X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FNetReader.pm;h=c1af03a3f8b972db9a8ff38f0b081652dbb90a53;hb=961690bae47c90a4a6960952587c6f4463fb4b19;hp=e305523e300b16c2b32b59442ecadf44be683c02;hpb=20c940a876728fe91892200dd874dd917cd677ac;p=public-inbox.git diff --git a/lib/PublicInbox/NetReader.pm b/lib/PublicInbox/NetReader.pm index e305523e..c1af03a3 100644 --- a/lib/PublicInbox/NetReader.pm +++ b/lib/PublicInbox/NetReader.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2021 all contributors +# Copyright (C) all contributors # License: AGPL-3.0+ # common reader code for IMAP and NNTP (and maybe JMAP) @@ -328,7 +328,7 @@ sub imap_common_init ($;$) { my $sec = uri_section($uri); # knobs directly for Mail::IMAPClient->new - for my $k (qw(Starttls Debug Compress Ignoresizeerrors)) { + for my $k (qw(Starttls Debug Compress)) { my $bool = cfg_bool($cfg, "imap.$k", $$uri) // next; $mic_common->{$sec}->{$k} = $bool; } @@ -344,10 +344,10 @@ sub imap_common_init ($;$) { } my $k = 'imap.fetchBatchSize'; my $bs = $cfg->urlmatch($k, $$uri) // next; - if ($bs =~ /\A([0-9]+)\z/) { + if ($bs =~ /\A([0-9]+)\z/ && $bs > 0) { $self->{cfg_opt}->{$sec}->{batch_size} = $bs; } else { - warn "$k=$bs is not an integer\n"; + warn "$k=$bs is not a positive integer\n"; } } # make sure we can connect and cache the credentials in memory @@ -359,6 +359,7 @@ sub imap_common_init ($;$) { mic_for($self, $uri, $mic_common, $lei) // die "Unable to continue\n"; next unless $self->isa('PublicInbox::NetWriter'); + next if $self->{-skip_creat}; my $dst = $orig_uri->mailbox // next; next if $mic->exists($dst); # already exists $mic->create($dst) or die "CREATE $dst failed <$orig_uri>: $@"; @@ -480,7 +481,7 @@ sub itrk_last ($$;$$) { my ($self, $uri, $r_uidval, $mic) = @_; return (undef, undef, $r_uidval) unless $self->{incremental}; my ($itrk, $l_uid, $l_uidval); - if (defined(my $lms = $self->{-lms_ro})) { # LeiMailSync or 0 + if (defined(my $lms = $self->{-lms_rw})) { # LeiMailSync or 0 $uri->uidvalidity($r_uidval) if defined $r_uidval; if ($mic) { my $auth = $mic->Authmechanism // ''; @@ -539,6 +540,41 @@ sub perm_fl_ok ($) { # may be overridden in NetWriter or Watch sub folder_select { $_[0]->{each_old} ? 'select' : 'examine' } +sub _imap_fetch_bodies ($$$$) { + my ($self, $mic, $uri, $uids) = @_; + my $req = $mic->imap4rev1 ? 'BODY.PEEK[]' : 'RFC822.PEEK'; + my $key = $req; + $key =~ s/\.PEEK//; + my $sec = uri_section($uri); + my $mbx = $uri->mailbox; + my $bs = $self->{cfg_opt}->{$sec}->{batch_size} // 1; + my ($last_uid, $err); + my $use_fl = $self->{-use_fl}; + + while (scalar @$uids) { + my @batch = splice(@$uids, 0, $bs); + my $batch = join(',', @batch); + local $0 = "UID:$batch $mbx $sec"; + my $r = $mic->fetch_hash($batch, $req, 'FLAGS'); + unless ($r) { # network error? + last if $!{EINTR} && $self->{quit}; + $err = "E: $uri UID FETCH $batch error: $!"; + last; + } + for my $uid (@batch) { + # messages get deleted, so holes appear + my $per_uid = delete $r->{$uid} // next; + my $raw = delete($per_uid->{$key}) // next; + my $fl = $use_fl ? $per_uid->{FLAGS} : undef; + _imap_do_msg($self, $uri, $uid, \$raw, $fl); + $last_uid = $uid; + last if $self->{quit}; + } + last if $self->{quit}; + } + ($last_uid, $err); +} + sub _imap_fetch_all ($$$) { my ($self, $mic, $orig_uri) = @_; my $sec = uri_section($orig_uri); @@ -586,6 +622,7 @@ EOF $mic->Uid(1); # the default, we hope my $err; my $use_fl = perm_fl_ok($perm_fl); + local $self->{-use_fl} = $use_fl; if (!defined($single_uid) && $self->{each_old} && $use_fl) { $err = each_old_flags($self, $mic, $uri, $l_uid); return $err if $err; @@ -597,15 +634,12 @@ EOF my $m = $mod ? " [(UID % $mod) == $shard]" : ''; warn "# $uri fetching UID $l_uid:$r_uid$m\n"; } - my $bs = $self->{cfg_opt}->{$sec}->{batch_size} // 1; - my $req = $mic->imap4rev1 ? 'BODY.PEEK[]' : 'RFC822.PEEK'; - my $key = $req; - $key =~ s/\.PEEK//; - my ($uids, $batch); + my $fetch_cb = \&_imap_fetch_bodies; do { # I wish "UID FETCH $START:*" could work, but: # 1) servers do not need to return results in any order # 2) Mail::IMAPClient doesn't offer a streaming API + my $uids; if (defined $single_uid) { $uids = [ $single_uid ]; } elsif (!($uids = $mic->search("UID $l_uid:*"))) { @@ -623,31 +657,8 @@ EOF return if $uids->[0] < $l_uid; $l_uid = $uids->[-1] + 1; # for next search - my $last_uid; - my $n = $self->{max_batch}; - @$uids = grep { ($_ % $mod) == $shard } @$uids if $mod; - while (scalar @$uids) { - my @batch = splice(@$uids, 0, $bs); - $batch = join(',', @batch); - local $0 = "UID:$batch $mbx $sec"; - my $r = $mic->fetch_hash($batch, $req, 'FLAGS'); - unless ($r) { # network error? - last if $!{EINTR} && $self->{quit}; - $err = "E: $uri UID FETCH $batch error: $!"; - last; - } - for my $uid (@batch) { - # messages get deleted, so holes appear - my $per_uid = delete $r->{$uid} // next; - my $raw = delete($per_uid->{$key}) // next; - my $fl = $use_fl ? $per_uid->{FLAGS} : undef; - _imap_do_msg($self, $uri, $uid, \$raw, $fl); - $last_uid = $uid; - last if $self->{quit}; - } - last if $self->{quit}; - } + (my $last_uid, $err) = $fetch_cb->($self, $mic, $uri, $uids); run_commit_cb($self); $itrk->update_last($r_uidval, $last_uid) if $itrk; } until ($err || $self->{quit} || defined($single_uid)); @@ -729,7 +740,10 @@ sub _nntp_fetch_all ($$$) { } (defined($num_a) && defined($num_b) && $num_a > $num_b) and return "E: $uri: backwards range: $num_a > $num_b"; - + if (defined($num_a)) { # no article numbers in mail_sync.sqlite3 + $uri = $uri->clone; + $uri->group($group); + } # IMAPTracker is also used for tracking NNTP, UID == article number # LIST.ACTIVE can get the equivalent of UIDVALIDITY, but that's # expensive. So we assume newsgroups don't change: