From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <e@80x24.org>
X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net
X-Spam-Level: 
X-Spam-ASN:  
X-Spam-Status: No, score=-4.0 required=3.0 tests=ALL_TRUSTED,BAYES_00
	shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2
Received: from localhost (dcvr.yhbt.net [127.0.0.1])
	by dcvr.yhbt.net (Postfix) with ESMTP id 37AC41FA10
	for <meta@public-inbox.org>; Wed, 13 Oct 2021 10:16:12 +0000 (UTC)
From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH 4/7] eml: avoid Encode 2.87..3.12 leak
Date: Wed, 13 Oct 2021 10:16:08 +0000
Message-Id: <20211013101611.22962-5-e@80x24.org>
In-Reply-To: <20211013101611.22962-1-e@80x24.org>
References: <20211013101611.22962-1-e@80x24.org>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
List-Id: <meta.public-inbox.org>

Encode::FB_CROAK leaks memory in old versions of Encode:
<https://rt.cpan.org/Public/Bug/Display.html?id=139622>

Since I expect there's still many users on old systems and old
Perls, we can use "$SIG{__WARN__} = \&croak" here with
Encode::FB_WARN to emulate Encode::FB_CROAK behavior.
---
 lib/PublicInbox/Eml.pm | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/lib/PublicInbox/Eml.pm b/lib/PublicInbox/Eml.pm
index 0867a016..69c26932 100644
--- a/lib/PublicInbox/Eml.pm
+++ b/lib/PublicInbox/Eml.pm
@@ -28,7 +28,7 @@ package PublicInbox::Eml;
 use strict;
 use v5.10.1;
 use Carp qw(croak);
-use Encode qw(find_encoding decode encode); # stdlib
+use Encode qw(find_encoding); # stdlib
 use Text::Wrap qw(wrap); # stdlib, we need Perl 5.6+ for $huge
 use MIME::Base64 3.05; # Perl 5.10.0 / 5.9.2
 use MIME::QuotedPrint 3.05; # ditto
@@ -334,9 +334,14 @@ sub body_set {
 
 sub body_str_set {
 	my ($self, $body_str) = @_;
-	my $charset = ct($self)->{attributes}->{charset} or
+	my $cs = ct($self)->{attributes}->{charset} //
 		croak('body_str was given, but no charset is defined');
-	body_set($self, \(encode($charset, $body_str, Encode::FB_CROAK)));
+	my $enc = find_encoding($cs) // croak "unknown encoding `$cs'";
+	$body_str = do {
+		local $SIG{__WARN__} = \&croak;
+		$enc->encode($body_str, Encode::FB_WARN);
+	};
+	body_set($self, \$body_str);
 }
 
 sub content_type { scalar header($_[0], 'Content-Type') }
@@ -452,15 +457,17 @@ sub body {
 sub body_str {
 	my ($self) = @_;
 	my $ct = ct($self);
-	my $charset = $ct->{attributes}->{charset};
-	if (!$charset) {
-		if ($STR_TYPE{$ct->{type}} && $STR_SUBTYPE{$ct->{subtype}}) {
+	my $cs = $ct->{attributes}->{charset} // do {
+		($STR_TYPE{$ct->{type}} && $STR_SUBTYPE{$ct->{subtype}}) and
 			return body($self);
-		}
 		croak("can't get body as a string for ",
 			join("\n\t", header_raw($self, 'Content-Type')));
-	}
-	decode($charset, body($self), Encode::FB_CROAK);
+	};
+	my $enc = find_encoding($cs) or croak "unknown encoding `$cs'";
+	my $tmp = body($self);
+	# workaround https://rt.cpan.org/Public/Bug/Display.html?id=139622
+	local $SIG{__WARN__} = \&croak;
+	$enc->decode($tmp, Encode::FB_WARN);
 }
 
 sub as_string {