msgiter: attempt to decode all text/* bodies

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

diff --git a/lib/PublicInbox/MsgIter.pm b/lib/PublicInbox/MsgIter.pm
index 5be06a44be16dc96d064aa7b3eefbaff8e021f97..d9df32abc37c7fc7da68c0e09ed2f942ddbb4827 100644 (file)
--- a/lib/PublicInbox/MsgIter.pm
+++ b/lib/PublicInbox/MsgIter.pm
@@ -1,11 +1,12 @@
-# Copyright (C) 2016 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2019 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
-#
+
+# read-only utilities for Email::MIME
 package PublicInbox::MsgIter;
 use strict;
 use warnings;
 use base qw(Exporter);
-our @EXPORT = qw(msg_iter);
+our @EXPORT = qw(msg_iter msg_part_text);
 use PublicInbox::MIME;
 
 # Like Email::MIME::walk_parts, but this is:
@@ -34,4 +35,34 @@ sub msg_iter ($$) {
        }
 }
 
+sub msg_part_text ($$) {
+       my ($part, $ct) = @_;
+
+       my $s = eval { $part->body_str };
+       my $err = $@;
+
+       # text/plain is the default, multipart/mixed happened a few
+       # times when it should not have been:
+       #   <87llgalspt.fsf@free.fr>
+       #   <200308111450.h7BEoOu20077@mail.osdl.org>
+       if ($err && ($ct =~ m!\btext/\b!i ||
+                       $ct =~ m!\bmultipart/mixed\b!i)) {
+               my $cte = $part->header_raw('Content-Transfer-Encoding');
+               if (defined($cte) && $cte =~ /\b7bit\b/i) {
+                       $s = $part->body;
+                       $err = undef if $s =~ /\A[[:ascii:]]+\z/s;
+               } else {
+                       # Try to assume UTF-8 because Alpine seems to
+                       # do wacky things and set charset=X-UNKNOWN
+                       $part->charset_set('UTF-8');
+                       $s = eval { $part->body_str };
+               }
+
+               # If forcing charset=UTF-8 failed,
+               # caller will warn further down...
+               $s = $part->body if $@;
+       }
+       ($s, $err);
+}
+
 1;