imap+nntp: share COMPRESS implementation

[public-inbox.git] / lib / PublicInbox / LeiConvert.pm

diff --git a/lib/PublicInbox/LeiConvert.pm b/lib/PublicInbox/LeiConvert.pm
index 0aa1322997fb8381ae34c132b6f1ade02b0f598f..59af40dea1b11a64ca8e074412c978ab1e98020d 100644 (file)
--- a/lib/PublicInbox/LeiConvert.pm
+++ b/lib/PublicInbox/LeiConvert.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2021 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # front-end for the "lei convert" sub-command
@@ -6,105 +6,69 @@ package PublicInbox::LeiConvert;
 use strict;
 use v5.10.1;
 use parent qw(PublicInbox::IPC PublicInbox::LeiInput);
-use PublicInbox::Eml;
-use PublicInbox::LeiStore;
 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 $kw = PublicInbox::MboxReader::mbox_keywords($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 }, $eml);
+sub input_eml_cb { # used by PublicInbox::LeiInput::input_fh
+       my ($self, $eml) = @_;
+       $self->{wcb}->(undef, {}, $eml);
 }
 
-sub mdir_cb {
-       my ($f, $kw, $eml, $self) = @_;
+sub input_maildir_cb {
+       my (undef, $kw, $eml, $self) = @_; # $_[0] $filename ignored
        $self->{wcb}->(undef, { kw => $kw }, $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, $self);
-       }
-}
-
-sub do_convert { # via wq_do
+sub process_inputs { # via wq_do
        my ($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, '<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) {
-                       my $m = $lei->{opt}->{'lock'} //
-                                       ($ifmt eq 'eml' ? ['none'] :
-                                       PublicInbox::MboxLock->defaults);
-                       my $mbl = PublicInbox::MboxLock->acq($input, 0, $m);
-                       convert_fh($self, $ifmt, $mbl->{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 lei_convert { # the main "lei convert" method
        my ($lei, @inputs) = @_;
        $lei->{opt}->{kw} //= 1;
        $lei->{opt}->{dedupe} //= 'none';
-       my $self = $lei->{cnv} = bless {}, __PACKAGE__;
+       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");
-       $lei->{opt}->{augment} = 1 unless $ovv->{dst} eq '/dev/stdout';
+               $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;
-       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}) { $op->event_step }
+       # n.b. {net} {auth} is handled by l2m worker
+       my ($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);