X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FFilter.pm;h=5a023cf7032b0e27573932238c6349b1464f5004;hb=f76f265a851944b5dedcc3be5f3b5224b6ebda89;hp=49ba5cb2df452ea72a8991dbfbeabf5fad997745;hpb=f66473b1a41606d8b4cda74c551aa85d12ec37ef;p=public-inbox.git

diff --git a/lib/PublicInbox/Filter.pm b/lib/PublicInbox/Filter.pm
index 49ba5cb2..5a023cf7 100644
--- a/lib/PublicInbox/Filter.pm
+++ b/lib/PublicInbox/Filter.pm
@@ -1,4 +1,4 @@
-# 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)
 #
 # This only exposes one function: run
@@ -15,7 +15,7 @@ our $VERSION = '0.0.1';
 use constant NO_HTML => '*** We only accept plain-text email, no HTML ***';
 
 # start with the same defaults as mailman
-our $BAD_EXT = qr/\.(?:exe|bat|cmd|com|pif|scr|vbs|cpl)\z/i;
+our $BAD_EXT = qr/\.(exe|bat|cmd|com|pif|scr|vbs|cpl|zip)\s*\z/i;
 our $MIME_HTML = qr!\btext/html\b!i;
 our $MIME_TEXT_ANY = qr!\btext/[a-z0-9\+\._-]+\b!i;
 
@@ -127,6 +127,7 @@ sub strip_multipart {
 		# some extensions are just bad, reject them outright
 		my $fn = $part->filename;
 		if (defined($fn) && $fn =~ $BAD_EXT) {
+			$filter->reject("Bad file type: $1") if $filter;
 			$rejected++;
 			return;
 		}
@@ -144,16 +145,18 @@ sub strip_multipart {
 
 			push @keep, $part;
 		} elsif ($part_type eq '' ||
-		         $part_type =~ m!\Aapplication/octet-stream\z!i) {
+		         $part_type =~ m!\bapplication/octet-stream\b!i) {
 			# unfortunately, some mailers don't set correct types,
 			# let messages of unknown type through but do not
 			# change the sender-specified type
 			if (recheck_type_ok($part)) {
 				push @keep, $part;
+			} elsif ($filter) {
+				$filter->reject('no attachments')
 			} else {
 				$rejected++;
 			}
-		} elsif ($part_type =~ m!\Aapplication/pgp-signature\z!i) {
+		} elsif ($part_type =~ m!\bapplication/pgp-signature\b!i) {
 			# PGP signatures are not huge, we may keep them.
 			# They can only be valid if it's the last element,
 			# so we keep them iff the message is unmodified:
@@ -161,6 +164,7 @@ sub strip_multipart {
 				push @keep, $part;
 			}
 		} else {
+			$filter->reject('no attachments') if $filter;
 			# reject everything else, including non-PGP signatures
 			$rejected++;
 		}
@@ -207,7 +211,7 @@ sub collapse {
 	$mime->header_set('Content-Type', $part->content_type);
 	$mime->body_set($part->body_raw);
 	my $cte = $part->header('Content-Transfer-Encoding');
-	if (defined($cte) && length($cte)) {
+	if (defined($cte) && $cte ne '') {
 		$mime->header_set('Content-Transfer-Encoding', $cte);
 	}
 	mark_changed($mime);