#!/usr/bin/perl -w
# $Revision: 1.28 $
# $Date: 2006-12-15 18:28:24 $
#
# Luis Mondesi < lemsx1@gmail.com >
#
# DESCRIPTION: A simple script to rename Music files in a consistent manner
# USAGE: cd ~/Music; normalize_music.pl or simply: $0 --help
# LICENSE: GPL
use strict;
$|++;
my $revision='1.28'; # version
# standard Perl modules
use utf8;
use Getopt::Long;
Getopt::Long::Configure('bundling');
use File::Spec::Functions qw/ splitdir catdir catfile / ; # abs2rel() and other dir/filename specific
use File::Find; # find();
use File::Basename; # basename() && dirname()
#use FileHandle; # for progressbar
my $got_ogg = 0;
my $got_mp3 = 0;
eval "use MP3::Tag";
if ( not $@ )
{
$got_mp3 = 1;
}
eval "use Ogg::Vorbis::Header::PurePerl";
if ( not $@ )
{
$got_ogg = 1;
}
die("We need Ogg::Vorbis::Header::PurePerl or MP3::Tag installed
")
if (not $got_ogg and not $got_mp3);
# Globals (no need to change any variables @see $0 --help)
my $MUSIC_FILES = '\.(mp3|ogg)$'; # files we will find
my @TAGS = ('song','track','artist','album');
my @ls=();
# allows for removing dirs
my @ls_dirs=();
my $longest_path = 0;
# Args:
my $PVERSION=0;
my $HELP=0;
my $DEBUG=0;
my $VERBOSE=0;
my $SHOW_DUPS=0;
my $REMOVE_EMPTY_DIRS=0;
my $REMOVE_DUPS=0;
my $FILE=undef;
# get options
GetOptions(
# flags
'v|version' => \$PVERSION,
'h|help' => \$HELP,
'D|debug' => sub { $DEBUG++; $VERBOSE++; $SHOW_DUPS++; },
'V|verbose' => sub { $VERBOSE++; $SHOW_DUPS++; },
'S|show-duplicates' => \$SHOW_DUPS,
'R|remove-empty-dirs' => \$REMOVE_EMPTY_DIRS,
'r|remove-duplicates' => \$REMOVE_DUPS,
# strings
#'o|option=s' => \$NEW_OPTION,
# numbers
#'a|another-option=i' => \$NEW_ANOTHER_OPTION,
) and $FILE = shift;
if ( $HELP ) {
use Pod::Text;
my $parser = Pod::Text->new (sentence => 0, width => 78);
$parser->parse_from_file($0,\*STDOUT);
exit 0;
}
if ( $PVERSION ) { print STDOUT ($revision,"
"); exit 0; }
# main
umask(0022); # fix anal permissions
if ( defined ($FILE) and $FILE !~ /^\s*$/ )
{
die ("No such file $FILE
") if ( not -f $FILE );
my $err = _rename($FILE);
print STDERR ("$err
") if ( $VERBOSE );
# if we were passed more files from the command line, do those as well:
foreach ( @ARGV )
{
next if ( ! -f $_ );
my $err = _rename($_);
print STDERR ("$err
") if ( $VERBOSE );
}
} else {
# if we were passed a directory name, use it, else defaults to current directory
my $_root = ( $FILE !~ /^\s*$/ and -d $FILE ) ? $FILE : ".";
# are we running from Nautilus?
# Get Nautilus current working directory, if under Natilus:
if ( exists $ENV{'NAUTILUS_SCRIPT_CURRENT_URI'}
and $ENV{'NAUTILUS_SCRIPT_CURRENT_URI'} =~ m#^file:///# )
{
$_root = $ENV{'NAUTILUS_SCRIPT_CURRENT_URI'};
$_root =~ s#%([0-9A-Fa-f]{2})#chr(hex($1))#ge; # fixes %20 and other URL thingies
$_root =~ s#^file://##g;
}
chdir($_root) or die ("Could not change to directory $_root. $!
");
my $aryref = do_file_ary(".");
foreach(@$aryref)
{
my $err = _rename($_);
print STDOUT ($err,"
") if ( $DEBUG or $VERBOSE );
}
_remove_empty_dirs() if ($REMOVE_EMPTY_DIRS > 0);
}
# support functions
sub do_file_ary {
# uses find() to recur thru directories
# returns an array of files
# i.e. in directory "a" with the files:
# /a/file.txt
# /a/b/file-b.txt
# /a/b/c/file-c.txt
# /a/b2/c2/file-c2.txt
#
# my $aryref = do_file_ary(".");
#
# will yield:
# a/file.txt
# a/b/file-b.txt
# a/b/c/file-c.txt
# a/b2/c2/file-c2.txt
#
my $ROOT = shift;
my %opt = (wanted => \&process_file, no_chdir=>1);
find(\%opt,$ROOT);
return \@ls;
}
sub process_file {
# remove empty dirs before we rename files (in case there is no files in this)
if ($REMOVE_EMPTY_DIRS > 0 and -d $_)
{
my @_dirs = splitdir($_);
$longest_path = ($#_dirs > $longest_path) ? ($#_dirs+1) : $longest_path;
push(@ls_dirs,$_);
return;
}
push (@ls,$_) if ( $_ =~ m($MUSIC_FILES)i and -f $_ );
}
sub _remove_empty_dirs
{
# removes all empty directories from our list of dirs
# you have to do $longest_path number of passes using rmdir()
# to get them all
for (my $i=0;$i<$longest_path;$i++)
{
for (my $j=0;$j<$#ls_dirs;$j++)
{
warn("removing dir $ls_dirs[$j]
") if ($DEBUG);
rmdir($ls_dirs[$j]) and
splice(@ls_dirs,$j,1); # remove this item from our array
}
}
}
sub _mkdir
{
# @desc implements `mkdir -p`
my $path = shift;
my $root = ( $path =~ m,^([/|\\|:]), ) ? $1 : ""; # relative or full path?
my @dirs = splitdir($path);
my $last = "";
my $flag=1;
foreach (@dirs)
{
next if ( $_ =~ m/^\s*$/ );
$last = ( $flag > 1 ) ? catdir($last,$_) : "$root"."$_" ;
mkdir ($last) if ( ! -d $last);
$flag++;
}
return $flag; # number of directories created
}
sub _rename
{
my $orig_filename=shift;
my ($_name,$_path,$_suffix) = fileparse($orig_filename,('\.mp3','\.ogg'));
my $hashref = undef;
# deal with ogg files
if ( $_suffix eq ".ogg" )
{
my $ogg = Ogg::Vorbis::Header::PurePerl->new($orig_filename);
foreach my $k ($ogg->comment_tags) {
foreach my $v ( $ogg->comment($k) )
{
$hashref->{lc($k)} = $v;
$hashref->{'song'} = $v if ( lc($k) eq 'title' );
}
}
} elsif ( $_suffix eq ".mp3" ) {
# deal with mp3 files
no warnings;
my $mp3 = MP3::Tag->new($orig_filename);
$hashref = $mp3->autoinfo();
} else {
print STDERR ("Unknown fileformat for $orig_filename
");
return "";
}
print STDOUT ("_"x69,"
") if ( $VERBOSE );
print STDOUT ("file\t$orig_filename
") if ( $VERBOSE );
if ( $DEBUG )
{
use Data::Dumper;
print Dumper($hashref);
}
# tracks,artist,album are not that essential:
#'song','track','artist','album'
if ( ! defined($hashref->{'track'}) or $hashref->{'track'} =~ m/^\s*$/ )
{
$hashref->{'track'}="00/00";
}
if ( ! defined($hashref->{'artist'}) or $hashref->{'artist'} =~ m/^\s*$/ )
{
$hashref->{'artist'} = "noartist";
}
if ( ! defined($hashref->{'album'}) or $hashref->{'album'} =~ m/^\s*$/ )
{
$hashref->{'album'} = "noalbum";
}
foreach(@TAGS)
{
return "
*** $_ tag missing for file '$orig_filename'
Bailing out
"
if ( ! defined($hashref->{$_}) or $hashref->{$_} =~ m/^\s*$/ );
# clean chars that might not be good for filenames
#?|?|?|?|?|?|
$hashref->{$_} =~ s/([^[:alnum:]\!\@\*\#\%\(\)\[\]\_\-\:\,\.\'\"\{\}\=\+ ])//gi;
print STDOUT ($_, "\t", $hashref->{$_}, "
") if ( $VERBOSE );
}
my ($track,$garbage) = split(/\//,$hashref->{'track'});
$track =~ s/^(\d{1,2}).*$/$1/g;
my $path = lc( catdir($hashref->{'artist'},$hashref->{'album'}) );
my $new_filename = lc( catfile($path,$track."-".$hashref->{'song'}.$_suffix) );
print STDOUT ("to file\t$new_filename
") if ( $VERBOSE );
# silently bail out if we have done this file before
$orig_filename =~ s,^(\./|\.),,;
if ( $new_filename eq $orig_filename )
{
print STDERR ("Skipping filename: same file ...
") if ($DEBUG);
return "";
}
if ( -f $new_filename )
{
print STDERR ("New filename already exist ...
") if ($DEBUG);
print STDOUT ("DUP:\t$orig_filename\t$new_filename
") if ( $SHOW_DUPS );
unlink($orig_filename) if ( $REMOVE_DUPS );
return "";
}
# make the path where we will be putting our new_filename
_mkdir($path) if ( ! -d $path );
if ( ! rename ( $orig_filename,$new_filename ) )
{
print STDOUT ("Renaming $orig_filename to $new_filename failed. Do you have permissions to write in $path?
");
return "";
}
}
__END__
=head1 NAME
normalize_music.pl - a simple mp3/ogg file renaming script
=head1 SYNOPSIS
B<normalize_music.pl> [-v,--version]
[-D,--debug]
[-h,--help]
[-V,--verbose]
[-S,--show-duplicates]
[-R,--remove-empty-dirs]
[-r,--remove-duplicates]
[[directory] | [file1 [file2] [...]]]
=head1 DESCRIPTION
This script finds all music files in a given directory and renames them according to the tags found in them. Renaming is consistent with iTunes naming convention with minor additions:
Artist/Album/track-song_name-artist.$ext
If a directory name is given as argument, then it will rename music files found in that path, otherwise the current directory is used.
If a file name is given as argument, or more a series of files, only those files will be processed.
=head1 EXAMPLES
cd /path/to/dir && normalize_music.pl
normalize_music.pl /path/to/dir
normalize_music.pl file.ogg file.mp3 ...
=head1 OPTIONS
=over 8
=item -v,--version
prints version and exits
=item -D,--debug
enables debug mode
=item -h,--help
prints this help and exits
=item -V,--verbose
print all tags about each file
=item -S,--show-duplicates
print files which have the same id3 tags but on different locations. The lines will be prefix with "DUP:"
=item -R,--remove-empty-dirs
removes empty directories found in given path
=item -r,--remove-duplicates
removes duplicate files. This option implies --show-duplicates. Backup your files first. Be extra careful with this option! Use --show-duplicates to know what will be deleted.
=back
=head1 AUTHORS
Luis Mondesi <L<lemsx1@gmail.com>>
=cut
Tweets by @13m5
Advertisement