From fc62e40e4e3d00a00377ba26aeca010880158caa Mon Sep 17 00:00:00 2001
From: "Eric Wong (Contractor, The Linux Foundation)" <e@80x24.org>
Date: Fri, 16 Feb 2018 02:59:11 +0000
Subject: search: stop assuming Message-ID is unique

In general, they are, but there's no way for or general purpose
mail server to enforce that.  This is a step in allowing us
to handle more corner cases which existing lists throw at us.
---
 lib/PublicInbox/ExtMsg.pm    |  2 +-
 lib/PublicInbox/Search.pm    | 14 ++++++++++++--
 lib/PublicInbox/SearchIdx.pm | 10 ++++++----
 3 files changed, 19 insertions(+), 7 deletions(-)

(limited to 'lib')

diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm
index 4e31ef0f..90d68db8 100644
--- a/lib/PublicInbox/ExtMsg.pm
+++ b/lib/PublicInbox/ExtMsg.pm
@@ -46,7 +46,7 @@ sub ext_msg {
 		}
 
 		# try to find the URL with Xapian to avoid forking
-		my $doc_id = eval { $s->find_unique_doc_id('XMID' . $mid) };
+		my $doc_id = eval { $s->find_first_doc_id('XMID' . $mid) };
 		if ($@) {
 			# xapian not configured properly for this repo
 			push @nox, $other;
diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index 3ec96ca9..33a1f2d3 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -56,7 +56,7 @@ my %bool_pfx_internal = (
 );
 
 my %bool_pfx_external = (
-	mid => 'XMID', # uniQue id (Message-ID)
+	mid => 'XMID', # Message-ID (full/exact)
 );
 
 my %prob_prefix = (
@@ -285,7 +285,7 @@ sub lookup_message {
 	my ($self, $mid) = @_;
 	$mid = mid_clean($mid);
 
-	my $doc_id = $self->find_unique_doc_id('XMID' . $mid);
+	my $doc_id = $self->find_first_doc_id('XMID' . $mid);
 	my $smsg;
 	if (defined $doc_id) {
 		# raises on error:
@@ -327,6 +327,16 @@ sub find_doc_ids {
 	($db->postlist_begin($termval), $db->postlist_end($termval));
 }
 
+sub find_first_doc_id {
+	my ($self, $termval) = @_;
+
+	my ($begin, $end) = $self->find_doc_ids($termval);
+
+	return undef if $begin->equal($end); # not found
+
+	$begin->get_docid;
+}
+
 # normalize subjects so they are suitable as pathnames for URLs
 # XXX: consider for removal
 sub subject_path {
diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index fa5057fd..265403a3 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -366,12 +366,14 @@ sub remove_message {
 	$mid = mid_clean($mid);
 
 	eval {
-		$doc_id = $self->find_unique_doc_id('XMID' . $mid);
-		if (defined $doc_id) {
-			$db->delete_document($doc_id);
-		} else {
+		my ($head, $tail) = $self->find_doc_ids('XMID' . $mid);
+		if ($head->equal($tail)) {
 			warn "cannot remove non-existent <$mid>\n";
 		}
+		for (; $head != $tail; $head->inc) {
+			my $docid = $head->get_docid;
+			$db->delete_document($docid);
+		}
 	};
 
 	if ($@) {
-- 
cgit v1.2.3-24-ge0c7