X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FMDA.pm;h=33696528f8c1468e8b2b05d597df4e84675c21cd;hb=95bdac7f09c69036efed537a4d03d5bdd2ae4eb6;hp=bb14ae50cf27a8d28dd9bf87d40686814a2f8650;hpb=a8d9e2f1853032016db6ff177979873b3bdadd85;p=public-inbox.git

diff --git a/lib/PublicInbox/MDA.pm b/lib/PublicInbox/MDA.pm
index bb14ae50..33696528 100644
--- a/lib/PublicInbox/MDA.pm
+++ b/lib/PublicInbox/MDA.pm
@@ -1,13 +1,27 @@
-# Copyright (C) 2013, Eric Wong <normalperson@yhbt.net> and all contributors
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2013-2020 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <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::Address;
-use Encode qw/decode encode/;
-use Date::Parse qw(strptime);
-use constant MAX_SIZE => 1024 * 500; # same as spamc default
-use constant cmd => qw/ssoma-mda -1/;
+use Email::Simple;
+use PublicInbox::MsgTime;
+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
+
+our @BAD_HEADERS = (
+	# postfix
+	qw(delivered-to x-original-to), # prevent training loops
+
+	# The rest are taken from Mailman 2.1.15:
+	# could contain passwords:
+	qw(approved approve x-approved x-approve urgent),
+	# could be used phishing:
+	qw(return-receipt-to disposition-notification-to x-confirm-reading-to),
+	# Pegasus mail:
+	qw(x-pmrqc)
+);
 
 # drop plus addressing for matching
 sub __drop_plus {
@@ -18,15 +32,17 @@ 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 $mid = $simple->header("Message-ID");
+	my ($klass, $simple, $address) = @_;
+	my @mid = $simple->header('Message-ID');
+	return 0 if scalar(@mid) != 1;
+	my $mid = $mid[0];
+	return 0 if (length($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('u@h'), $simple->header("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, $recipient);
+	alias_specified($simple, $address);
 }
 
 sub usable_str {
@@ -35,52 +51,56 @@ sub usable_str {
 }
 
 sub usable_date {
-	my @t = eval { strptime(@_) };
-	scalar @t;
+	defined(eval { PublicInbox::MsgTime::str2date_zone($_[0]) });
 }
 
-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) =~ $oaddr) {
-			return 1;
+sub alias_specified {
+	my ($simple, $address) = @_;
+
+	my @address = ref($address) eq 'ARRAY' ? @$address : ($address);
+	my %ok = map {
+		lc(__drop_plus($_)) => 1;
+	} @address;
+
+	foreach my $line ($simple->header('Cc'), $simple->header('To')) {
+		my @addrs = ($line =~ /([^,<\s]+\@[^,>\s]+)/g);
+		foreach my $addr (@addrs) {
+			if ($ok{lc(__drop_plus($addr))}) {
+				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 $url = $dst->{url};
-	if (defined $url) {
-		$simple->header_set("List-Archive", "<$url>");
-		$simple->header_set("List-Help", "<${url}help>");
+	unless (defined $simple->header('List-Id')) {
+		my $pa = $dst->{-primary_address};
+		$pa =~ tr/@/./; # RFC2919
+		$simple->header_set("List-Id", "<$pa>");
 	}
 }
 
-# returns a 3-element array: name, email, date
-sub author_info {
-	my ($class, $simple) = @_;
+sub inboxes_for_list_id ($$) {
+	my ($klass, $config, $simple) = @_;
 
-	my $from = decode('MIME-Header', $simple->header('From'));
-	$from = encode('utf8', $from);
-	my @from = Email::Address->parse($from);
-	my $name = $from[0]->name;
-	defined $name or $name = '';
-	my $email = $from[0]->address;
-	defined $email or $email = '';
-	($name, $email, $simple->header('Date'));
+	# newer Email::Simple allows header_raw, as does Email::MIME:
+	my @list_ids = $simple->can('header_raw') ?
+			$simple->header_raw('List-Id') :
+			$simple->header('List-Id');
+	my @dests;
+	for my $list_id (@list_ids) {
+		$list_id =~ /<[ \t]*(.+)?[ \t]*>/ or next;
+		if (my $ibx = $config->lookup_list_id($1)) {
+			push @dests, $ibx;
+		}
+	}
+	if (scalar(@list_ids) > 1) {
+		warn "W: multiple List-IDs in message:\n";
+		warn "W: List-ID: $_\n" for @list_ids
+	}
+	\@dests;
 }
 
 1;