X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiConvert.pm;h=59af40dea1b11a64ca8e074412c978ab1e98020d;hb=23af251dd607c4e75ab1e68063f2c885c48cc035;hp=32aa2edb0079fbc47f83e8d9c44d4d9313a1446a;hpb=9d1d7c2b505454fba331666a951e0f7997500b3c;p=public-inbox.git diff --git a/lib/PublicInbox/LeiConvert.pm b/lib/PublicInbox/LeiConvert.pm index 32aa2edb..59af40de 100644 --- a/lib/PublicInbox/LeiConvert.pm +++ b/lib/PublicInbox/LeiConvert.pm @@ -1,158 +1,74 @@ -# Copyright (C) 2021 all contributors +# Copyright (C) all contributors # License: AGPL-3.0+ # front-end for the "lei convert" sub-command package PublicInbox::LeiConvert; use strict; use v5.10.1; -use parent qw(PublicInbox::IPC); -use PublicInbox::Eml; -use PublicInbox::InboxWritable qw(eml_from_path); -use PublicInbox::LeiStore; +use parent qw(PublicInbox::IPC PublicInbox::LeiInput); use PublicInbox::LeiOverview; +use PublicInbox::DS; -sub mbox_cb { +# /^input_/ subs are used by PublicInbox::LeiInput + +sub input_mbox_cb { # MboxReader callback my ($eml, $self) = @_; - my @kw = PublicInbox::LeiStore::mbox_keywords($eml); + my $kw = PublicInbox::MboxReader::mbox_keywords($eml); $eml->header_set($_) for qw(Status X-Status); - $self->{wcb}->(undef, { kw => \@kw }, $eml); -} - -sub net_cb { # callback for ->imap_each, ->nntp_each - my (undef, undef, $kw, $eml, $self) = @_; # @_[0,1]: url + uid ignored $self->{wcb}->(undef, { kw => $kw }, $eml); } -sub mdir_cb { - my ($kw, $eml, $self) = @_; - $self->{wcb}->(undef, { kw => $kw }, $eml); +sub input_eml_cb { # used by PublicInbox::LeiInput::input_fh + my ($self, $eml) = @_; + $self->{wcb}->(undef, {}, $eml); } -sub convert_fh ($$$$) { - my ($self, $ifmt, $fh, $name) = @_; - if ($ifmt eq 'eml') { - my $buf = do { local $/; <$fh> } // - return $self->{lei}->child_error(1 << 8, <<""); -error reading $name: $! - - my $eml = PublicInbox::Eml->new(\$buf); - $self->{wcb}->(undef, { kw => [] }, $eml); - } else { - PublicInbox::MboxReader->$ifmt($fh, \&mbox_cb, $self); - } +sub input_maildir_cb { + my (undef, $kw, $eml, $self) = @_; # $_[0] $filename ignored + $self->{wcb}->(undef, { kw => $kw }, $eml); } -sub do_convert { # via wq_do +sub process_inputs { # via wq_do my ($self) = @_; + local $PublicInbox::DS::in_loop = 0; # force synchronous dwaitpid + $self->SUPER::process_inputs; my $lei = $self->{lei}; - my $in_fmt = $lei->{opt}->{'in-format'}; - my $mics; - if (my $stdin = delete $self->{0}) { - convert_fh($self, $in_fmt, $stdin, ''); - } - for my $input (@{$self->{inputs}}) { - my $ifmt = lc($in_fmt // ''); - if ($input =~ m!\Aimaps?://!) { - $lei->{net}->imap_each($input, \&net_cb, $self); - next; - } elsif ($input =~ m!\A(?:nntps?|s?news)://!) { - $lei->{net}->nntp_each($input, \&net_cb, $self); - next; - } elsif ($input =~ s!\A([a-z0-9]+):!!i) { - $ifmt = lc $1; - } - if (-f $input) { - open my $fh, '<', $input or - return $lei->fail("open $input: $!"); - convert_fh($self, $ifmt, $fh, $input); - } elsif (-d _) { - PublicInbox::MdirReader::maildir_each_eml($input, - \&mdir_cb, $self); - } else { - die "BUG: $input unhandled"; # should've failed earlier - } - } delete $lei->{1}; delete $self->{wcb}; # commit + my $nr_w = delete($lei->{-nr_write}) // 0; + my $d = (delete($lei->{-nr_seen}) // 0) - $nr_w; + $d = $d ? " ($d duplicates)" : ''; + $lei->qerr("# converted $nr_w messages$d"); } -sub call { # the main "lei convert" method - my ($cls, $lei, @inputs) = @_; - my $opt = $lei->{opt}; - $opt->{kw} //= 1; - my $self = $lei->{cnv} = bless {}, $cls; - my $in_fmt = $opt->{'in-format'}; - my (@f, @d); - $opt->{dedupe} //= 'none'; +sub lei_convert { # the main "lei convert" method + my ($lei, @inputs) = @_; + $lei->{opt}->{kw} //= 1; + $lei->{opt}->{dedupe} //= 'none'; + my $self = bless {}, __PACKAGE__; my $ovv = PublicInbox::LeiOverview->new($lei, 'out-format'); $lei->{l2m} or return - $lei->fail("output not specified or is not a mail destination"); - my $net = $lei->{net}; # NetWriter may be created by l2m - $opt->{augment} = 1 unless $ovv->{dst} eq '/dev/stdout'; - if ($opt->{stdin}) { - @inputs and return $lei->fail("--stdin and @inputs do not mix"); - $lei->check_input_format(undef, 'in-format') or return; - $self->{0} = $lei->{0}; - } - # e.g. Maildir:/home/user/Mail/ or imaps://example.com/INBOX - for my $input (@inputs) { - my $input_path = $input; - if ($input =~ m!\A(?:imaps?|nntps?|s?news)://!i) { - require PublicInbox::NetReader; - $net //= PublicInbox::NetReader->new; - $net->add_url($input); - } elsif ($input_path =~ s/\A([a-z0-9]+)://is) { - my $ifmt = lc $1; - if (($in_fmt // $ifmt) ne $ifmt) { - return $lei->fail(<<""); ---in-format=$in_fmt and `$ifmt:' conflict - - } - if (-f $input_path) { - require PublicInbox::MboxReader; - PublicInbox::MboxReader->can($ifmt) or return - $lei->fail("$ifmt not supported"); - } elsif (-d _) { - require PublicInbox::MdirReader; - $ifmt eq 'maildir' or return - $lei->fail("$ifmt not supported"); - } else { - return $lei->fail("Unable to handle $input"); - } - } elsif (-f $input) { push @f, $input } - elsif (-d _) { push @d, $input } - else { return $lei->fail("Unable to handle $input") } - } - if (@f) { $lei->check_input_format(\@f, 'in-format') or return } - if (@d) { # TODO: check for MH vs Maildir, here - require PublicInbox::MdirReader; - } - $self->{inputs} = \@inputs; - if ($net) { - if (my $err = $net->errors) { - return $lei->fail($err); - } - $net->{quiet} = $opt->{quiet}; - $lei->{net} //= $net; - } - my $op = $lei->workers_start($self, 'lei_convert', 1, { - '' => [ $lei->can('dclose'), $lei ] - }); - $self->wq_io_do('do_convert', []); - $self->wq_close(1); - while ($op && $op->{sock}) { $op->event_step } + $lei->fail('--output unspecified or is not a mail destination'); + my $devfd = $lei->path_to_fd($ovv->{dst}) // return; + $lei->{opt}->{augment} = 1 if $devfd < 0; + $self->prepare_inputs($lei, \@inputs) or return; + # n.b. {net} {auth} is handled by l2m worker + my ($op_c, $ops) = $lei->workers_start($self, 1); + $lei->{wq1} = $self; + $self->wq_io_do('process_inputs', []); + $self->wq_close; + $lei->wait_wq_events($op_c, $ops); } sub ipc_atfork_child { my ($self) = @_; my $lei = $self->{lei}; - $lei->lei_atfork_child; + $lei->_lei_atfork_child; my $l2m = delete $lei->{l2m}; if (my $net = $lei->{net}) { # may prompt user once $net->{mics_cached} = $net->imap_common_init($lei); $net->{nn_cached} = $net->nntp_common_init($lei); } - $SIG{__WARN__} = PublicInbox::Eml::warn_ignore_cb(); $l2m->pre_augment($lei); $l2m->do_augment($lei); $l2m->post_augment($lei);