X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiXSearch.pm;h=5965274ca211396df49e263ec6ade1508e3ba458;hb=7b654d175cf2e31b4354929ea678563f534947e5;hp=4aa2a81c00256698b87241bfe7c19578210aad4e;hpb=299b40d252cf4d4db6fa29ad18cb78777f1f55fc;p=public-inbox.git

diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm
index 4aa2a81c..5965274c 100644
--- a/lib/PublicInbox/LeiXSearch.pm
+++ b/lib/PublicInbox/LeiXSearch.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2020-2021 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Combine any combination of PublicInbox::Search,
@@ -103,13 +103,6 @@ sub smsg_for {
 	$smsg;
 }
 
-sub recent {
-	my ($self, $qstr, $opt) = @_;
-	$opt //= {};
-	$opt->{relevance} //= -2;
-	$self->mset($qstr //= 'z:1..', $opt);
-}
-
 sub over {}
 
 sub _check_mset_limit ($$$) {
@@ -142,11 +135,11 @@ sub wait_startq ($) {
 				delete $lei->{opt}->{verbose};
 				delete $lei->{-progress};
 			} else {
-				$lei->fail("$$ WTF `$do_augment_done'");
+				die "BUG: do_augment_done=`$do_augment_done'";
 			}
 			return;
 		}
-		return $lei->fail("$$ wait_startq: $!") unless $!{EINTR};
+		die "wait_startq: $!" unless $!{EINTR};
 	}
 }
 
@@ -163,8 +156,9 @@ sub mset_progress {
 }
 
 sub l2m_progress {
-	my ($lei, $nr) = @_;
-	$lei->{-nr_write} += $nr;
+	my ($lei, $nr_write, $nr_seen) = @_;
+	$lei->{-nr_write} += $nr_write;
+	$lei->{-nr_seen} += $nr_seen;
 }
 
 sub query_one_mset { # for --threads and l2m w/o sort
@@ -283,20 +277,22 @@ sub each_remote_eml { # callback for MboxReader->mboxrd
 	my $smsg = bless {}, 'PublicInbox::Smsg';
 	if ($self->{import_sto} && !$xoids) {
 		my ($res, $kw) = $self->{import_sto}->wq_do('add_eml', $eml);
-		$smsg = $res if ref($res) eq ref($smsg); # totally new message
+		if (ref($res) eq ref($smsg)) { # totally new message
+			$smsg = $res;
+			$self->{-sto_imported} = 1;
+		}
 		$smsg->{kw} = $kw; # short-circuit xsmsg_vmd
 	}
 	$smsg->{blob} //= $xoids ? (keys(%$xoids))[0]
 				: $lei->git_oid($eml)->hexdigest;
 	_smsg_fill($smsg, $eml);
 	wait_startq($lei);
+	my $nr = ++$lei->{-nr_remote_eml}; # needed for lss->cfg_set
 	if ($lei->{-progress}) {
-		++$lei->{-nr_remote_eml};
 		my $now = now();
 		my $next = $lei->{-next_progress} //= ($now + 1);
 		if ($now > $next) {
 			$lei->{-next_progress} = $now + 1;
-			my $nr = $lei->{-nr_remote_eml};
 			mset_progress($lei, $lei->{-current_url}, $nr, '?');
 		}
 	}
@@ -340,7 +336,7 @@ sub query_remote_mboxrd {
 	local $SIG{TERM} = sub { exit(0) }; # for DESTROY (File::Temp, $reap)
 	my $lei = $self->{lei};
 	my $opt = $lei->{opt};
-	my $qstr = $lei->{mset_opt}->{qstr};
+	chomp(my $qstr = $lei->{mset_opt}->{qstr});
 	$qstr =~ s/[ \n\t]+/ /sg; # make URLs less ugly
 	my @qform = (x => 'm');
 	push(@qform, t => 1) if $opt->{threads};
@@ -374,13 +370,14 @@ sub query_remote_mboxrd {
 		$fh = IO::Uncompress::Gunzip->new($fh, MultiStream => 1);
 		PublicInbox::MboxReader->mboxrd($fh, \&each_remote_eml, $self,
 						$lei, $each_smsg);
-		my $nr = $lei->{-nr_remote_eml};
-		my $wait = $lei->{sto}->wq_do('done') if $nr && $lei->{sto};
+		if (delete($self->{-sto_imported})) {
+			my $wait = $self->{import_sto}->wq_do('done');
+		}
 		$reap_curl->join;
 		if ($? == 0) {
 			# don't update if no results, maybe MTA is down
-			$key && $nr and
-				$lei->{lss}->cfg_set($key, $start);
+			my $nr = $lei->{-nr_remote_eml};
+			$lei->{lss}->cfg_set($key, $start) if $key && $nr;
 			mset_progress($lei, $lei->{-current_url}, $nr, $nr);
 			next;
 		}
@@ -403,10 +400,12 @@ sub query_remote_mboxrd {
 
 sub git { $_[0]->{git} // die 'BUG: git uninitialized' }
 
-sub xsearch_done_wait { # dwaitpid callback
-	my ($arg, $pid) = @_;
-	my ($wq, $lei) = @$arg;
-	$lei->child_error($?, 'non-fatal error from '.ref($wq)) if $?;
+sub xsearch_done_wait { # awaitpid cb
+	my ($pid, $wq, $lei) = @_;
+	return if !$?;
+	my $s = $? & 127;
+	return $lei->child_error($?) if $s == 13 || $s == 15;
+	$lei->child_error($?, 'non-fatal error from '.ref($wq)." \$?=$?");
 }
 
 sub query_done { # EOF callback for main daemon
@@ -441,13 +440,16 @@ Error closing $lei->{ovv}->{dst}: \$!=$! \$?=$?
 		}
 		if ($lei->{-progress}) {
 			my $tot = $lei->{-mset_total} // 0;
-			my $nr = $lei->{-nr_write} // 0;
+			my $nr_w = $lei->{-nr_write} // 0;
+			my $d = ($lei->{-nr_seen} // 0) - $nr_w;
+			my $x = "$tot matches";
+			$x .= ", $d duplicates" if $d;
 			if ($l2m) {
-				my $m = "# $nr written to " .
-					"$lei->{ovv}->{dst} ($tot matches)";
-				$nr ? $lei->qfin($m) : $lei->qerr($m);
+				my $m = "# $nr_w written to " .
+					"$lei->{ovv}->{dst} ($x)";
+				$nr_w ? $lei->qfin($m) : $lei->qerr($m);
 			} else {
-				$lei->qerr("# $tot matches");
+				$lei->qerr("# $x");
 			}
 		}
 		$lei->start_mua if $l2m && !$l2m->lock_free;
@@ -460,10 +462,11 @@ sub do_post_augment {
 	my ($lei) = @_;
 	local $PublicInbox::LEI::current_lei = $lei;
 	my $l2m = $lei->{l2m} or return; # client disconnected
-	$lei->fchdir or return;
-	my $err;
-	eval { $l2m->post_augment($lei) };
-	$err = $@;
+	eval {
+		$lei->fchdir;
+		$l2m->post_augment($lei);
+	};
+	my $err = $@;
 	if ($err) {
 		if (my $lxs = delete $lei->{lxs}) {
 			$lxs->wq_kill('-TERM');
@@ -472,7 +475,8 @@ sub do_post_augment {
 		$lei->fail("$err");
 	}
 	if (!$err && delete $lei->{early_mua}) { # non-augment case
-		$lei->start_mua;
+		eval { $lei->start_mua };
+		$lei->fail($@) if $@;
 	}
 	close(delete $lei->{au_done}); # triggers wait_startq in lei_xsearch
 }
@@ -546,7 +550,7 @@ sub do_query {
 		'child_error' => [ $lei ],
 		'incr_start_query' => [ $self, $lei ],
 	};
-	$lei->{auth}->op_merge($ops, $l2m) if $l2m && $lei->{auth};
+	$lei->{auth}->op_merge($ops, $l2m, $lei) if $l2m && $lei->{auth};
 	my $end = $lei->pkt_op_pair;
 	$lei->{1}->autoflush(1);
 	$lei->start_pager if delete $lei->{need_pager};
@@ -569,15 +573,15 @@ sub do_query {
 			$l2m->{au_peers} = [ $a_r, $a_w, $b_r, $b_w ];
 		}
 		$l2m->wq_workers_start('lei2mail', undef,
-					$lei->oldset, { lei => $lei });
-		$l2m->wq_wait_async(\&xsearch_done_wait, $lei);
+					$lei->oldset, { lei => $lei },
+					\&xsearch_done_wait, $lei);
 		pipe($lei->{startq}, $lei->{au_done}) or die "pipe: $!";
 		fcntl($lei->{startq}, $F_SETPIPE_SZ, 4096) if $F_SETPIPE_SZ;
 		delete $l2m->{au_peers};
 	}
 	$self->wq_workers_start('lei_xsearch', undef,
-				$lei->oldset, { lei => $lei });
-	$self->wq_wait_async(\&xsearch_done_wait, $lei);
+				$lei->oldset, { lei => $lei },
+				\&xsearch_done_wait, $lei);
 	my $op_c = delete $lei->{pkt_op_c};
 	delete $lei->{pkt_op_p};
 	@$end = ();
@@ -600,34 +604,40 @@ sub add_uri {
 		require IO::Uncompress::Gunzip;
 		require PublicInbox::LeiCurl;
 		push @{$self->{remotes}}, $uri;
+		$uri;
 	} else {
 		warn "curl missing, ignoring $uri\n";
+		undef;
 	}
 }
 
+# returns URI or PublicInbox::Inbox-like object
 sub prepare_external {
 	my ($self, $loc, $boost) = @_; # n.b. already ordered by boost
 	if (ref $loc) { # already a URI, or PublicInbox::Inbox-like object
 		return add_uri($self, $loc) if $loc->can('scheme');
+		# fall-through on Inbox-like objects
 	} elsif ($loc =~ m!\Ahttps?://!) {
 		require URI;
 		return add_uri($self, URI->new($loc));
-	} elsif (-f "$loc/ei.lock") {
+	} elsif (-f "$loc/ei.lock" && -d "$loc/ALL.git/objects") {
 		require PublicInbox::ExtSearch;
 		die "`\\n' not allowed in `$loc'\n" if index($loc, "\n") >= 0;
 		$loc = PublicInbox::ExtSearch->new($loc);
-	} elsif (-f "$loc/inbox.lock" || -d "$loc/public-inbox") {
+	} elsif ((-f "$loc/inbox.lock" && -d "$loc/all.git/objects") ||
+			(-d "$loc/public-inbox" && -d "$loc/objects")) {
 		die "`\\n' not allowed in `$loc'\n" if index($loc, "\n") >= 0;
 		require PublicInbox::Inbox; # v2, v1
 		$loc = bless { inboxdir => $loc }, 'PublicInbox::Inbox';
 	} elsif (!-e $loc) {
 		warn "W: $loc gone, perhaps run: lei forget-external $loc\n";
-		return;
+		return undef;
 	} else {
 		warn "W: $loc ignored, unable to determine external type\n";
-		return;
+		return undef;
 	}
 	push @{$self->{locals}}, $loc;
+	$loc;
 }
 
 sub _lcat_i { # LeiMailSync->each_src iterator callback