[PATCH] search: retry document loading from Xapian

[PATCH] search: retry document loading from Xapian

[Eric Wong]

In addition to needing to retry enquire queries, we also need
to protect document loading from the Xapian DB and retry on
modification, as it seems to throw the same errors.

Checking the $@ ref for Search::Xapian::DatabaseModifiedError
is actually in the test suite for both the XS and SWIG Xapian
bindings, so we should be good as far as forward/backwards
compatibility.
---
 lib/PublicInbox/Search.pm     | 15 ++++++++++-----
 lib/PublicInbox/SearchView.pm | 39 +++++++++++++++++++++++++++++----------
 lib/PublicInbox/View.pm       | 19 +++++++++++--------
 3 files changed, 50 insertions(+), 23 deletions(-)

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index 5e6bfc6..24cb266 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -166,22 +166,27 @@ sub get_thread {
 	_do_enquire($self, $qtid, $opts);
 }
 
-sub _do_enquire {
-	my ($self, $query, $opts) = @_;
+sub retry_reopen {
+	my ($self, $cb) = @_;
 	my $ret;
 	for (1..10) {
-		eval { $ret = _enquire_once($self, $query, $opts) };
+		eval { $ret = $cb->() };
 		return $ret unless $@;
 		# Exception: The revision being read has been discarded -
 		# you should call Xapian::Database::reopen()
-		if (index($@, 'Xapian::Database::reopen') >= 0) {
+		if (ref($@) eq 'Search::Xapian::DatabaseModifiedError') {
 			reopen($self);
 		} else {
-			die $@;
+			die;
 		}
 	}
 }
 
+sub _do_enquire {
+	my ($self, $query, $opts) = @_;
+	retry_reopen($self, sub { _enquire_once($self, $query, $opts) });
+}
+
 sub _enquire_once {
 	my ($self, $query, $opts) = @_;
 	my $enquire = $self->enquire;
diff --git a/lib/PublicInbox/SearchView.pm b/lib/PublicInbox/SearchView.pm
index 6af151a..50a2c01 100644
--- a/lib/PublicInbox/SearchView.pm
+++ b/lib/PublicInbox/SearchView.pm
@@ -59,6 +59,17 @@ sub sres_top_html {
 	PublicInbox::WwwStream->response($ctx, $code, $cb);
 }
 
+# allow undef for individual doc loads...
+sub load_doc_retry {
+	my ($srch, $mitem) = @_;
+
+	eval {
+		$srch->retry_reopen(sub {
+			PublicInbox::SearchMsg->load_doc($mitem->get_document)
+		});
+	}
+}
+
 # display non-threaded search results similar to what users expect from
 # regular WWW search engines:
 sub mset_summary {
@@ -68,10 +79,18 @@ sub mset_summary {
 	my $pad = length("$total");
 	my $pfx = ' ' x $pad;
 	my $res = \($ctx->{-html_tip});
+	my $srch = $ctx->{srch};
 	foreach my $m ($mset->items) {
 		my $rank = sprintf("%${pad}d", $m->get_rank + 1);
 		my $pct = $m->get_percent;
-		my $smsg = PublicInbox::SearchMsg->load_doc($m->get_document);
+		my $smsg = load_doc_retry($srch, $m);
+		unless ($smsg) {
+			eval {
+				$m = "$m ".$m->get_docid . " expired\n";
+				$ctx->{env}->{'psgi.errors'}->print($m);
+			};
+			next;
+		}
 		my $s = ascii_html($smsg->subject);
 		my $f = ascii_html($smsg->from_name);
 		my $ts = PublicInbox::View::fmt_ts($smsg->ts);
@@ -145,14 +164,14 @@ sub search_nav_bot {
 sub mset_thread {
 	my ($ctx, $mset, $q) = @_;
 	my %pct;
-	my @m = map {
+	my $msgs = $ctx->{srch}->retry_reopen(sub { [ map {
 		my $i = $_;
-		my $m = PublicInbox::SearchMsg->load_doc($i->get_document);
-		$pct{$m->mid} = $i->get_percent;
-		$m;
-	} ($mset->items);
+		my $smsg = PublicInbox::SearchMsg->load_doc($i->get_document);
+		$pct{$smsg->mid} = $i->get_percent;
+		$smsg;
+	} ($mset->items) ]});
 
-	my $th = PublicInbox::SearchThread->new(\@m);
+	my $th = PublicInbox::SearchThread->new($msgs);
 	$th->thread;
 	if ($q->{r}) { # order by relevance
 		$th->order(sub {
@@ -175,12 +194,11 @@ sub mset_thread {
 	$ctx->{prev_attr} = '';
 	$ctx->{prev_level} = 0;
 	$ctx->{seen} = {};
-	$ctx->{s_nr} = scalar(@m).'+ results';
+	$ctx->{s_nr} = scalar(@$msgs).'+ results';
 
 	PublicInbox::View::walk_thread($th, $ctx,
 		*PublicInbox::View::pre_thread);
 
-	my $msgs = \@m;
 	my $mime;
 	sub {
 		return unless $msgs;
@@ -217,9 +235,10 @@ sub adump {
 	my $ibx = $ctx->{-inbox};
 	my @items = $mset->items;
 	$ctx->{search_query} = $q;
+	my $srch = $ctx->{srch};
 	PublicInbox::WwwAtomStream->response($ctx, 200, sub {
 		while (my $x = shift @items) {
-			$x = PublicInbox::SearchMsg->load_doc($x->get_document);
+			$x = load_doc_retry($srch, $x);
 			$x = $ibx->msg_by_smsg($x) and
 					return Email::MIME->new($x);
 		}
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index ec5f7e0..fa47a16 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -327,8 +327,9 @@ sub stream_thread ($$) {
 sub thread_html {
 	my ($ctx) = @_;
 	my $mid = $ctx->{mid};
-	my $sres = $ctx->{srch}->get_thread($mid);
-	my $msgs = load_results($sres);
+	my $srch = $ctx->{srch};
+	my $sres = $srch->get_thread($mid);
+	my $msgs = load_results($srch, $sres);
 	my $nr = $sres->{total};
 	return missing_thread($ctx) if $nr == 0;
 	my $skel = '<hr><pre>';
@@ -574,7 +575,8 @@ sub thread_skel {
 	$ctx->{prev_attr} = '';
 	$ctx->{prev_level} = 0;
 	$ctx->{dst} = $dst;
-	walk_thread(thread_results(load_results($sres)), $ctx, *skel_dump);
+	$sres = load_results($srch, $sres);
+	walk_thread(thread_results($sres), $ctx, *skel_dump);
 	$ctx->{parent_msg} = $parent;
 }
 
@@ -733,9 +735,9 @@ sub indent_for {
 }
 
 sub load_results {
-	my ($sres) = @_;
-
-	[ map { $_->ensure_metadata; $_ } @{delete $sres->{msgs}} ];
+	my ($srch, $sres) = @_;
+	my $msgs = delete $sres->{msgs};
+	$srch->retry_reopen(sub { [ map { $_->ensure_metadata; $_ } @$msgs ] });
 }
 
 sub msg_timestamp {
@@ -975,10 +977,11 @@ sub index_topics {
 	my $opts = { offset => $off, limit => 200 };
 
 	$ctx->{order} = [];
-	my $sres = $ctx->{srch}->query('', $opts);
+	my $srch = $ctx->{srch};
+	my $sres = $srch->query('', $opts);
 	my $nr = scalar @{$sres->{msgs}};
 	if ($nr) {
-		$sres = load_results($sres);
+		$sres = load_results($srch, $sres);
 		walk_thread(thread_results($sres), $ctx, *acc_topic);
 	}
 	$ctx->{-next_o} = $off+ $nr;
-- 
EW