From 8ea217f4452e34776ba294b5090827c99656cada Mon Sep 17 00:00:00 2001
From: Eric Wong <e@80x24.org>
Date: Sun, 30 Aug 2015 00:38:05 +0000
Subject: search: do not index references and inreplyto terms

We no longer need them, as we can rely on index-time thread
resolution and thread merging.  This allows us to index less
data and hopefully increase efficiency.
---
 lib/PublicInbox/Search.pm    | 17 ++---------------
 lib/PublicInbox/SearchIdx.pm |  5 -----
 2 files changed, 2 insertions(+), 20 deletions(-)

(limited to 'lib/PublicInbox')

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index 4b3830e2..d3faaebb 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -24,7 +24,8 @@ use constant {
 	# 4 - change "Re: " normalization, avoid circular Reference ghosts
 	# 5 - subject_path drops trailing '.'
 	# 6 - preserve References: order in document data
-	SCHEMA_VERSION => 6,
+	# 7 - remove references and inreplyto terms
+	SCHEMA_VERSION => 7,
 	QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD,
 };
 
@@ -37,8 +38,6 @@ my %bool_pfx_internal = (
 my %bool_pfx_external = (
 	path => 'XPATH',
 	thread => 'G', # newsGroup (or similar entity - e.g. a web forum name)
-	references => 'XREFS',
-	inreplyto => 'XIRT',
 );
 
 my %prob_prefix = (
@@ -87,18 +86,6 @@ sub get_subject_path {
 	$self->do_enquire($query, $opts);
 }
 
-# given a message ID, get followups to a message
-sub get_followups {
-	my ($self, $mid, $opts) = @_;
-	$mid = mid_clean($mid);
-	$mid = mid_compress($mid);
-	my $qp = $self->qp;
-	my $irt = $qp->parse_query("inreplyto:$mid", 0);
-	my $ref = $qp->parse_query("references:$mid", 0);
-	my $query = Search::Xapian::Query->new(OP_OR, $irt, $ref);
-	$self->do_enquire($query, $opts);
-}
-
 sub get_thread {
 	my ($self, $mid, $opts) = @_;
 	my $smsg = eval { $self->lookup_message($mid) };
diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index 2ac53a73..dec33334 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -238,20 +238,15 @@ sub link_message_to_parents {
 		}
 	}
 	if (@refs) {
-		$doc->add_term(xpfx('inreplyto') . $irt) if defined $irt;
 		$smsg->{references_sorted} = '<'.join('><', @refs).'>';
 
-		my $ref_pfx = xpfx('references');
-
 		# first ref *should* be the thread root,
 		# but we can never trust clients to do the right thing
 		my $ref = shift @refs;
-		$doc->add_term($ref_pfx . $ref);
 		$tid = $self->_resolve_mid_to_tid($ref);
 
 		# the rest of the refs should point to this tid:
 		foreach $ref (@refs) {
-			$doc->add_term($ref_pfx . $ref);
 			my $ptid = $self->_resolve_mid_to_tid($ref);
 			if ($tid ne $ptid) {
 				$self->merge_threads($tid, $ptid);
-- 
cgit v1.2.3-24-ge0c7