#!/usr/bin/perl -w
#
#  A simple utility which will process each "*.txt" file inside a named
# directory.
#
#  Any entries with a duplicate title will be alerted.
#
# Steve
#

use strict;
use warnings;


#
#  Get the directory and ensure it exists.
#
my $dir = shift;
die "Usage: $0 directory" unless ( defined($dir) );
die "Not a directory : $dir" unless ( -d $dir );


#
#  The hash of arrays
#
my $titles;


#
#  Find the files.
#
foreach my $file ( sort( glob( $dir . "/*" ) ) )
{

    #
    #  Skip non-files
    #
    next if ( -d $file );
    next if ( $file =~ /~$/ );

    #
    #  Title for this entry.
    #
    my $title = undef;

    #
    #  Read each file, and look for the title.
    #
    open( IN, "<", $file )
      or die "Failed to read $file - $!";
    foreach my $line (<IN>)
    {
        $title = $2
          if ( !defined($title)
               && ( $line =~ /^(Subject|Title):(.*)/i ) );

    }
    close(IN);

    #
    #  Ignore entries with no title.
    #
    next if ( !$title );

    #
    #  Get current entries we might have found with
    # this title; and add on the new one.
    #
    my $a = $titles->{ $title };
    push( @$a, $file );
    $titles->{ $title } = $a;
}


#
#  Now look for dupes
#
my $fail = 0;
foreach my $title ( keys %$titles )
{
    my $a = $titles->{ $title };

    if ( ( scalar @$a ) > 1 )
    {
        print "Title: $title\n";
        foreach my $file (@$a)
        {
            print "\t" . $file . "\n";
        }

        $fail = 1;
    }
}


exit($fail);