X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiXSearch.pm;h=41e798562d424b097ec611ef5f283f2251d9563b;hb=e85ea061aa504e490bccbaad8d249ce039c0ab53;hp=88540bc22367525b3523711c55e8a71daad31b71;hpb=b319bb7276457d01ab4bc779d867368f884bd522;p=public-inbox.git diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm index 88540bc2..41e79856 100644 --- a/lib/PublicInbox/LeiXSearch.pm +++ b/lib/PublicInbox/LeiXSearch.pm @@ -15,7 +15,9 @@ use PublicInbox::Search qw(xap_terms); use PublicInbox::Spawn qw(popen_rd spawn which); use PublicInbox::MID qw(mids); use PublicInbox::Smsg; +use PublicInbox::AutoReap; use PublicInbox::Eml; +use PublicInbox::LEI; use Fcntl qw(SEEK_SET F_SETFL O_APPEND O_RDWR); use PublicInbox::ContentHash qw(git_sha); use POSIX qw(strftime); @@ -32,7 +34,7 @@ sub new { sub attach_external { my ($self, $ibxish) = @_; # ibxish = ExtSearch or Inbox my $desc = $ibxish->{inboxdir} // $ibxish->{topdir}; - my $srch = $ibxish->search or + my $srch = $ibxish->search // return warn("$desc not indexed for Xapian ($@ $!)\n"); my @shards = $srch->xdb_shards_flat or return warn("$desc has no Xapian shards\n"); @@ -140,11 +142,11 @@ 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}; } } @@ -183,11 +185,10 @@ sub query_one_mset { # for --threads and l2m w/o sort my $maxk = "external.$dir.maxuid"; my $stop_at = $lss ? $lss->{-cfg}->{$maxk} : undef; if (defined $stop_at) { - die "$maxk=$stop_at has multiple values" if ref $stop_at; - my @e; - local $SIG{__WARN__} = sub { push @e, @_ }; - $stop_at += 0; - return warn("$maxk=$stop_at: @e") if @e; + ref($stop_at) and + return warn("$maxk=$stop_at has multiple values\n"); + ($stop_at =~ /[^0-9]/) and + return warn("$maxk=$stop_at not numeric\n"); } my $first_ids; do { @@ -281,23 +282,23 @@ sub each_remote_eml { # callback for MboxReader->mboxrd my $xoids = $lei->{ale}->xoids_for($eml, 1); my $smsg = bless {}, 'PublicInbox::Smsg'; if ($self->{import_sto} && !$xoids) { - my $res = $self->{import_sto}->wq_do('add_eml', $eml); + my ($res, $kw) = $self->{import_sto}->wq_do('add_eml', $eml); if (ref($res) eq ref($smsg)) { # totally new message $smsg = $res; - $smsg->{kw} = []; # short-circuit xsmsg_vmd + $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, '?'); } } @@ -324,7 +325,7 @@ sub fudge_qstr_time ($$$) { } $lr -= ($rft || (48 * 60 * 60)); $lei->qerr("# $uri limiting to ". - strftime('%Y-%m-%d %k:%M', localtime($lr)). ' and newer'); + strftime('%Y-%m-%d %k:%M %z', localtime($lr)). ' and newer'); # this should really be rt: (received-time), but no stable # public-inbox releases support it, yet. my $dt = 'dt:'.strftime('%Y%m%d%H%M%S', gmtime($lr)).'..'; @@ -341,23 +342,22 @@ 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}; my $verbose = $opt->{verbose}; - my ($reap_tail, $reap_curl); + my $reap_tail; my $cerr = File::Temp->new(TEMPLATE => 'curl.err-XXXX', TMPDIR => 1); fcntl($cerr, F_SETFL, O_APPEND|O_RDWR) or warn "set O_APPEND: $!"; - my $rdr = { 2 => $cerr, pgid => 0 }; - my $sigint_reap = $lei->can('sigint_reap'); + my $rdr = { 2 => $cerr }; if ($verbose) { # spawn a process to force line-buffering, otherwise curl # will write 1 character at-a-time and parallel outputs # mmmaaayyy llloookkk llliiikkkeee ttthhhiiisss - my $o = { 1 => $lei->{2}, 2 => $lei->{2}, pgid => 0 }; + my $o = { 1 => $lei->{2}, 2 => $lei->{2} }; my $pid = spawn(['tail', '-f', $cerr->filename], undef, $o); - $reap_tail = PublicInbox::OnDestroy->new($sigint_reap, $pid); + $reap_tail = PublicInbox::AutoReap->new($pid); } my $curl = PublicInbox::LeiCurl->new($lei, $self->{curl}) or return; push @$curl, '-s', '-d', ''; @@ -372,32 +372,30 @@ sub query_remote_mboxrd { my $cmd = $curl->for_uri($lei, $uri); $lei->qerr("# $cmd"); my ($fh, $pid) = popen_rd($cmd, undef, $rdr); - $reap_curl = PublicInbox::OnDestroy->new($sigint_reap, $pid); + my $reap_curl = PublicInbox::AutoReap->new($pid); $fh = IO::Uncompress::Gunzip->new($fh, MultiStream => 1); PublicInbox::MboxReader->mboxrd($fh, \&each_remote_eml, $self, $lei, $each_smsg); - my $err = waitpid($pid, 0) == $pid ? undef - : "BUG: waitpid($cmd): $!"; - @$reap_curl = (); # cancel OnDestroy - die $err if $err; - 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; } - $err = ''; + my $err; if (-s $cerr) { - seek($cerr, 0, SEEK_SET) or - $lei->err("seek($cmd stderr): $!"); + seek($cerr, 0, SEEK_SET) // + warn "seek($cmd stderr): $!"; $err = do { local $/; <$cerr> } // - "read($cmd stderr): $!"; - truncate($cerr, 0) or - $lei->err("truncate($cmd stderr): $!"); + warn "read($cmd stderr): $!"; + truncate($cerr, 0) // warn "truncate($cmd stderr): $!"; } + $err //= ''; next if (($? >> 8) == 22 && $err =~ /\b404\b/); $uri->query_form(q => $qstr); $lei->child_error($?, "E: <$uri> $err"); @@ -411,71 +409,78 @@ 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 $?; + 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 my ($lei) = @_; - my $l2m = delete $lei->{l2m}; - $l2m->wq_wait_old(\&xsearch_done_wait, $lei) if $l2m; - if (my $lxs = delete $lei->{lxs}) { - $lxs->wq_wait_old(\&xsearch_done_wait, $lei); - } - if ($lei->{opt}->{'mail-sync'} && !$lei->{sto}) { - warn "BUG: {sto} missing with --mail-sync"; - } - $lei->sto_done_request if $lei->{sto}; - my $wait = $lei->{v2w} ? $lei->{v2w}->wq_do('done') : undef; - $lei->{ovv}->ovv_end($lei); - my $start_mua; - if ($l2m) { # close() calls LeiToMail reap_compress - if (my $out = delete $lei->{old_1}) { - if (my $mbout = $lei->{1}) { - close($mbout) or return $lei->fail(<<""); -Error closing $lei->{ovv}->{dst}: $! + local $PublicInbox::LEI::current_lei = $lei; + eval { + my $l2m = delete $lei->{l2m}; + delete $lei->{lxs}; + ($lei->{opt}->{'mail-sync'} && !$lei->{sto}) and + warn "BUG: {sto} missing with --mail-sync"; + $lei->sto_done_request if $lei->{sto}; + if (my $v2w = delete $lei->{v2w}) { + my $wait = $v2w->wq_do('done'); # may die + $v2w->wq_close; + } + $lei->{ovv}->ovv_end($lei); + if ($l2m) { # close() calls LeiToMail reap_compress + if (my $out = delete $lei->{old_1}) { + if (my $mbout = $lei->{1}) { + close($mbout) or die <<""; +Error closing $lei->{ovv}->{dst}: \$!=$! \$?=$? + } + $lei->{1} = $out; + } + if ($l2m->lock_free) { + $l2m->poke_dst; + $lei->poke_mua; + } else { # mbox users + delete $l2m->{mbl}; # drop dotlock } - $lei->{1} = $out; - } - if ($l2m->lock_free) { - $l2m->poke_dst; - $lei->poke_mua; - } else { # mbox users - delete $l2m->{mbl}; # drop dotlock - $start_mua = 1; } - } - if ($lei->{-progress}) { - my $tot = $lei->{-mset_total} // 0; - my $nr = $lei->{-nr_write} // 0; - if ($l2m) { - my $m = "# $nr written to " . - "$lei->{ovv}->{dst} ($tot matches)"; - $nr ? $lei->qfin($m) : $lei->qerr($m); - } else { - $lei->qerr("# $tot matches"); + if ($lei->{-progress}) { + my $tot = $lei->{-mset_total} // 0; + my $nr = $lei->{-nr_write} // 0; + if ($l2m) { + my $m = "# $nr written to " . + "$lei->{ovv}->{dst} ($tot matches)"; + $nr ? $lei->qfin($m) : $lei->qerr($m); + } else { + $lei->qerr("# $tot matches"); + } } - } - $lei->start_mua if $start_mua; - $lei->dclose; + $lei->start_mua if $l2m && !$l2m->lock_free; + $lei->dclose; + }; + $lei->fail($@) if $@; } 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; - $lxs->wq_close(0, undef, $lei); + $lxs->wq_kill('-TERM'); + $lxs->wq_close; } $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 } @@ -497,9 +502,10 @@ sub concurrency { $nl + $nr; } -sub start_query ($;$) { # always runs in main (lei-daemon) process - my ($self, $l2m) = @_; - if ($self->{opt_threads} || ($l2m && !$self->{opt_sort})) { +sub start_query ($$) { # always runs in main (lei-daemon) process + my ($self, $lei) = @_; + local $PublicInbox::LEI::current_lei = $lei; + if ($self->{opt_threads} || ($lei->{l2m} && !$self->{opt_sort})) { for my $ibxish (locals($self)) { $self->wq_io_do('query_one_mset', [], $ibxish); } @@ -517,13 +523,14 @@ sub start_query ($;$) { # always runs in main (lei-daemon) process if ($self->{-do_lcat}) { $self->wq_io_do('lcat_dump', []); } - $self->wq_close(1); # lei_xsearch workers stop when done + $self->wq_close; # lei_xsearch workers stop when done } sub incr_start_query { # called whenever an l2m shard starts do_post_auth - my ($self, $l2m) = @_; + my ($self, $lei) = @_; + my $l2m = $lei->{l2m}; return if ++$self->{nr_start_query} != $l2m->{-wq_nr_workers}; - start_query($self, $l2m); + start_query($self, $lei); } sub ipc_atfork_child { @@ -536,18 +543,18 @@ sub do_query { my ($self, $lei) = @_; my $l2m = $lei->{l2m}; my $ops = { - '|' => [ $lei->can('sigpipe_handler'), $lei ], - '!' => [ $lei->can('fail_handler'), $lei ], - '.' => [ \&do_post_augment, $lei ], - '+' => [ \&incr_post_augment, $lei ], + 'sigpipe_handler' => [ $lei ], + 'fail_handler' => [ $lei ], + 'do_post_augment' => [ \&do_post_augment, $lei ], + 'incr_post_augment' => [ \&incr_post_augment, $lei ], '' => [ \&query_done, $lei ], 'mset_progress' => [ \&mset_progress, $lei ], 'l2m_progress' => [ \&l2m_progress, $lei ], - 'x_it' => [ $lei->can('x_it'), $lei ], - 'child_error' => [ $lei->can('child_error'), $lei ], - 'incr_start_query' => [ \&incr_start_query, $self, $l2m ], + 'x_it' => [ $lei ], + '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}; @@ -571,12 +578,14 @@ sub do_query { } $l2m->wq_workers_start('lei2mail', undef, $lei->oldset, { lei => $lei }); + $l2m->wq_wait_async(\&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); my $op_c = delete $lei->{pkt_op_c}; delete $lei->{pkt_op_p}; @$end = (); @@ -586,7 +595,7 @@ sub do_query { if ($l2m) { $l2m->net_merge_all_done($lei) unless $lei->{auth}; } else { - start_query($self); + start_query($self, $lei); } $lei->event_step_init; # wait for shutdowns $lei->wait_wq_events($op_c, $ops);