnings ;

use Text::Balanced qw(extract_multiple) ;

die "Uso: $0 filtro\n" unless @ARGV ;

my ($begop,$begin,$end) = (qr/\([&|!]\s*/, qr/\(\s*/, qr/\)\s*/) ;

my $filter = shift @ARGV ;
my @blocks = extract_multiple($filter,[$begop]) ;

my $step = 0 ;
foreach my $block (@blocks) {
  if ($block =~ $begop) {
    # Inizia un operatore
    print_chunk($step++,$block,1) ;
  } else {
    # E` un blocco di match, probabilmente sbilanciato
    my @matches = extract_multiple($block,[$begin,$end]) ;

    # Questi sono match
    while (@matches >= 3) {
      my @chunks = splice(@matches,0,3) ;

      # Fai check sui "chunk" e agisci di conseguenza:
      if ($chunks[1] =~ /=/) {
        # E` un match:
        print_chunk($step,join("",@chunks),1) ;
      } else {
        # Sfiga
    while (my $chunk = shift @chunks) {
      if ($chunk =~ $end) {
            print_chunk(--$step,$chunk,1) ;
      } else {
        # Ricarica gli elementi in @matches e riparti
        unshift @matches,$chunk,@chunks ;
        last ;
      }
    }
      }
    }

    # Queste sono parentesi che si chiudono
    drop_parenses(@matches) ;
  }
}

sub print_chunk {
  my ($step,$string,$newline) = @_ ;
  print "    "x$step ;
  print $string ;
  print "\n" if $newline ;
}

sub drop_parenses {
  while (my $parens = shift @_) {
    print_chunk(--$step,$parens,1) ;
  }
}
[download]

I am pretty sure that there are far better ways to do that, and I am interested on how you'd do it. Anyone?

Ciao!
--bronto

In theory, there is no difference between theory and practice. In practice, there is.

Back to Seekers of Perl Wisdom