path: root/bin/hashl

#!/usr/bin/env perl
## Copyright © 2010 by Daniel Friesel <derf@finalrewind.org>
## License: WTFPL <http://sam.zoy.org/wtfpl>
##   0. You just DO WHAT THE FUCK YOU WANT TO.
use strict;
use warnings;
use 5.010;

use App::Hashl;
use Cwd;
use File::Copy;
use File::Find;
use Getopt::Long;
use IO::Handle;
use Time::Progress;

my $add_unignore  = 0;
my $base          = getcwd();
my $rel_paths     = 1;
my $db_file       = '.hashl.db';
my $total         = 0;
my $cur           = 0;
my $show_progress = 1;
my $timer;
my $incoming_dir;
my $read_size;
my ( $find_ref, $find_db_write );

my $hashl;

our $VERSION = '0.2';

STDERR->autoflush(1);

GetOptions(
	'd|database=s'  => \$db_file,
	'f|force'       => \$add_unignore,
	'n|no-progress' => sub { $show_progress = 0 },
	's|read-size=i' => sub { $read_size = $_[1] * 1024 },
	'V|version'     => sub { say "hashl version ${VERSION}"; exit 0 },
) or usage();

if ( substr( $db_file, 0, 1 ) ne q{/} ) {
	$db_file = "${base}/${db_file}";
}

my $action = shift;

sub usage {
	die(<<"EOF");
Usage: $0 [options] <update|list|info|...> [args]
See 'perldoc -F $0' (or 'man hashl' if it is properly installed)
EOF
}

if ( not defined $action ) {
	usage();
}

if ( -r $db_file ) {
	$hashl = App::Hashl->new_from_file($db_file);
}
else {
	$hashl = App::Hashl->new( read_size => $read_size );
}

local $SIG{INT}  = \&quit_save_db;
local $SIG{TERM} = \&quit_save_db;

sub quit_save_db {
	$hashl->save($db_file);
	exit 0;
}

sub get_total {
	my $file = $File::Find::name;

	if ( -f $file and not -l $file and $file ne $db_file ) {
		$total++;
	}

	return;
}

sub drop_deleted {
	for my $file ( $hashl->files() ) {
		if ( not -e $file ) {
			$hashl->delete_file($file);
		}
	}

	return;
}

sub copy_file {
	my ( $file, $to ) = @_;

	my $cp_base = substr( $file, length($base) + 1 );
	if ( $base =~ s{ / [^/]+ $}{}x ) {
		mkdirs( $incoming_dir, $cp_base );
	}

	copy( $file, "${to}/${cp_base}" )
	  or die("Cannot copy ${file} to ${to}/${cp_base}: $!\n");

	return;
}

sub db_find_new {
	my ( $file, $path ) = @_;

	if ( not $hashl->file_in_db($path) ) {
		say "\r\e[2K${file}";
	}

	return;
}

sub db_find_known {
	my ( $file, $path ) = @_;

	if ( $hashl->file_in_db($path) ) {
		say "\r\e[2K${file}";
	}

	return;
}

sub db_info {
	printf(
		"Read size: %d bytes (%s)\n",
		$hashl->read_size(), $hashl->si_size( $hashl->read_size ),
	);

	return;
}

sub file_info {
	my ($file) = @_;

	my $entry = $hashl->file($file);

	if ( not $entry ) {
		die("No such file in database\n");
	}

	printf(
		"File: %s\nSize: %d bytes (%s)\nHash: %s\n",
		$file, $entry->{size}, $hashl->si_size( $entry->{size} ),
		$entry->{hash},
	);

	return;
}

sub process_file {
	my ( $code, $write ) = ( $find_ref, $find_db_write );

	my $file = $File::Find::name;
	my $path = $file;

	if ( not -f $file or -l $file or $file eq $db_file ) {
		return;
	}

	if ($rel_paths) {
		$file = substr( $file, length($base) + 1 );
	}

	$cur++;

	if ($show_progress) {
		print STDERR $timer->report(
			"\r\e[2KScanning directory: %p done, %L elapsed, %E remaining",
			$cur, );
	}

	&{$code}( $file, $path );

	if ( $write and ( ( $cur % 5000 ) == 0 ) ) {
		$hashl->save($db_file);
	}

	return;
}

sub db_update {
	my ( $file, $path ) = @_;

	$hashl->add_file(
		file     => $file,
		path     => $path,
		unignore => $add_unignore,
	);

	return;
}

sub db_ignore {
	my ( $file, $path ) = @_;

	$hashl->ignore( $file, $path );

	return;
}

sub db_copy {
	my ( $file, $path ) = @_;

	if ( not $hashl->file_in_db($path) ) {
		copy_file( $path, $incoming_dir );
	}

	return;
}

sub mkdirs {
	my ( $dir_base, $new ) = @_;

	for my $dir ( split( qr{/}, $new ) ) {
		$dir_base .= "/$dir";
		if ( !-d $dir_base ) {
			mkdir($dir_base)
			  or die("Cannot create ${dir_base}: $!\n");
		}
	}

	return;
}

sub prepare_db_run {
	my ($dir) = @_;

	$dir //= $base;

	if ( not $show_progress ) {
		return;
	}

	find( \&get_total, $dir );

	$timer = Time::Progress->new();
	$timer->attr(
		min => 1,
		max => $total,
	);

	return;
}

sub cmd_copy {
	($incoming_dir) = @_;

	prepare_db_run();

	if ( not $incoming_dir ) {
		usage();
	}

	if ( substr( $incoming_dir, 0, 1 ) ne q{/} ) {
		$incoming_dir = $base . q{/} . $incoming_dir;
	}

	$find_ref      = \&db_copy;
	$find_db_write = 0;
	find( \&process_file, $base );
	print "\n";

	return;
}

sub cmd_find_known {
	my ($dir) = @_;

	$dir //= $base;

	if ( substr( $dir, 0, 1 ) ne q{/} ) {
		$dir = $base . q{/} . $dir;
	}

	prepare_db_run($dir);

	$find_ref      = \&db_find_known;
	$find_db_write = 0;
	find( \&process_file, $dir );
	print "\n";

	return;
}

sub cmd_find_new {
	my ($new_dir) = @_;

	$new_dir //= $base;

	if ( substr( $new_dir, 0, 1 ) ne q{/} ) {
		$new_dir = $base . q{/} . $new_dir;
	}

	prepare_db_run($new_dir);

	$find_ref      = \&db_find_new;
	$find_db_write = 0;
	find( \&process_file, $new_dir );
	print "\n";

	return;
}

sub cmd_ignore {
	my ($ign_dir) = @_;

	$ign_dir //= $base;

	prepare_db_run();

	if ( substr( $ign_dir, 0, 1 ) ne q{/} ) {
		$ign_dir = $base . q{/} . $ign_dir;
	}

	$find_ref      = \&db_ignore;
	$find_db_write = 1;
	find( \&process_file, $ign_dir );
	$hashl->save($db_file);
	print "\n";

	return;
}

sub cmd_info {
	my ($file) = @_;

	if ($file) {
		file_info($file);
	}
	else {
		db_info();
	}

	return;
}

sub cmd_list {
	my ($re) = @_;

	printf(
		"# hashl v%s   Read Size %d bytes (%s)\n",
		$VERSION, $hashl->read_size(), $hashl->si_size( $hashl->read_size() ),
	);

	for my $name ( sort $hashl->files() ) {
		my $file = $hashl->file($name);

		if ( $re and $name !~ m{$re} ) {
			next;
		}

		printf( "%s %-7s %s\n",
			$file->{hash}, $hashl->si_size( $file->{size} ), $name );
	}

	return;
}

sub cmd_list_files {
	say join( "\n", sort $hashl->files() );

	return;
}

sub cmd_list_ignored {
	for my $hash ( $hashl->ignored() ) {
		say $hash;
	}

	return;
}

sub cmd_update {
	drop_deleted();
	prepare_db_run();

	$find_ref      = \&db_update;
	$find_db_write = 1;

	find( \&process_file, $base );

	print "\n";

	$hashl->save($db_file);

	return;
}

given ($action) {
	when ('copy')         { cmd_copy(@ARGV) }
	when ('find-known')   { cmd_find_known(@ARGV) }
	when ('find-new')     { cmd_find_new(@ARGV) }
	when ('ignore')       { cmd_ignore(@ARGV) }
	when ('info')         { cmd_info(@ARGV) }
	when ('list')         { cmd_list(@ARGV) }
	when ('list-files')   { cmd_list_files(@ARGV) }
	when ('list-ignored') { cmd_list_ignored(@ARGV) }
	when ('update')       { cmd_update(@ARGV) }
	default               { usage() }
}

__END__

=head1 NAME

B<hashl> - Create database with partial file hashes, check if other files are in it

=head1 SYNOPSIS

B<hashl> [B<-fn>] [B<-d> I<dbfile>] [B<-s> I<read-size>] I<action> [I<args>]

=head1 VERSION

This manual documents hashl version 0.2

=head1 DESCRIPTION

Actions:

=over

=item B<copy> I<newdir>

Copy all files in the current directory which are not in the database to
I<newdir>.

=item B<find-known> [I<directory>]

List all files which are already in the database.  Scans either the current
directory or I<directory>.

=item B<find-new> [I<directory>]

List all files which are not in the database.  Scans either the current
directory or I<directory>.

=item B<ignore> [I<directory>]

Add all files in I<directory> (or the current directory) as "ignored" to the
database.  This means that hashl will save the file's hash and skip matching
files for B<copy> or B<find-new>.

=item B<info> [I<file>]

Show information on I<file> (or the database, if I<file> is not specified).

=item B<list> [I<regex>]

List all files and their hashes.  The list format is C<< hash size file >>.

If I<regex> (a perl regular expression) is specifed, only matching files will
be listed.

=item B<list-files>

List all filenames, one file per line.

=item B<list-ignored>

List ignored hashes.

=item B<update>

Update or create hash database.  Iterates over all files below the current
directory.

=back

=head1 OPTIONS

=over

=item B<-d>|B<--database> I<dbfile>

Use I<dbfile> instead of F<.hashl.db>

=item B<-f>|B<--force>

For use with C<< hashl add >>: If there are ignored files in the directory,
unignore and add them.

=item B<-n>|B<--no-progress>

Do not show progress information.  Most useful with C<< hashl find-new >>.

=item B<-s>|B<--read-size> I<kibibytes>

Change size of the part of each file which is hashed.  By default, B<hashl>
hashes the first 4 MiB.  Note that this option only makes sense when using C<<
hashl update >> to create a new database.

=item B<-V>|B<--version>

Print version information.

=back

=head1 EXIT STATUS

Unless an error occured, B<hashl> will always return zero.

=head1 CONFIGURATION

None, so far

=head1 DEPENDENCIES

=over

=item * Digest::SHA

=item * Time::Progress

=back

=head1 BUGS AND LIMITATIONS

Unknown.  This is beta software.

=head1 EXAMPLES

=head2 LEECHING

First, create a database of your local files:

C<< cd /media/videos; hashl update >>

Now, assume you have a (possibly slow) external share mounted at
F</tmp/mnt/ext>.  You do not want to copy all files to your disk and then use
B<fdupes> or similar to weed out the duplicates.  Since you just used B<hashl>
to create a database with the hashes of the first 4MB of all your files, you
can now use it to check if you (very probably) already have any remote file.
For that, you only need to leech the first 4MB of every file on the share, and
not the whole file.  For example:

C<< cd /tmp/mnt/ext; hashl copy /media/videos/incoming >>

=head2 EXTERNAL HARD DISK

Personally, I have all my videos on an external hard disk, which I usually do
not carry with me.  So, when I get new videos, I put them into F<~/lib/videos>
on my netboo, and then later copy them to the external disk.  Of course, it
can always happen that I get a movie I already have, or forget to move
something from F<~/lib/videos> to the external disk, especially since I also
always have some stuff from the disk in F<~/lib/videos>.

However, I can use B<hashl> to conveniently solve this issue.  Run
periodically:

C<< cd /media/argon; hashl -d ~/lib/video/.argon update >>

Now, I always have a list of files on the external disk with me.  When I get a
new file:

C<< hashl -d ~/lib/video/.argon new-file $file >>

And to find out which files are not on the external disk:

C<< cd ~/lib/video; print -l **/*(.) | hashl -d .argon new-file >>

=head1 AUTHOR

Copyright (C) 2010 by Daniel Friesel E<lt>derf@finalrewind.orgE<gt>

=head1 LICENSE

  0. You just DO WHAT THE FUCK YOU WANT TO.