X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiXSearch.pm;h=fba168613d96a2cdec8f8baec1d350e8dff76d40;hb=4b5a1b5787edee2a3b6cc10a3ccc5721f1414268;hp=99f887ce730f50f21c45f742576164c528217415;hpb=9a32c4433b87eddc0f07e979faf64e0f8e025593;p=public-inbox.git diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm index 99f887ce..fba16861 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"); @@ -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 { @@ -346,18 +347,17 @@ sub query_remote_mboxrd { 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', ''; @@ -372,16 +372,13 @@ 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}; + $reap_curl->join; if ($? == 0) { # don't update if no results, maybe MTA is down $key && $nr and @@ -389,15 +386,15 @@ sub query_remote_mboxrd { mset_progress($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> $err"); @@ -416,14 +413,14 @@ sub xsearch_done_wait { # dwaitpid callback sub query_done { # EOF callback for main daemon my ($lei) = @_; + local $PublicInbox::LEI::current_lei = $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'} && !$lei->{sto}) { + ($lei->{opt}->{'mail-sync'} && !$lei->{sto}) and 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); @@ -462,6 +459,7 @@ Error closing $lei->{ovv}->{dst}: $! sub do_post_augment { my ($lei) = @_; + local $PublicInbox::LEI::current_lei = $lei; my $l2m = $lei->{l2m} or return; # client disconnected $lei->fchdir or return; my $err; @@ -469,7 +467,7 @@ sub do_post_augment { $err = $@; if ($err) { if (my $lxs = delete $lei->{lxs}) { - $lxs->wq_kill; + $lxs->wq_kill('-TERM'); $lxs->wq_close(0, undef, $lei); } $lei->fail("$err"); @@ -497,9 +495,10 @@ 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', [], $ibxish); } @@ -521,9 +520,10 @@ sub start_query ($;$) { # always runs in main (lei-daemon) process } sub incr_start_query { # called whenever an l2m shard starts do_post_auth - my ($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, $lei); } sub ipc_atfork_child { @@ -536,16 +536,16 @@ sub do_query { my ($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}; my $end = $lei->pkt_op_pair; @@ -586,7 +586,7 @@ sub do_query { if ($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);