#!/usr/bin/perl -w
#
# Converts a diff between two Audible.Com catalogue .tab files
# such as are found at
#    http://www.audible.com/catalog/catalog-audiobooks.tab
# into a human-readable format.
#
# Suggested to be used with the 'audible_newbooks' shell script.
#
# Last updated by rb on Wed Feb 25 17:25:12 EST 2004
#

use strict;
use vars qw/%TITLES %ALL_DATA $DEBUG @categories/;


$DEBUG = 0;
#$DEBUG = 1;

# Order of categories in diff file (fixed)
@categories = ( 'Title', 'Author', 'Author Surname', 'Narrator', 'Running Time', 'Price', 'Abridgedness', 'Category', 'Formats');

#
# Main code begins here
#

while (<>) {
   #$ALL_DATA{$_}++ if s/^> //;
   #$ALL_DATA{$_}-- if s/^< //;
   if (s/^< //) {
      $ALL_DATA{$_}--;
     
      my @tmp = split(/\t/);
      my $title = shift(@tmp);
      #$TITLES{$title}--;
      
      #$DEBUG && print "Score down to $ALL_DATA{$_}/" .
      #                $TITLES{$title} ? $TITLES{$title} : 0 .
      #                " for $title\n";
   } elsif (s/^> //) {
      $ALL_DATA{$_}++;
      
      # For new books we check the title isn't replicated.
      my @tmp = split(/\t/);
      my $title = shift(@tmp);
      $TITLES{$title}++;
      
     # $DEBUG && print "Score up to $ALL_DATA{$_}/$TITLES{$title} for $title\n";
   }
}

=pod
foreach (keys %TITLES) {
   my $title = $_;
   my $score = $TITLES{$_};

   next unless $score > 1;  # Only >1 titles are doubles

   $DEBUG && print "Double Title Detected:  $title\n";

   my @matchlines;
   foreach (sort keys %ALL_DATA) {
      my @data = split(/\t/);
      my $test_title = shift(@data);

      if ($test_title eq $title) {
        push @matchlines, $title;
        $DEBUG && "Match number $#matchlines for $title!\n";
      }
   }
   
   # XXX TODO
   # deal with @matchlines here - find differences and point out.
   # put into printable form in an array then just join the array
   # on "\n" and print it!

}
=cut

foreach (sort keys %ALL_DATA) {
   my $score = $ALL_DATA{$_};
   next unless $score > 0;                  # Only +ve scores are new books
   
   $DEBUG && print "Score for this book:  $score\n";
   
   my @data = split(/\t/);
   my $tmp; 
   foreach (@categories) {
      if ($_ eq "Author Surname") {
         # don't bother printing.
         $tmp = shift(@data);
      } elsif ($_ eq "Title") {
         # take off (Unabridged) at EOL - double info annoys
         $tmp = shift(@data);
         $tmp =~ s/\s+\((un)?abridged\)//i;
         print "$_: $tmp\n";
      } elsif ($_ eq "Abridgedness") {
         # sometimes not specified
         $tmp = shift(@data);
         $tmp ||= "Unknown";
         print "$_: $tmp\n";
      } elsif ($_ eq "Formats") {
         # handle specially - at end of line so contains all the rest
         $tmp = join(" ", @data);
         print "$_: $tmp\n";
      } else {
         # Most ones.
         $tmp = shift(@data);
         print "$_: $tmp\n";
      }
   }
   print "\n";
}

#
# Ende.
#