#!/usr/bin/perl
#
# Find duplicate files in specified directories using md5sum values to
# identify duplicates.
#
# (C) 2009 S.M.Mahesh

use strict;
use warnings;
use File::Find;
use Digest::MD5;

my $version = 0.1;
my %md5sums;
my $md5 = Digest::MD5->new();

sub Usage() {
    print<<USAGEDOC;

    $0 v$version - FindDuplicate script

    USAGE:
        $0  <DIR1> [DIR2...DIRn]

        where,
            DIR1..DIRn      Specifies the directories to search
        
    EXAMPLE:
        $0  /home/user/downloads /home/user/documents

USAGEDOC

    exit 1;
}

sub wanted {

    return unless -f $File::Find::name; # Return if it is not a plain file
    return if -l $File::Find::name; # Return in case this is a symlink

    if (open(FILE, $File::Find::name) )
    {
        binmode(FILE);
        my $sum = $md5->addfile(*FILE)->hexdigest();
        close(FILE);

        my $aref = $md5sums{$sum};
        if ( defined $aref )
        {
            push @$aref, $File::Find::name;
        }
        else
        {
            my @list = ($File::Find::name);
            $md5sums{$sum} = \@list;
        }
    }
    else
    {
        print "ERROR: Could not open '$File::Find::name' for reading\n";
    }

    return;
}


Usage() if( $#ARGV < 0 );

foreach my $dir (@ARGV) {
    print "$dir \n";

    unless ( -d $dir ) {
        print "ERROR: '$dir' is not a valid directory\n";
        next;
    }

    find(\&wanted, $dir);
}


print "\n", '-'x25, "\n";
print "Printing duplicate files (if any)\n";
print '-'x25, "\n\n";

foreach my $sum (sort keys %md5sums) {
    my $list = $md5sums{$sum};

    if ($#$list > 0)
    {
        print "$sum :\n";
        foreach my $file (@$list) {
            print "\t $file\n";
        }
        print"\n"
    }
}

print '-'x25, "\n";