X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiXSearch.pm;h=9f7f3885a3a6fce2a120bb23416d109495c324ad;hb=57fed2e4b78ed394;hp=b3fd79d005c9ad00954c6d6e9b5edfce4e653c3d;hpb=63e2751ace2dd35a661620a9ad30d6c92948a009;p=public-inbox.git diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm index b3fd79d0..9f7f3885 100644 --- a/lib/PublicInbox/LeiXSearch.pm +++ b/lib/PublicInbox/LeiXSearch.pm @@ -9,7 +9,6 @@ use strict; use v5.10.1; use parent qw(PublicInbox::LeiSearch PublicInbox::IPC); use PublicInbox::DS qw(now); -use PublicInbox::PktOp qw(pkt_do); use File::Temp 0.19 (); # 0.19 for ->newdir use File::Spec (); use PublicInbox::Search qw(xap_terms); @@ -33,7 +32,7 @@ sub attach_external { my ($self, $ibxish) = @_; # ibxish = ExtSearch or Inbox my $desc = $ibxish->{inboxdir} // $ibxish->{topdir}; my $srch = $ibxish->search or - return warn("$desc not indexed for Xapian\n"); + return warn("$desc not indexed for Xapian ($@ $!)\n"); my @shards = $srch->xdb_shards_flat or return warn("$desc has no Xapian shards\n"); @@ -67,15 +66,21 @@ sub remotes { @{$_[0]->{remotes} // []} } # called by PublicInbox::Search::xdb (usually via ->mset) sub xdb_shards_flat { @{$_[0]->{shards_flat} // []} } -sub mitem_kw ($$;$) { - my ($smsg, $mitem, $flagged) = @_; - my $kw = xap_terms('K', my $doc = $mitem->get_document); +sub _mitem_kw { # retry_reopen callback + my ($srch, $smsg, $mitem, $flagged) = @_; + my $doc = $mitem->get_document; + my $kw = xap_terms('K', $doc); $kw->{flagged} = 1 if $flagged; + my @L = xap_terms('L', $doc); # we keep the empty {kw} array here to prevent expensive work in # ->xsmsg_vmd, _unbless_smsg will clobber it iff it's empty $smsg->{kw} = [ sort keys %$kw ]; - my $L = xap_terms('L', $doc); - $smsg->{L} = [ sort keys %$L ] if scalar(keys %$L); + $smsg->{L} = \@L if scalar(@L); +} + +sub mitem_kw ($$$;$) { + my ($srch, $smsg, $mitem, $flagged) = @_; + $srch->retry_reopen(\&_mitem_kw, $smsg, $mitem, $flagged); } # like over->get_art @@ -90,7 +95,7 @@ sub smsg_for { my $smsg = $ibx->over->get_art($num); return if $smsg->{bytes} == 0; # external message if ($ibx->can('msg_keywords')) { - mitem_kw($smsg, $mitem); + mitem_kw($self, $smsg, $mitem); } $smsg; } @@ -136,14 +141,19 @@ sub mset_progress { my $lei = shift; return if $lei->{early_mua} || !$lei->{-progress}; if ($lei->{pkt_op_p}) { - pkt_do($lei->{pkt_op_p}, 'mset_progress', @_); + $lei->{pkt_op_p}->pkt_do('mset_progress', @_); } else { # single lei-daemon consumer my ($desc, $mset_size, $mset_total_est) = @_; - $lei->{-mset_total} += $mset_size; + $lei->{-mset_total} += $mset_size if $mset_total_est ne '?'; $lei->qerr("# $desc $mset_size/$mset_total_est"); } } +sub l2m_progress { + my ($lei, $nr) = @_; + $lei->{-nr_write} += $nr; +} + sub query_one_mset { # for --threads and l2m w/o sort my ($self, $ibxish) = @_; local $0 = "$0 query_one_mset"; @@ -151,6 +161,7 @@ sub query_one_mset { # for --threads and l2m w/o sort my ($srch, $over) = ($ibxish->search, $ibxish->over); my $dir = $ibxish->{inboxdir} // $ibxish->{topdir}; return warn("$dir not indexed by Xapian\n") unless ($srch && $over); + bless $srch, 'PublicInbox::LeiSearch'; # for ->qparse_new my $mo = { %{$lei->{mset_opt}} }; # copy my $mset; my $each_smsg = $lei->{ovv}->ovv_each_smsg_cb($lei); @@ -189,7 +200,8 @@ sub query_one_mset { # for --threads and l2m w/o sort my $mitem = delete $n2item{$n}; next if $smsg->{bytes} == 0; if ($mitem && $can_kw) { - mitem_kw($smsg, $mitem, $fl); + mitem_kw($srch, $smsg, $mitem, + $fl); } elsif ($mitem && $fl) { # call ->xsmsg_vmd, later $smsg->{lei_q_tt_flagged} = 1; @@ -205,7 +217,7 @@ sub query_one_mset { # for --threads and l2m w/o sort my $mitem = $items[$i++]; my $smsg = $over->get_art($n) or next; next if $smsg->{bytes} == 0; - mitem_kw($smsg, $mitem, $fl) if $can_kw; + mitem_kw($srch, $smsg, $mitem, $fl) if $can_kw; $each_smsg->($smsg, $mitem); } } @@ -233,7 +245,7 @@ sub query_combined_mset { # non-parallel for non-"--threads" users do { $mset = $self->mset($mo->{qstr}, $mo); mset_progress($lei, 'xsearch', $mset->size, - $mset->size, $mset->get_matches_estimated); + $mset->get_matches_estimated); wait_startq($lei); # wait for keyword updates for my $mitem ($mset->items) { my $smsg = smsg_for($self, $mitem) or next; @@ -244,19 +256,27 @@ sub query_combined_mset { # non-parallel for non-"--threads" users $lei->{ovv}->ovv_atexit_child($lei); } +sub _smsg_fill ($$) { + my ($smsg, $eml) = @_; + $smsg->populate($eml); + $smsg->parse_references($eml, mids($eml)); + $smsg->{$_} //= '' for qw(from to cc ds subject references mid); + delete @$smsg{qw(From Subject -ds -ts)}; +} + sub each_remote_eml { # callback for MboxReader->mboxrd my ($eml, $self, $lei, $each_smsg) = @_; my $xoids = $lei->{ale}->xoids_for($eml, 1); + my $smsg = bless {}, 'PublicInbox::Smsg'; if ($self->{import_sto} && !$xoids) { - $self->{import_sto}->ipc_do('add_eml', $eml); + my $res = $self->{import_sto}->ipc_do('add_eml', $eml); + if (ref($res) eq ref($smsg)) { # totally new message + $smsg = $res; + $smsg->{kw} = []; # short-circuit xsmsg_vmd + } } - my $smsg = bless {}, 'PublicInbox::Smsg'; - $smsg->{blob} = $xoids ? (keys(%$xoids))[0] - : git_sha(1, $eml)->hexdigest; - $smsg->populate($eml); - $smsg->parse_references($eml, mids($eml)); - $smsg->{$_} //= '' for qw(from to cc ds subject references mid); - delete @$smsg{qw(From Subject -ds -ts)}; + $smsg->{blob} //= $xoids ? (keys(%$xoids))[0] : $lei->git_blob_id($eml); + _smsg_fill($smsg, $eml); wait_startq($lei); if ($lei->{-progress}) { ++$lei->{-nr_remote_eml}; @@ -277,7 +297,9 @@ sub query_remote_mboxrd { local $SIG{TERM} = sub { exit(0) }; # for DESTROY (File::Temp, $reap) my $lei = $self->{lei}; my $opt = $lei->{opt}; - my @qform = (q => $lei->{mset_opt}->{qstr}, x => 'm'); + my $qstr = $lei->{mset_opt}->{qstr}; + $qstr =~ s/[ \n\t]+/ /sg; # make URLs less ugly + my @qform = (q => $qstr, x => 'm'); push(@qform, t => 1) if $opt->{threads}; my $verbose = $opt->{verbose}; my ($reap_tail, $reap_curl); @@ -352,11 +374,14 @@ sub query_done { # EOF callback for main daemon if (my $lxs = delete $lei->{lxs}) { $lxs->wq_wait_old(\&xsearch_done_wait, $lei); } - my $wait = $lei->{sto} ? $lei->{sto}->ipc_do('done') : undef; + if ($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}->ipc_do('done') : undef; $lei->{ovv}->ovv_end($lei); - my @out; + my $start_mua; if ($l2m) { # close() calls LeiToMail reap_compress - @out = (" in $lei->{ovv}->{dst}"); if (my $out = delete $lei->{old_1}) { if (my $mbout = $lei->{1}) { close($mbout) or return $lei->fail(<<""); @@ -370,17 +395,28 @@ Error closing $lei->{ovv}->{dst}: $! $lei->poke_mua; } else { # mbox users delete $l2m->{mbl}; # drop dotlock - $lei->start_mua; + $start_mua = 1; } } - $lei->{-progress} and - $lei->err('# ', $lei->{-mset_total} // 0, " matches", @out); + if ($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; $lei->dclose; } sub do_post_augment { my ($lei) = @_; my $l2m = $lei->{l2m} or return; # client disconnected + $lei->fchdir or return; my $err; eval { $l2m->post_augment($lei) }; $err = $@; @@ -431,6 +467,9 @@ sub start_query ($;$) { # always runs in main (lei-daemon) process for my $uris (@$q) { $self->wq_io_do('query_remote_mboxrd', [], $uris); } + if ($self->{-do_lcat}) { + $self->wq_io_do('lcat_dump', []); + } $self->wq_close(1); # lei_xsearch workers stop when done } @@ -456,6 +495,7 @@ sub do_query { '+' => [ \&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, $l2m ], @@ -471,11 +511,22 @@ sub do_query { if ($lei->{opt}->{augment} && delete $lei->{early_mua}) { $lei->start_mua; } + my $F_SETPIPE_SZ = $^O eq 'linux' ? 1031 : undef; + if ($l2m->{-wq_nr_workers} > 1 && + $l2m->{base_type} =~ /\A(?:maildir|mbox)\z/) { + # setup two barriers to coordinate ->has_entries + # between l2m workers + pipe(my ($a_r, $a_w)) or die "pipe: $!"; + fcntl($a_r, $F_SETPIPE_SZ, 4096) if $F_SETPIPE_SZ; + pipe(my ($b_r, $b_w)) or die "pipe: $!"; + fcntl($b_r, $F_SETPIPE_SZ, 4096) if $F_SETPIPE_SZ; + $l2m->{au_peers} = [ $a_r, $a_w, $b_r, $b_w ]; + } $l2m->wq_workers_start('lei2mail', undef, $lei->oldset, { lei => $lei }); pipe($lei->{startq}, $lei->{au_done}) or die "pipe: $!"; - # 1031: F_SETPIPE_SZ - fcntl($lei->{startq}, 1031, 4096) if $^O eq 'linux'; + fcntl($lei->{startq}, $F_SETPIPE_SZ, 4096) if $F_SETPIPE_SZ; + delete $l2m->{au_peers}; } $self->wq_workers_start('lei_xsearch', undef, $lei->oldset, { lei => $lei }); @@ -484,13 +535,14 @@ sub do_query { @$end = (); $self->{opt_threads} = $lei->{opt}->{threads}; $self->{opt_sort} = $lei->{opt}->{'sort'}; + $self->{-do_lcat} = $lei->{lcat_blob} // $lei->{lcat_fid}; if ($l2m) { - $l2m->net_merge_complete unless $lei->{auth}; + $l2m->net_merge_all_done unless $lei->{auth}; } else { start_query($self); } $lei->event_step_init; # wait for shutdowns - $op_c->op_wait_event($ops); + $lei->wait_wq_events($op_c, $ops); } sub add_uri { @@ -527,5 +579,48 @@ sub prepare_external { push @{$self->{locals}}, $loc; } +sub _lcat_i { # LeiMailSync->each_src iterator callback + my ($oidbin, $id, $each_smsg) = @_; + $each_smsg->({blob => unpack('H*', $oidbin), pct => 100}); +} + +sub _lcat2smsg { # git->cat_async callback + my ($bref, $oid, $type, $size, $smsg) = @_; + if ($bref) { + my $eml = PublicInbox::Eml->new($bref); + my $json_dump = delete $smsg->{-json_dump}; + bless $smsg, 'PublicInbox::Smsg'; + _smsg_fill($smsg, $eml); + $json_dump->($smsg, undef, $eml); + } +} + +sub lcat_dump { + my ($self) = @_; + my $lei = $self->{lei}; + my $each_smsg = $lei->{ovv}->ovv_each_smsg_cb($lei); + my $git = $lei->{ale}->git; + if (!$lei->{l2m}) { + my $json_dump = $each_smsg; + $each_smsg = sub { + my ($smsg) = @_; + use Data::Dumper; + $smsg->{-json_dump} = $json_dump; + $git->cat_async($smsg->{blob}, \&_lcat2smsg, $smsg); + }; + } + for my $oid (@{$lei->{lcat_blob} // []}) { + $each_smsg->({ blob => $oid, pct => 100 }); + } + if (my $fids = delete $lei->{lcat_fid}) { + my $lms = $lei->{lse}->lms; + for my $fid (@$fids) { + $lms->each_src({fid => $fid}, \&_lcat_i, $each_smsg); + } + } + $git->async_wait_all; + undef $each_smsg; # may commit + $lei->{ovv}->ovv_atexit_child($lei); +} 1;