From: Eric Wong <e@80x24.org>
Date: Sat, 3 Apr 2021 10:48:26 +0000 (+0000)
Subject: lei: improve handling of Message-ID-less draft messages
X-Git-Tag: v1.7.0~827
X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=commitdiff_plain;h=b3e2975029ae938bb232aaa0cbc3dabda55d57d6

lei: improve handling of Message-ID-less draft messages

We need a stable fallback time for digest2mid in the presence
of messages without Received/Date headers.  Furthermore, we
must avoid using uninitialized smsg->{mid} when parsing
References for draft replies.
---

diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 34738279..46f57e27 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -510,8 +510,8 @@ sub atfork_child {
 	}
 }
 
-sub digest2mid ($$) {
-	my ($dig, $hdr) = @_;
+sub digest2mid ($$;$) {
+	my ($dig, $hdr, $fallback_time) = @_;
 	my $b64 = $dig->clone->b64digest;
 	# Make our own URLs nicer:
 	# See "Base 64 Encoding with URL and Filename Safe Alphabet" in RFC4648
@@ -520,7 +520,7 @@ sub digest2mid ($$) {
 	# Add a date prefix to prevent a leading '-' in case that trips
 	# up some tools (e.g. if a Message-ID were a expected as a
 	# command-line arg)
-	my $dt = msg_datestamp($hdr);
+	my $dt = msg_datestamp($hdr, $fallback_time);
 	$dt = POSIX::strftime('%Y%m%d%H%M%S', gmtime($dt));
 	"$dt.$b64" . '@z';
 }
diff --git a/lib/PublicInbox/LeiSearch.pm b/lib/PublicInbox/LeiSearch.pm
index 69ba8303..148aa185 100644
--- a/lib/PublicInbox/LeiSearch.pm
+++ b/lib/PublicInbox/LeiSearch.pm
@@ -58,7 +58,7 @@ sub content_key ($) {
 			qw(Message-ID X-Alt-Message-ID Resent-Message-ID));
 	unless (@$mids) {
 		$eml->{-lei_fake_mid} = $mids->[0] =
-				PublicInbox::Import::digest2mid($dig, $eml);
+				PublicInbox::Import::digest2mid($dig, $eml, 0);
 	}
 	($chash, $mids);
 }
diff --git a/lib/PublicInbox/OverIdx.pm b/lib/PublicInbox/OverIdx.pm
index e1cd31b9..66dec099 100644
--- a/lib/PublicInbox/OverIdx.pm
+++ b/lib/PublicInbox/OverIdx.pm
@@ -264,8 +264,10 @@ sub add_overview {
 	$smsg->{lines} = $eml->body_raw =~ tr!\n!\n!;
 	my $mids = mids_for_index($eml);
 	my $refs = $smsg->parse_references($eml, $mids);
-	$mids->[0] //= $smsg->{mid} //= $eml->{-lei_fake_mid};
-	$smsg->{mid} //= '';
+	$mids->[0] //= do {
+		$smsg->{mid} //= '';
+		$eml->{-lei_fake_mid};
+	};
 	my $subj = $smsg->{subject};
 	my $xpath;
 	if ($subj ne '') {
diff --git a/lib/PublicInbox/Smsg.pm b/lib/PublicInbox/Smsg.pm
index b4cc2ecb..da8ce590 100644
--- a/lib/PublicInbox/Smsg.pm
+++ b/lib/PublicInbox/Smsg.pm
@@ -76,7 +76,7 @@ sub parse_references ($$$) {
 	return $refs if scalar(@$refs) == 0;
 
 	# prevent circular references here:
-	my %seen = ( $smsg->{mid} => 1 );
+	my %seen = ( ($smsg->{mid} // '') => 1 );
 	my @keep;
 	foreach my $ref (@$refs) {
 		if (length($ref) > PublicInbox::MID::MAX_MID_SIZE) {
diff --git a/t/lei-import.t b/t/lei-import.t
index 99289748..9bb4e1fa 100644
--- a/t/lei-import.t
+++ b/t/lei-import.t
@@ -79,6 +79,27 @@ is($res->[1], undef, 'only one result');
 is($res->[0]->{'m'}, 'k@y', 'got expected message');
 is_deeply($res->[0]->{kw}, ['seen'], "`seen' keywords set");
 
+# no From, Sender, or Message-ID
+$eml_str = <<'EOM';
+Subject: draft message with no sender
+References: <y@y>
+
+No use for a name
+EOM
+lei_ok([qw(import -F eml -)], undef, { %$lei_opt, 0 => \$eml_str });
+lei_ok(['q', 's:draft message with no sender']);
+my $draft_a = json_utf8->decode($lei_out);
+ok(!exists $draft_a->[0]->{'m'}, 'no fake mid stored or exposed');
+lei_ok([qw(tag -F eml - +kw:draft)], undef, { %$lei_opt, 0 => \$eml_str });
+lei_ok(['q', 's:draft message with no sender']);
+my $draft_b = json_utf8->decode($lei_out);
+my $kw = delete $draft_b->[0]->{kw};
+is_deeply($kw, ['draft'], 'draft kw set');
+is_deeply($draft_a, $draft_b, 'fake Message-ID lookup') or
+				diag explain($draft_a, $draft_b);
+lei_ok('blob', '--mail', $draft_b->[0]->{blob});
+is($lei_out, $eml_str, 'draft retrieved by blob');
+
 # see t/lei_to_mail.t for "import -F mbox*"
 });
 done_testing;