use raw header for Message-ID

[public-inbox.git] / lib / PublicInbox / MDA.pm

diff --git a/lib/PublicInbox/MDA.pm b/lib/PublicInbox/MDA.pm
index bb14ae50cf27a8d28dd9bf87d40686814a2f8650..003bac659e6166ef73d29613ac372fc6cb382702 100644 (file)
--- a/lib/PublicInbox/MDA.pm
+++ b/lib/PublicInbox/MDA.pm
@@ -1,12 +1,15 @@
-# Copyright (C) 2013, Eric Wong <normalperson@yhbt.net> and all contributors
+# Copyright (C) 2013-2015 all contributors <meta@public-inbox.org>
 # License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+#
+# For the -mda script (mail delivery agent)
 package PublicInbox::MDA;
 use strict;
 use warnings;
+use Email::Simple;
 use Email::Address;
-use Encode qw/decode encode/;
 use Date::Parse qw(strptime);
-use constant MAX_SIZE => 1024 * 500; # same as spamc default
+use constant MAX_SIZE => 1024 * 500; # same as spamc default, should be tunable
+use constant MAX_MID_SIZE => 244; # max term size - 1 in Xapian
 use constant cmd => qw/ssoma-mda -1/;
 
 # drop plus addressing for matching
@@ -18,15 +21,16 @@ sub __drop_plus {
 
 # do not allow Bcc, only Cc and To if recipient is set
 sub precheck {
-       my ($klass, $filter, $recipient) = @_;
-       my $simple = $filter->simple;
+       my ($klass, $filter, $address) = @_;
+       my Email::Simple $simple = $filter->simple;
        my $mid = $simple->header("Message-ID");
+       return 0 if (length($mid) > MAX_MID_SIZE);
        return 0 unless usable_str(length('<m@h>'), $mid) && $mid =~ /\@/;
        return 0 unless usable_str(length('u@h'), $filter->from);
        return 0 unless usable_str(length(':o'), $simple->header("Subject"));
        return 0 unless usable_date($simple->header("Date"));
        return 0 if length($simple->as_string) > MAX_SIZE;
-       recipient_specified($filter, $recipient);
+       alias_specified($filter, $address);
 }
 
 sub usable_str {
@@ -39,48 +43,49 @@ sub usable_date {
        scalar @t;
 }
 
-sub recipient_specified {
-       my ($filter, $recipient) = @_;
-       defined($recipient) or return 1; # for mass imports
-       my @recip = Email::Address->parse($recipient);
-       my $oaddr = __drop_plus($recip[0]->address);
-       $oaddr = qr/\b\Q$oaddr\E\b/i;
-       my @to = Email::Address->parse($filter->to);
-       my @cc = Email::Address->parse($filter->cc);
-       foreach my $addr (@to, @cc) {
-               if (__drop_plus($addr->address) =~ $oaddr) {
-                       return 1;
+sub alias_specified {
+       my ($filter, $address) = @_;
+
+       my @address = ref($address) eq 'ARRAY' ? @$address : ($address);
+       my %ok = map {
+               my @recip = Email::Address->parse($_);
+               lc(__drop_plus($recip[0]->address)) => 1;
+       } @address;
+
+       foreach my $line ($filter->cc, $filter->to) {
+               foreach my $addr (Email::Address->parse($line)) {
+                       if ($ok{lc(__drop_plus($addr->address))}) {
+                               return 1;
+                       }
                }
        }
        return 0;
 }
 
-# RFC2919 and RFC2369
 sub set_list_headers {
        my ($class, $simple, $dst) = @_;
-       my $pa = "<$dst->{-primary_address}>";
-       $simple->header_set("List-Id", $pa);
-       $simple->header_set("List-Post", $pa);
+       my $pa = $dst->{-primary_address};
+
+       $simple->header_set("List-Id", "<$pa>"); # RFC2919
 
-       my $url = $dst->{url};
-       if (defined $url) {
-               $simple->header_set("List-Archive", "<$url>");
-               $simple->header_set("List-Help", "<${url}help>");
+       # remove Delivered-To: prevent training loops
+       # The rest are taken from Mailman 2.1.15, some may be used for phishing
+       foreach my $h (qw(delivered-to approved approve x-approved x-approve
+                       urgent return-receipt-to disposition-notification-to
+                       x-confirm-reading-to x-pmrqc)) {
+               $simple->header_set($h);
        }
 }
 
 # returns a 3-element array: name, email, date
 sub author_info {
-       my ($class, $simple) = @_;
+       my ($class, $mime) = @_;
 
-       my $from = decode('MIME-Header', $simple->header('From'));
-       $from = encode('utf8', $from);
+       my $from = $mime->header('From');
        my @from = Email::Address->parse($from);
        my $name = $from[0]->name;
-       defined $name or $name = '';
        my $email = $from[0]->address;
-       defined $email or $email = '';
-       ($name, $email, $simple->header('Date'));
+       ($name, $email, $mime->header('Date'));
 }
 
 1;