[PATCH] threading: deal with improperly-terminated References headers

[PATCH] threading: deal with improperly-terminated References headers

[Eric Wong]

We should not blindly join References and In-Reply-To headers
as a single string, because some messages can have an open
angle brace '<' in References: without a corresponding '>'.
---
 lib/PublicInbox/SearchIdx.pm | 5 ++---
 lib/PublicInbox/View.pm      | 5 ++---
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index 0824db0..cfb9a08 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -414,9 +414,8 @@ sub link_message {
 
 	# last References should be IRT, but some mail clients do things
 	# out of order, so trust IRT over References iff IRT exists
-	my @refs = ($hdr->header_raw('References'),
-			$hdr->header_raw('In-Reply-To'));
-	@refs = ((join(' ', @refs)) =~ /<([^>]+)>/g);
+	my @refs = (($hdr->header_raw('References') || '') =~ /<([^>]+)>/g);
+	push(@refs, (($hdr->header_raw('In-Reply-To') || '') =~ /<([^>]+)>/g));
 
 	my $tid;
 	if (@refs) {
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 7454acb..b39c820 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -104,9 +104,8 @@ EOF
 sub in_reply_to {
 	my ($hdr) = @_;
 	my %mid = map { $_ => 1 } $hdr->header_raw('Message-ID');
-	my @refs = ($hdr->header_raw('References'),
-			$hdr->header_raw('In-Reply-To'));
-	@refs = ((join(' ', @refs)) =~ /<([^>]+)>/g);
+	my @refs = (($hdr->header_raw('References') || '') =~ /<([^>]+)>/g);
+	push(@refs, (($hdr->header_raw('In-Reply-To') || '') =~ /<([^>]+)>/g));
 	while (defined(my $irt = pop @refs)) {
 		next if $mid{"<$irt>"};
 		return $irt;
-- 
EW