#!/usr/bin/perl -w

my @score;
my @from;
my @subj;

# Setup a whitelist hash because it is faster than grepping the file every time.
my %whitelist;
open(WL, "/rigel4/rreid/mail/good.list")
  or warn "Could not open /rigel4/rreid/mail/good.list for reading";
while(<WL>){
  chomp;
  $whitelist{$_} = 1;
}
close(WL);

my @wlscore;
my @wlfrom;
my @wlsubj;

print "Proofpoint suspected spam report\n";
print "********************************\n\n";

$/ = ">";              # Turn on quasiparagraph mode.
while(<>){
  if(/^\s*Report\s+(\d+)\s+([-a-zA-Z0-9.+_]+\@[-a-zA-Z0-9.+_]+)\s+(.*)/){
    my $currscore = $1;
    my $currfrom  = $2;
    my $currsubj  = $3;

    # Shorten $currfrom if necessary by removing all but .domain.abc
    if(length($currfrom) > 32){
      $currfrom =~ s/\@.+(\.[^.]+\.[^.]+)$/\@..$1/;
    }

    if(defined $whitelist{$currfrom}){
      push @wlscore, $currscore;
      push @wlfrom,  $currfrom;
      push @wlsubj,  $currsubj;
      next;
    }

    # Skip this spam if $currsubj has an excess of funny characters...
    my @funnychars = ($currsubj =~ /[[:^ascii:]]/g);

    # ...or would if transliterated into the proper charset.
    if($currsubj =~ /^=\?(iso-2022|euc)-kr/){
      push @funnychars, ("a", "b", "c", "d", "e");  # Any old 5 will do.
    }

    if(@funnychars < 5){
      # Truncate those long UTF-8 and Windows-1251 Subject:s.
      if($currsubj =~ /^=\?utf-8\?/ || $currsubj =~ /^=\?Windows-1251\?/){
	$currsubj =~ s/^=\?([^?]+).*/$1/;
      }

      push @score, $currscore;
      push @from,  $currfrom;
      push @subj,  $currsubj;
    }
  }
}

my $maxlength = 0;
my $formatstr;

if(@wlscore){
  print "Whitelisted:\n";
  print "------------\n";
  foreach(@wlfrom){
    my $templength = length;
    if($maxlength < $templength){
      $maxlength = $templength;
    }
  }
  $formatstr = "!!( \%3d  \%-${maxlength}s  \%s )!!\n";

  for (my $i = 0; $i < @wlfrom; ++$i){
    printf($formatstr, $wlscore[$i], $wlfrom[$i], $wlsubj[$i]);
  }

  if(@from){
    $maxlength = 0;
    print "\nSuspected spam:\n";
    print   "---------------\n";
  }
}

foreach(@from){
  my $templength = length;
  if($maxlength < $templength){
    $maxlength = $templength;
  }
}

$formatstr = "\%3d  \%-${maxlength}s  \%s\n";
for (my $i = 0; $i < @from; ++$i){
  printf($formatstr, $score[$i], $from[$i], $subj[$i]);
}