X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiImport.pm;h=2d91e4c4aa25e1d428a595712efc1bd6a0cb1256;hb=71eb48442e0f5403e9a5c3f0ccf8257600efe392;hp=9ad2ff12cd396380442aae9e24ab4358e8ca30c0;hpb=e7d13d7bc4a7b1e990602e796b7c2acbddb99a7b;p=public-inbox.git diff --git a/lib/PublicInbox/LeiImport.pm b/lib/PublicInbox/LeiImport.pm index 9ad2ff12..2d91e4c4 100644 --- a/lib/PublicInbox/LeiImport.pm +++ b/lib/PublicInbox/LeiImport.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2021 all contributors +# Copyright (C) all contributors # License: AGPL-3.0+ # front-end for the "lei import" sub-command @@ -6,137 +6,131 @@ package PublicInbox::LeiImport; use strict; use v5.10.1; use parent qw(PublicInbox::IPC PublicInbox::LeiInput); -use PublicInbox::Eml; -use PublicInbox::PktOp qw(pkt_do); +use PublicInbox::InboxWritable qw(eml_from_path); -sub eml_cb { # used by PublicInbox::LeiInput::input_fh +# /^input_/ subs are used by (or override) PublicInbox::LeiInput superclass + +sub input_eml_cb { # used by PublicInbox::LeiInput::input_fh my ($self, $eml, $vmd) = @_; my $xoids = $self->{lei}->{ale}->xoids_for($eml); - $self->{lei}->{sto}->ipc_do('set_eml', $eml, $vmd, $xoids); + if (my $all_vmd = $self->{all_vmd}) { + @$vmd{keys %$all_vmd} = values %$all_vmd; + } + $self->{lei}->{sto}->wq_do('set_eml', $eml, $vmd, $xoids); } -sub mbox_cb { # MboxReader callback used by PublicInbox::LeiInput::input_fh +sub input_mbox_cb { # MboxReader callback my ($eml, $self) = @_; my $vmd; if ($self->{-import_kw}) { my $kw = PublicInbox::MboxReader::mbox_keywords($eml); $vmd = { kw => $kw } if scalar(@$kw); } - eml_cb($self, $eml, $vmd); -} - -sub import_done_wait { # dwaitpid callback - my ($arg, $pid) = @_; - my ($imp, $lei) = @$arg; - $lei->child_error($?, 'non-fatal errors during import') if $?; - my $sto = delete $lei->{sto}; - my $wait = $sto->ipc_do('done') if $sto; # PublicInbox::LeiStore::done - $lei->dclose; + input_eml_cb($self, $eml, $vmd); } -sub import_done { # EOF callback for main daemon - my ($lei) = @_; - my $imp = delete $lei->{imp} or return; - $imp->wq_wait_old(\&import_done_wait, $lei); +sub pmdir_cb { # called via wq_io_do from LeiPmdir->each_mdir_fn + my ($self, $f, $fl) = @_; + my ($folder, $bn) = ($f =~ m!\A(.+?)/(?:new|cur)/([^/]+)\z!) or + die "BUG: $f was not from a Maildir?\n"; + my $kw = PublicInbox::MdirReader::flags2kw($fl); + substr($folder, 0, 0) = 'maildir:'; # add prefix + my $lse = $self->{lse} //= $self->{lei}->{sto}->search; + my $lms = $self->{-lms_rw} //= $self->{lei}->lms; # may be 0 or undef + my @oidbin = $lms ? $lms->name_oidbin($folder, $bn) : (); + @oidbin > 1 and warn("W: $folder/*/$$bn not unique:\n", + map { "\t".unpack('H*', $_)."\n" } @oidbin); + my %seen; + my @docids = sort { $a <=> $b } grep { !$seen{$_}++ } + map { $lse->over->oidbin_exists($_) } @oidbin; + my $vmd = $self->{-import_kw} ? { kw => $kw } : undef; + if (scalar @docids) { + $lse->kw_changed(undef, $kw, \@docids) or return; + } + if (my $eml = eml_from_path($f)) { + $vmd->{sync_info} = [ $folder, \$bn ] if $self->{-mail_sync}; + $self->input_eml_cb($eml, $vmd); + } } -sub net_merge_complete { # callback used by LeiAuth - my ($self) = @_; - for my $input (@{$self->{inputs}}) { - $self->wq_io_do('import_path_url', [], $input); +sub input_net_cb { # imap_each / nntp_each + my ($uri, $uid, $kw, $eml, $self) = @_; + if (defined $eml) { + my $vmd = $self->{-import_kw} ? { kw => $kw } : undef; + $vmd->{sync_info} = [ $$uri, $uid ] if $self->{-mail_sync}; + $self->input_eml_cb($eml, $vmd); + } elsif (my $ikw = $self->{lei}->{ikw}) { # old message, kw only + # we send $uri as a bare SCALAR and not a URIimap ref to + # reduce socket traffic: + $ikw->wq_io_do('ck_update_kw', [], $$uri, $uid, $kw); } - $self->wq_close(1); } -sub import_start { - my ($lei) = @_; - my $self = $lei->{imp}; - $lei->ale; # initialize for workers to read - my $j = $lei->{opt}->{jobs} // scalar(@{$self->{inputs}}) || 1; - if (my $net = $lei->{net}) { +sub do_import_index ($$@) { + my ($self, $lei, @inputs) = @_; + my $sto = $lei->_lei_store(1); + $sto->write_prepare($lei); + $self->{-import_kw} = $lei->{opt}->{kw} // 1; + my $vmd_mod = $self->vmd_mod_extract(\@inputs); + return $lei->fail(join("\n", @{$vmd_mod->{err}})) if $vmd_mod->{err}; + $self->{all_vmd} = $vmd_mod if scalar keys %$vmd_mod; + $lei->ale; # initialize for workers to read (before LeiPmdir->new) + $self->{-mail_sync} = $lei->{opt}->{'mail-sync'} // 1; + $self->prepare_inputs($lei, \@inputs) or return; + + my $j = $lei->{opt}->{jobs} // 0; + $j =~ /\A([0-9]+),[0-9]+\z/ and $j = $1 + 0; + $j ||= scalar(@{$self->{inputs}}) || 1; + my $ikw; + my $net = $lei->{net}; + if ($net) { # $j = $net->net_concurrency($j); TODO + if ($lei->{opt}->{incremental} // 1) { + $net->{incremental} = 1; + $net->{-lms_rw} = $lei->lms // 0; + if ($self->{-import_kw} && $net->{-lms_rw} && + !$lei->{opt}->{'new-only'} && + $net->{imap_order}) { + require PublicInbox::LeiImportKw; + $ikw = PublicInbox::LeiImportKw->new($lei); + $net->{each_old} = 1; + } + } } else { my $nproc = $self->detect_nproc; $j = $nproc if $j > $nproc; } - my $ops = { '' => [ \&import_done, $lei ] }; - $lei->{auth}->op_merge($ops, $self) if $lei->{auth}; - $self->{-wq_nr_workers} = $j // 1; # locked - my $op = $lei->workers_start($self, 'lei_import', undef, $ops); - $self->wq_io_do('import_stdin', []) if $self->{0}; - net_merge_complete($self) unless $lei->{auth}; - while ($op && $op->{sock}) { $op->event_step } + ($lei->{opt}->{'new-only'} && (!$net || !$net->{imap_order})) and + warn "# --new-only is only for IMAP\n"; + $lei->{-eml_noisy} = 1; + $lei->{-err_type} = 'non-fatal'; + $lei->wq1_start($self, $j); } sub lei_import { # the main "lei import" method my ($lei, @inputs) = @_; - my $sto = $lei->_lei_store(1); - $sto->write_prepare($lei); - my $self = $lei->{imp} = bless {}, __PACKAGE__; - $self->{-import_kw} = $lei->{opt}->{kw} // 1; - $self->prepare_inputs($lei, \@inputs) or return; - import_start($lei); -} - -sub ipc_atfork_child { - my ($self) = @_; - my $lei = $self->{lei}; - delete $lei->{imp}; # drop circular ref - $lei->lei_atfork_child; - $self->SUPER::ipc_atfork_child; - $lei->{auth}->do_auth_atfork($self) if $lei->{auth}; - undef; -} - -sub _import_maildir { # maildir_each_eml cb - my ($f, $kw, $eml, $sto, $set_kw) = @_; - $sto->ipc_do('set_eml', $eml, $set_kw ? { kw => $kw }: ()); -} - -sub _import_net { # imap_each, nntp_each cb - my ($url, $uid, $kw, $eml, $sto, $set_kw) = @_; - $sto->ipc_do('set_eml', $eml, $set_kw ? { kw => $kw } : ()); + my $self = bless {}, __PACKAGE__; + do_import_index($self, $lei, @inputs); } -sub import_path_url { - my ($self, $input) = @_; - my $lei = $self->{lei}; - my $ifmt = lc($lei->{opt}->{'in-format'} // ''); - # TODO auto-detect? - if ($input =~ m!\Aimaps?://!i) { - $lei->{net}->imap_each($input, \&_import_net, $lei->{sto}, - $self->{-import_kw}); - return; - } elsif ($input =~ m!\A(?:nntps?|s?news)://!i) { - $lei->{net}->nntp_each($input, \&_import_net, $lei->{sto}, 0); - return; - } elsif ($input =~ s!\A([a-z0-9]+):!!i) { - $ifmt = lc $1; - } - if (-f $input) { - my $m = $lei->{opt}->{'lock'} // ($ifmt eq 'eml' ? ['none'] : - PublicInbox::MboxLock->defaults); - my $mbl = PublicInbox::MboxLock->acq($input, 0, $m); - $self->input_fh($ifmt, $mbl->{fh}, $input); - } elsif (-d _ && (-d "$input/cur" || -d "$input/new")) { - return $lei->fail(<{sto}, $self->{-import_kw}); - } else { - $lei->fail("$input unsupported (TODO)"); +sub _complete_import { + my ($lei, @argv) = @_; + my ($re, $cur, $match_cb) = $lei->complete_url_prepare(\@argv); + my @k = $lei->url_folder_cache->keys($argv[-1] // undef, 1); + my @L = eval { $lei->_lei_store->search->all_terms('L') }; + push(@k, map { "+L:$_" } @L); + my @m = map { $match_cb->($_) } @k; + my %f = map { $_ => 1 } (@m ? @m : @k); + if (my $lms = $lei->lms) { + @k = $lms->folders($argv[-1] // undef, 1); + @m = map { $match_cb->($_) } @k; + if (@m) { @f{@m} = @m } else { @f{@k} = @k } } + keys %f; } -sub import_stdin { - my ($self) = @_; - my $lei = $self->{lei}; - my $in = delete $self->{0}; - $self->input_fh($lei->{opt}->{'in-format'}, $in, ''); -} +no warnings 'once'; +*ipc_atfork_child = \&PublicInbox::LeiInput::input_only_atfork_child; +*net_merge_all_done = \&PublicInbox::LeiInput::input_only_net_merge_all_done; -no warnings 'once'; # the following works even when LeiAuth is lazy-loaded -*net_merge_all = \&PublicInbox::LeiAuth::net_merge_all; 1;