#!/usr/bin/perl -w =head1 NAME gnump3d-index - Create a database of all tag information for audio files. =head1 SYNOPSIS gnump3d-index [help options] [debugging options] [global options] Help Options: --manual Show the manual. --help Show this help. --version Show the version number. Debugging Options: --verbose Display all the tag values read. --debug Show the files indexed. Global Options: --config file Set the GNUMP3d config file. --output file Write the output to the given file. --root directory Start the indexing at the given directory. --stats Don't update the cache file, just display audio stats. =cut =head1 DESCRIPTION gnump3d-index is a simple script to index the tag information located within the audio files of your archive. The script will index the files found beneath your root, as defined in the gnump3d.conf file - and write out a cache for speedy access. =cut =head1 OPTIONS =over 8 =item B<--config> Set the GNUMP3d configuration file to read. =item B<--debug> Show the filenames of each file which is indexed. =item B<--help> Show a brief help overview. =item B<--manual> Show the manual to the user. =item B<--output> Write the tag cache to the given fil. =item B<--root> Start the indexing at the given directory, this overrided anything given in the configuration file passed to the program. =item B<--stats> Don't update the cache file, just display audio stats and exit. =item B<--verbose> Display all the tag values read from each indexed track. =item B<--version> Show the version number of this script, and exit.. =item B<--dir> Specify the output directory where images were previously saved. =back =cut =head1 AUTHOR Steve -- http://www.steve.org.uk/ $Id: gnump3d-index,v 1.27 2006/04/25 16:50:56 skx Exp $ =cut =head1 CONTRIBUTORS Contributors to this code: =over 8 =item Gordon Haverland 2003/11/11 =back =head1 LICENSE Copyright (c) 2005 by Steve Kemp. All rights reserved. This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The LICENSE file contains the full text of the license. =cut use strict; # # Standard Perl modules. # use File::Find; use Getopt::Long; use Pod::Usage; # # GNUMPd modules. # use gnump3d::config; use gnump3d::files; use gnump3d::filetypes; use gnump3d::readtags; # Version identifier for this script. my $VERSION_NUMBER = '$Revision: 1.27 $'; # # Options set by the command line arguments. # my %CONFIG; # # Global file-type tester # my $tester = gnump3d::filetypes->new(); # # Determine which configuration file to read. # if ( ( $ENV{"HOME"} ) && ( -e $ENV{"HOME"} . "/.gnump3drc" ) ) { $CONFIG{'file'} = $ENV{"HOME"} . "/.gnump3drc"; } elsif ( -e "/etc/gnump3d/gnump3d.conf" ) { $CONFIG{'file'} = "/etc/gnump3d/gnump3d.conf"; } elsif ( -e "gnump3d.conf" ) { # This is mainly here for Windows users. $CONFIG{'file'} = "gnump3d.conf"; } # # Global variables. # my $cache = "" ; my %cache; my @FOUND = ( ); # An array to hold filenames of audio files we locate. # # Initialize ourselves from the configuration file. # readConfig( $CONFIG{'file'} ); # # Parse the command line arguments. # parseArguments(); # # Make sure that we can read a configuration file. # if ( ! -e $CONFIG{'file'} ) { print "The configuration file which I've tried to read doesn't exist:\n"; print "'" . $CONFIG{'file'} . "'\n"; print "Aborting.\n"; exit; } # # Read various options from the configuration file - unless they have been # specified upon the command line. # if ( !$CONFIG{'root'} || length( $CONFIG{'root'} ) ) { $CONFIG{'root'} = getConfig( "root", "/home/mp3" ); } if ( !length( $cache ) ) { $cache = getConfig( "tag_cache", "" ); } # # If we're to show statistics just do that and exit. # if ( $CONFIG{'stats'} ) { showStats( $cache ); exit; } # # Sanity check the code. # if ( ! -e $CONFIG{'root'} ) { print "The server root directory you are trying to index doesn't exist:\n"; print "'" . $CONFIG{'root'} . "'\n"; print "Aborting\n"; exit; } # Read in existing cache here, if present. readCache( $cache ) if( -e $cache ); # And the update file, if present. readCache( "$cache.updates" ) if( -e "$cache.updates" ); # # Do the indexing - following symlinks on non-Windows platforms. # if ( &isWindows() ) { find({ wanted => \&findAudio }, $CONFIG{'root'} ); } else { find({ wanted => \&findAudio, follow => 1 }, $CONFIG{'root'} ); } # # Process the list of found files. # &indexFiles( ); # # Remove the updates file if present. # unlink("$cache.updates") if( -e "$cache.updates" ); # # Finished exit; # # Use the excellent File::Find module to locate all the files beneath # our archive root. # sub findAudio { my ( $file ) = $File::Find::name; if ( $CONFIG{'debug'} ) { print $file . "\n"; } return if ( ! $tester->isAudio( $file ) ); return if ( -z $file ); push @FOUND, $file; } # # Show the number of files we have indexed, their total size, and # the total playlength. # sub showStats { my ( $file ) = (@_); my $COUNT = 0; my $SIZE = 0; my $TIME = "0"; my %DETAILS = ( ); if ( ! -e $file ) { print "Stats could not be displayed as the cache file doesnt exist"; print "The cache file we expect is $file\n"; exit; } open( FILY, "<$file" ); foreach () { chomp; my @NAMES = split( /\t/, $_); my $file = shift(@NAMES); $DETAILS{$file} = \@NAMES; } close( FILY ); # # Process the hash. # Add up song times. # Add up total song size. # foreach my $file ( keys %DETAILS ) { my $details = $DETAILS{$file}; # A new line == A new file in archive. $COUNT ++; foreach my $pair ( @$details ) { if ( ( $pair =~ /([A-Z]+)=(.*)/ ) && ( length( $2 ) ) ) { my $key = $1; my $val = $2; if ( $key eq "SIZE" ) { $SIZE += $val; } if ( $key eq "LENGTH" ) { if ( $val =~ /([0-9]+):([0-9]+):([0-9]+)/ ) { $TIME += $3 + ($2 * 60 ) + ($1 * 60 * 60 ); } elsif( $val =~ /([0-9]+):([0-9]+)/ ) { $TIME += $2 + ($1 * 60 ) } } } } } # # Fudge the size my $sizeTotal = $SIZE; $sizeTotal = $sizeTotal < (1024) ? $sizeTotal . " bytes" : ( $sizeTotal < (1024 ** 2) ? (int (10 * $sizeTotal/1024)/10) . "K" : ( $sizeTotal < (1024 ** 3) ? (int (10 * $sizeTotal/(1024 ** 2) )/10) . "Mb" : ((int (10 * $sizeTotal/(1024 ** 3) )/10) . "Gb"))); # # Fudge the time. # my $foo = ""; $foo .= int($TIME/(24*60*60)) . " days, "; $foo .= ($TIME/(60*60))%24 . " hours, "; $foo .= ($TIME/60)%60 . " mins "; $foo .= $TIME%60 . " seconds"; $TIME = $foo; # # Print the results # print <$cache" ) or $error = 1; if ( $error ) { print "Error opening the cache file '$cache' - $!\n"; exit; } foreach my $file ( sort @FOUND ) { # Update our progress count every ten tracks. if ( ( ( $count % 10 ) == 0 ) && $CONFIG{'verbose'} ) { my $PER = ( ( $total - $count ) / $total ) * 100.0; if ( $PER =~ /([0-9]+)\.([0-9]+)/ ) { $PER = $1; } print "Indexed $count entries of $total - (\%$PER remaining)\n"; } # If we looked for an old cache file, and it existed, we can now # look to see if the mtime in the cache is the same as the mtime # of the file. If they are the same, trust the cache. my $skip = 0; # Is $file a symlink, or a "file"? my @fstat; if( -l $file ) { # $file is a link @fstat = lstat( $file ) } else { @fstat = stat( $file ); } if ($#fstat < 0) { print "ERROR stat()ing $file\n"; } next if ($fstat[7] < 1); # skip file if it's 0 bytes. if ( ( exists( $cache{$file} ) ) && ( exists( $cache{$file}->{"MTIME"} ) ) && ( defined( $fstat[9] ) ) && ( $cache{$file}->{"MTIME"} == $fstat[9] ) ) { $skip = 1; } unless( $skip ) { # Holder for the tags within the files. my %TAGS = &main::getTags($file); # # Show the tags we've found if the user wanted to see # debugging output. # if ( $CONFIG{'debug'} ) { print $file . "\n"; foreach my $k ( keys %TAGS ) { print "\t$k\t" . $TAGS{ $k } . "\n"; } } # # Write the filename and tag details to the cache file, as # tab seperated values. # print OUT "$file"; foreach my $k ( keys %TAGS ) { my $value = $TAGS{ $k }; # Replace tabs in tag values with spaces so that the reading # code doesn't get confused by excessive deliminators. # $value =~ s/\t/ /g; # # Newlines in the tag values will cause us problems. # Strip them. # $value =~ s/\n//g; $value =~ s/\r//g; print OUT "\t" . $k . "=" . $value; } print OUT "\n"; } else { # We have a previously cached entry to write. print OUT "$file"; foreach my $k ( keys %{$cache{$file}} ) { my $value = defined( $cache{$file}->{ $k } ) ? $cache{$file}->{ $k } : ''; print OUT "\t" . $k . "="; if( length( $value ) > 0 ) { $value =~ s/\t/ /g; $value =~ s/\n//g; $value =~ s/\r//g; print OUT $value; } } print OUT "\n"; } # Update our processed count. $count += 1; } # # Close the output file. close( OUT ); } # # Parse the command line options. # sub parseArguments { my $HELP = 0; my $MANUAL = 0; my $VERSION = 0; GetOptions( # # Help options. # "help", \$HELP, "version", \$VERSION, "manual", \$MANUAL, # # Debug options # "debug", \$CONFIG{'debug'}, "verbose", \$CONFIG{'verbose'}, # # General options # "config=s", \$CONFIG{'file'}, "output=s", \$cache, "root=s", \$CONFIG{'root'}, "stats", \$CONFIG{'stats'}, ); pod2usage(1) if $HELP; pod2usage(-verbose => 2 ) if $MANUAL; if ( $VERSION ) { my $REVISION = '$Revision: 1.27 $'; if ( $REVISION =~ /1.([0-9.]+) / ) { $REVISION = $1; } print "gnump3d-index $REVISION\n"; exit; } } # # GH: subroutine to read the cache. # sub readCache { my $fname = shift; # Cache is 1 line per entry as /absolute/pathname\tKEY=VALUE... open( CACHE, "< $fname" ) || die "Can't open cache ($fname) for read. $!\n"; while( my $line = ) { chomp( $line ); my @fields = split( /\t/, $line ); my $pathname = shift( @fields ); if( exists( $cache{$pathname} ) ) { print "Error: duplicate entries for ($pathname)\n"; } my $entry = {}; foreach my $field (@fields) { my($key,$value,@junk) = split(/=/, $field); $key = 'MTIME' if ($key eq 'mtime'); # Use empty string for undef. $value = defined( $value ) ? $value : ''; # print "Junk in field ($field)\n" if( @junk ); $entry->{$key} = $value; } $cache{$pathname} = $entry; } close( CACHE ); }