Search code examples
regexperl

Regex Filename Not Matching


I'm running a Perl script that someone else wrote for me, and I've noticed that the following regex doesn't seem to match my files which are named "Film CD 1.avi", "Film2 CD 1.avi", etc.:

$name = (split (/[ ]*CD[ ]*1/i, $in_avi))[0];

Does anyone know what I need to do to correct it?

The full code is this:

#!/usr/bin/perl

use strict;

########################################################################
#                                                Configuration Details                                          #
########################################################################

# Array of movies to convert
my @MOVIES_TO_CONVERT; # = ('The Figher', 'The Fifth Element');

# Directory where the split AVI files are stored
my $MOVIE_IN_PATH = 'D:\split';

# Directory where the combined AVI files will be stored
my $MOVIE_OUT_PATH = 'D:\combined';

# Full path to the avidemux executable
my $AVIDEMUX = 'C:\Program Files (x86)\Avidemux\avidemux.exe';

########################################################################
#                                                          Functions                                                      #
########################################################################

# Return an array of all the AVI files in the specified directory.
sub get_avis_in_directory
{
        my $dh; # Directory handle
        my $dir; # Current directory
        my @avis; # Array of file names to return

        opendir ($dh, $dir = shift) or die "Failed to open directory $dir: $!\n";
        while (readdir $dh)
        {
                next if (/^\.{1,2}/);
                $_ = $dir . "\\" . $_;
                push (@avis, $_) if (-f $_ and /.*\.avi$/i);
        }
        closedir $dh;

        return (@avis);
}

########################################################################
#                                                         Entry Point                                                    #
########################################################################

die "Input directory $MOVIE_IN_PATH does not exist!\n" unless (-d $MOVIE_IN_PATH);
die "Output directory $MOVIE_OUT_PATH does not exist!\n" unless (-d $MOVIE_OUT_PATH);

# This variable represents the actual names and paths of movies to be converted.
# It will either be built from the files specified in @MOVIES_TO_CONVERT manually, or
# built dynamically based on the files in the source and destination paths.
my @movies_formatted; # Array of hashes of movies to convert

if ($#MOVIES_TO_CONVERT == -1)
{
        my @in_avis; # Array of AVI files in the input directory
        my @out_avis; # Array of AVI files in the ouput directory

        @in_avis = get_avis_in_directory ($MOVIE_IN_PATH);
        @out_avis = get_avis_in_directory ($MOVIE_OUT_PATH);

        for my $in_avi (@in_avis)
        {
                if ($in_avi =~ /.*[ ]*CD[ ]*1\.avi$/i)
                {
                        my $rec; # Temporary hash variable
                        my $name; # Name of the move we are processing
                        $name = (split (/[ ]*CD[ ]*1/i, $in_avi))[0];
                        $name = (split (/$MOVIE_IN_PATH[\\\/]{1}/i, $name))[1];

                        for my $in_avi_2 (@in_avis)
                        {
                                if ($in_avi_2 =~ /^$MOVIE_IN_PATH\\$name[ ]*CD[ ]*2\.avi$/i)
                                {
                                        $rec->{'part2'} = $in_avi_2;
                                        last;
                                }
                        }

                        if (defined $rec->{'part2'})
                        {
                                for my $out_avi (@out_avis)
                                {
                                        if ($out_avi =~ /$name\.avi$/i)
                                        {
                                                $rec->{'output'} = $out_avi;
                                                last;
                                        }
                                }

                                unless (defined $rec->{'output'})
                                {
                                        $rec->{'part1'} = $in_avi;
                                        $rec->{'output'} = "$MOVIE_OUT_PATH\\$name.avi";
                                        push (@movies_formatted, $rec);
                                }
                        }
                }
        }
}
else
{
        my $rec; # Temporary hash variable

        for my $name (@MOVIES_TO_CONVERT)
        {
                $rec = {};
                $rec->{'part1'} = "$MOVIE_IN_PATH\\$name CD 1.avi";
                $rec->{'part2'} = "$MOVIE_IN_PATH\\$name CD 2.avi";
                $rec->{'output'} = "$MOVIE_OUT_PATH\\$name.avi";
                push (@movies_formatted, $rec);
        }
}

for my $movie (@movies_formatted)
{
        my $convert_cmd = "\"$AVIDEMUX\" --load \"" . $movie->{'part1'} . "\" --append \"" . $movie->{'part2'} . "\" --force-smart --save \"" . $movie->{'output'} . "\" --quit";
        print "$convert_cmd\n";
        die "Unable to convert $movie->{'output'}!\n" if (system "$convert_cmd");
}

The script is from http://www.neowin.net/forum/topic/1128384-batch-file-to-run-command-on-joining-video-files by xorangekiller.


Solution

  • The backslash in $MOVIE_IN_PATH is being removed by interpolating it into the regular expression. Change both the split commands to

    $name = (split(/\Q$MOVIE_IN_PATH\E[\\\/]{1}/i, $name))[1];
    

    i.e. add the \Q ... \E to escape any regex metacharacters.

    Your friend doesn't appear to be a particularly good Perl programmer. You may want to try this code instead.

    #!/usr/bin/perl
    use strict;
    use warnings;
    
    use File::Basename;
    use File::Spec;
    
    # Configuration Details
    #######################
    
    # Array of movies to convert
    my @MOVIES_TO_CONVERT;    # = ('The Fighter', 'The Fifth Element');
    
    # Directory where the split AVI files are stored
    my $MOVIE_IN_PATH = 'D:\split';
    
    # Directory where the combined AVI files will be stored
    my $MOVIE_OUT_PATH = 'D:\combined';
    
    # Full path to the avidemux executable
    my $AVIDEMUX = 'C:\Program Files (x86)\Avidemux\avidemux.exe';
    
    sub get_avis_in_directory {
      opendir my $dh, $MOVIE_IN_PATH or die $!;
      return grep {
        /\.avi$/ and -f File::Spec->catfile($MOVIE_IN_PATH, $_)
      } readdir $dh;
    }
    
    -d $MOVIE_IN_PATH or die qq(Input directory "$MOVIE_IN_PATH" does not exist!\n);
    -d $MOVIE_OUT_PATH or die qq(Output directory "$MOVIE_OUT_PATH" does not exist!\n);
    
    my @movies_formatted;    # Array of hashes of movies to convert
    
    if (@MOVIES_TO_CONVERT) {
    
      for my $name (@MOVIES_TO_CONVERT) {
        my $rec = {};
        $rec->{part1}  = File::Spec->catfile($MOVIE_IN_PATH, "$name CD 1.avi");
        $rec->{part2}  = File::Spec->catfile($MOVIE_IN_PATH, "$name CD 2.avi");
        $rec->{output} = File::Spec->catfile($MOVIE_OUT_PATH, "$name.avi");
        push @movies_formatted, $rec;
      }
    }
    else {
    
      my @in_avis  = get_avis_in_directory($MOVIE_IN_PATH);
    
      for my $in_avi (@in_avis) {
    
        next unless $in_avi =~ /^(.+?)\s*CD\s*1\.avi$/i;
    
        my $name = $1;
    
        my $rec = {};
        $rec->{part1} = $in_avi;
        ($rec->{part2}) = grep /^\Q$name\E\s*CD\s*2\.avi$/i, @in_avis;
    
        if ($rec->{part2}) {
          $rec->{$_} = File::Spec->catfile($MOVIE_IN_PATH, $rec->{$_}) for qw/ part1 part2 /;
          $rec->{output} = File::Spec->catfile($MOVIE_OUT_PATH, "$name.avi");
          push @movies_formatted, $rec;
        }
    
      }
    }
    
    for my $movie (@movies_formatted) {
    
      my $convert_cmd = sprintf '"%s" --load "%s" --append "%s" --force-smart --save "%s" --quit',
          $AVIDEMUX, @{$movie}{qw/ part1 part2 output /};
      print "$convert_cmd\n";
    
      die "Unable to convert $movie->{'output'}!\n" if (system $convert_cmd);
    }