fdupes-pivotal

#!perl

unless($ARGV[1]){
	print "fdupes-pivotal: You need to define a first directory (the pivot) and a second one which will be traversed for duplicates of files from the pivot\n\n";
	exit;
}

use File::Find;
use File::Spec ();
use File::Basename ();
use Digest::MD5::File qw( file_md5_hex );

our $VERSION = 0.02;
our $check_md5 = 1;

my $path_pivot = File::Spec->rel2abs($ARGV[0]);
my $path_dir = File::Spec->rel2abs($ARGV[1]);

die "fdupes-pivotal: $path_pivot does not exist!" unless -e $path_pivot;

my %sizes;
my %mtimes;
my %hashes;
my $cntPivot;
if(-f $path_pivot){
	print "fdupes-pivotal: Using file $path_pivot as pivot:\n";

	my @stat = stat($path_pivot);
	$sizes{$stat[7]} = $path_pivot;

	$mtimes{$stat[9]} = $path_pivot;

	if($check_md5){
		my $md5 = file_md5_hex($path_pivot);
		$hashes{$md5} = $path_pivot;
	}

	$cntPivot++;
}else{
	print "fdupes-pivotal: Building file list from pivot dir: $path_pivot\n";
	opendir(my $dh, $path_pivot) || die "can't opendir $path_pivot: $!";
	 my @files = readdir($dh);
	closedir $dh;
	for(@files){
		next unless -f $path_pivot.'/'.$_;

		my @stat = stat($path_pivot.'/'.$_);
		warn "fdupes-pivotal: This size already exists!!\n" if defined($sizes{$stat[7]});
		$sizes{$stat[7]} = $path_pivot.'/'.$_;

		warn "fdupes-pivotal: This mtime already exists!!\n" if defined($mtimes{$stat[9]});
		$mtimes{$stat[9]} = $path_pivot.'/'.$_;

		if($check_md5){
			my $md5 = file_md5_hex($path_pivot.'/'.$_);
			warn "fdupes-pivotal: This hash already exists!!\n" if defined($hashes{$md5});
			$hashes{$md5} = $path_pivot.'/'.$_;
		}

	#	print "fdupes-pivotal: $path_pivot/$_ size:$stat[7] md5:$md5\n";
		$cntPivot++;
	}
}

my %results;

print "fdupes-pivotal: Traversing dir $path_dir:\n";
my $cnt=0;
my $cntMatches=0;
File::Find::find({ wanted => \&wanted }, $path_dir);
print "fdupes-pivotal: Done. $cntPivot pivot file(s) ; compared $cnt files ; $cntMatches matches.\n";


sub wanted {
	## only dirs with files qualify
	return if !-f $File::Find::name;

	$cnt++;

	my @stat = stat($File::Find::dir.'/'.$_);

	next unless defined($sizes{$stat[7]});

	print " ".File::Basename::basename($sizes{$stat[7]})."\n";
	print " $_ in $File::Find::dir - ";
	print "same name, " if File::Basename::basename($sizes{$stat[7]}) eq $_;
	print "same size";

	if( defined($mtimes{$stat[7]}) ){
		print ", same mtime";
	}else{
		print ", diff mtime";
	}

	if($check_md5){
		my $md5 = file_md5_hex($File::Find::dir.'/'.$_);
		next unless defined($hashes{$md5});
		print ", same md5";
	}

	$cntMatches++;
	print "\n";
}

__END__

=pod

=head1 NAME

fdupes-pivotal - (Perl) fdupes variant, where duplicates are searched from a central (pivotal) directory

=head1 SYNOPSIS

	$ fdupes-pivotal <some directory> <some other directory>

	You need to define a first directory (the pivot) and a second one which will be traversed for
	duplicates of files from the pivot directory.

=head1 DESCRIPTION

You might know the handy file-deduplication tool L<fdupes|http://en.wikipedia.org/wiki/Fdupes>. Well,
fdupes looks for duplicates I<within> a directory tree. Now this tool here, fdupes-pivotal, takes files
located in a defined central (or as we call it: "I<pivotal>") directory and looks for duplicates of these
files in another directory tree by traversing down that other directory tree, comparing sizes, timestamps
and MD5 hashes.

=head1 CAVEATS

This is alpha stage software! Be careful when you use it. Look into the source to understand
what it does and to make sure that it's safe for you to use. And feel free to send patches
or improvements by forking the public repository.

One current limitation is that it assumes that all sizes and mtimes in the pivot dir are unique.

=head1 AUTHOR

Clipland GmbH L<http://www.clipland.com/>

=head1 COPYRIGHT & LICENSE

Copyright 2013 Clipland GmbH. All rights reserved.

This library is free software, dual-licensed under L<GPLv3|http://www.gnu.org/licenses/gpl>/L<AL2|http://opensource.org/licenses/Artistic-2.0>.
You can redistribute it and/or modify it under the same terms as Perl itself.