use PublicInbox::Eml;
use PublicInbox::ProcessPipe;
use PublicInbox::Spawn qw(spawn);
-use PublicInbox::LeiDedupe;
-use PublicInbox::PktOp qw(pkt_do);
use Symbol qw(gensym);
use IO::Handle; # ->autoflush
use Fcntl qw(SEEK_SET SEEK_END O_CREAT O_EXCL O_WRONLY);
if (my $ent = $kw2status{$k}) {
push @{$hdr{$ent->[0]}}, $ent->[1];
} else { # X-Label?
- warn "TODO: keyword `$k' not supported for mbox\n";
+ warn "# keyword `$k' not supported for mbox\n";
}
}
- # Messages are always 'O' (non-\Recent in IMAP), it saves
- # MUAs the trouble of rewriting the mbox if no other
- # changes are made. We put 'O' at the end (e.g. "Status: RO")
- # to match mutt(1) output.
- $eml->header_set('Status', join('', sort(@{$hdr{Status}})). 'O');
+ # When writing to empty mboxes, messages are always 'O'
+ # (not-\Recent in IMAP), it saves MUAs the trouble of
+ # rewriting the mbox if no other changes are made.
+ # We put 'O' at the end (e.g. "Status: RO") to match mutt(1) output.
+ # We only set smsg->{-recent} if augmenting existing stores.
+ my $status = join('', sort(@{$hdr{Status}}));
+ $status .= 'O' unless $smsg->{-recent};
+ $eml->header_set('Status', $status) if $status;
if (my $chars = delete $hdr{'X-Status'}) {
$eml->header_set('X-Status', join('', sort(@$chars)));
}
sub git_to_mail { # git->cat_async callback
my ($bref, $oid, $type, $size, $arg) = @_;
- if ($type ne 'blob') {
- if ($type eq 'missing') {
- warn "missing $oid\n";
- } else {
- warn "unexpected type=$type for $oid\n";
- }
- }
my ($write_cb, $smsg) = @$arg;
- if ($smsg->{blob} ne $oid) {
- die "BUG: expected=$smsg->{blob} got=$oid";
+ if ($type eq 'missing' && $smsg->{-lms_ro}) {
+ if ($bref = $smsg->{-lms_ro}->local_blob($oid, 1)) {
+ $type = 'blob';
+ $size = length($$bref);
+ }
}
- $write_cb->($bref, $smsg) if $size > 0;
+ return warn("W: $oid is $type (!= blob)\n") if $type ne 'blob';
+ return warn("E: $oid is empty\n") unless $size;
+ die "BUG: expected=$smsg->{blob} got=$oid" if $smsg->{blob} ne $oid;
+ $write_cb->($bref, $smsg);
}
sub reap_compress { # dwaitpid callback
$lei->fail("@$cmd failed", $? >> 8);
}
-sub _post_augment_mbox { # open a compressor process
+sub _post_augment_mbox { # open a compressor process from top-level process
my ($self, $lei) = @_;
my $zsfx = $self->{zsfx} or return;
my $cmd = PublicInbox::MboxReader::zsfx2cmd($zsfx, undef, $lei);
$lei->{1} = $pp;
}
-sub dup_src ($) {
- my ($in) = @_;
- open my $dup, '+>>&', $in or die "dup: $!";
- $dup;
-}
-
# --augment existing output destination, with deduplication
sub _augment { # MboxReader eml_cb
my ($eml, $lei) = @_;
my $dedupe = $lei->{dedupe};
$dedupe->prepare_dedupe;
my $lse = $lei->{lse}; # may be undef
+ my $set_recent = $dedupe->dedupe_nr;
sub { # for git_to_mail
my ($buf, $smsg, $eml) = @_;
$eml //= PublicInbox::Eml->new($buf);
- return if $dedupe->is_dup($eml, $smsg->{blob});
+ return if $dedupe->is_dup($eml, $smsg);
$lse->xsmsg_vmd($smsg) if $lse;
+ $smsg->{-recent} = 1 if $set_recent;
$buf = $eml2mbox->($eml, $smsg);
- return atomic_append($lei, $buf) if $atomic_append;
- my $lk = $ovv->lock_for_scope;
- $lei->out($$buf);
+ if ($atomic_append) {
+ atomic_append($lei, $buf);
+ } else {
+ my $lk = $ovv->lock_for_scope;
+ $lei->out($$buf);
+ }
+ ++$lei->{-nr_write};
}
}
sprintf('%x,%x,%x,%x', rand(0xffffffff), time, $$, ++$seq);
}
-sub _buf2maildir {
- my ($dst, $buf, $smsg) = @_;
+sub kw2suffix ($;@) {
+ my $kw = shift;
+ join('', sort(map { $kw2char{$_} // () } @$kw, @_));
+}
+
+sub _buf2maildir ($$$$) {
+ my ($dst, $buf, $smsg, $dir) = @_;
my $kw = $smsg->{kw} // [];
- my $sfx = join('', sort(map { $kw2char{$_} // () } @$kw));
my $rand = ''; # chosen by die roll :P
- my ($tmp, $fh, $final, $ok);
+ my ($tmp, $fh, $base, $ok);
my $common = $smsg->{blob} // _rand;
if (defined(my $pct = $smsg->{pct})) { $common .= "=$pct" }
do {
} while (!($ok = sysopen($fh, $tmp, O_CREAT|O_EXCL|O_WRONLY)) &&
$!{EEXIST} && ($rand = _rand.','));
if ($ok && print $fh $$buf and close($fh)) {
- # ignore new/ and write only to cur/, otherwise MUAs
- # with R/W access to the Maildir will end up doing
- # a mass rename which can take a while with thousands
- # of messages.
- $dst .= 'cur/';
+ $dst .= $dir; # 'new/' or 'cur/'
$rand = '';
do {
- $final = $dst.$rand.$common.':2,'.$sfx;
- } while (!($ok = link($tmp, $final)) && $!{EEXIST} &&
+ $base = $rand.$common.':2,'.kw2suffix($kw);
+ } while (!($ok = link($tmp, $dst.$base)) && $!{EEXIST} &&
($rand = _rand.','));
- die "link($tmp, $final): $!" unless $ok;
+ die "link($tmp, $dst$base): $!" unless $ok;
unlink($tmp) or warn "W: failed to unlink $tmp: $!\n";
+ \$base;
} else {
my $err = "Error writing $smsg->{blob} to $dst: $!\n";
$_[0] = undef; # clobber dst
$dedupe->prepare_dedupe if $dedupe;
my $dst = $lei->{ovv}->{dst};
my $lse = $lei->{lse}; # may be undef
+ my $sto = $lei->{opt}->{'mail-sync'} ? $lei->{sto} : undef;
+ my $out = $sto ? 'maildir:'.$lei->abs_path($dst) : undef;
+
+ # Favor cur/ and only write to new/ when augmenting. This
+ # saves MUAs from having to do a mass rename when the initial
+ # search result set is huge.
+ my $dir = $dedupe && $dedupe->dedupe_nr ? 'new/' : 'cur/';
sub { # for git_to_mail
- my ($buf, $smsg, $eml) = @_;
+ my ($bref, $smsg, $eml) = @_;
$dst // return $lei->fail; # dst may be undef-ed in last run
- $buf //= \($eml->as_string);
+ return if $dedupe && $dedupe->is_dup($eml //
+ PublicInbox::Eml->new($$bref),
+ $smsg);
$lse->xsmsg_vmd($smsg) if $lse;
- return _buf2maildir($dst, $buf, $smsg) if !$dedupe;
- $eml //= PublicInbox::Eml->new($$buf); # copy buf
- return if $dedupe->is_dup($eml, $smsg->{blob});
- undef $eml;
- _buf2maildir($dst, $buf, $smsg);
+ my $n = _buf2maildir($dst, $bref // \($eml->as_string),
+ $smsg, $dir);
+ $sto->ipc_do('set_sync_info', $smsg->{blob}, $out, $n) if $sto;
+ ++$lei->{-nr_write};
}
}
my ($self, $lei) = @_;
my $dedupe = $lei->{dedupe};
$dedupe->prepare_dedupe if $dedupe;
- my $imap_append = $lei->{net}->can('imap_append');
- my $mic = $lei->{net}->mic_get($self->{uri});
- my $folder = $self->{uri}->mailbox;
+ my $append = $lei->{net}->can('imap_append');
+ my $uri = $self->{uri};
+ my $mic = $lei->{net}->mic_get($uri);
+ my $folder = $uri->mailbox;
+ $uri->uidvalidity($mic->uidvalidity($folder));
my $lse = $lei->{lse}; # may be undef
+ my $sto = $lei->{opt}->{'mail-sync'} ? $lei->{sto} : undef;
sub { # for git_to_mail
my ($bref, $smsg, $eml) = @_;
- $mic // return $lei->fail; # dst may be undef-ed in last run
- if ($dedupe) {
- $eml //= PublicInbox::Eml->new($$bref); # copy bref
- return if $dedupe->is_dup($eml, $smsg->{blob});
- }
+ $mic // return $lei->fail; # mic may be undef-ed in last run
+ return if $dedupe && $dedupe->is_dup($eml //
+ PublicInbox::Eml->new($$bref),
+ $smsg);
$lse->xsmsg_vmd($smsg) if $lse;
- eval { $imap_append->($mic, $folder, $bref, $smsg, $eml) };
+ my $uid = eval { $append->($mic, $folder, $bref, $smsg, $eml) };
if (my $err = $@) {
undef $mic;
die $err;
}
+ # imap_append returns UID if IMAP server has UIDPLUS extension
+ ($sto && $uid =~ /\A[0-9]+\z/) and
+ $sto->ipc_do('set_sync_info',
+ $smsg->{blob}, $$uri, $uid + 0);
+ ++$lei->{-nr_write};
+ }
+}
+
+sub _text_write_cb ($$) {
+ my ($self, $lei) = @_;
+ my $dedupe = $lei->{dedupe};
+ $dedupe->prepare_dedupe if $dedupe;
+ my $lvt = $lei->{lvt};
+ my $ovv = $lei->{ovv};
+ $lei->{1} // die "no stdout ($ovv->{dst})"; # redirected earlier
+ $lei->{1}->autoflush(1);
+ binmode $lei->{1}, ':utf8';
+ my $lse = $lei->{lse}; # may be undef
+ sub { # for git_to_mail
+ my ($bref, $smsg, $eml) = @_;
+ $lse->xsmsg_vmd($smsg) if $lse;
+ $eml //= PublicInbox::Eml->new($bref);
+ return if $dedupe && $dedupe->is_dup($eml, $smsg);
+ my $lk = $ovv->lock_for_scope;
+ $lei->out(${$lvt->eml_to_text($smsg, $eml)}, "\n");
+ }
+}
+
+sub _v2_write_cb ($$) {
+ my ($self, $lei) = @_;
+ my $dedupe = $lei->{dedupe};
+ $dedupe->prepare_dedupe if $dedupe;
+ sub { # for git_to_mail
+ my ($bref, $smsg, $eml) = @_;
+ $eml //= PublicInbox::Eml->new($bref);
+ return if $dedupe && $dedupe->is_dup($eml, $smsg);
+ $lei->{v2w}->ipc_do('add', $eml); # V2Writable->add
}
}
sub write_cb { # returns a callback for git_to_mail
my ($self, $lei) = @_;
- # _mbox_write_cb, _maildir_write_cb or _imap_write_cb
+ # _mbox_write_cb, _maildir_write_cb, _imap_write_cb, _v2_write_cb
my $m = "_$self->{base_type}_write_cb";
$self->$m($lei);
}
my $fmt = $lei->{ovv}->{fmt};
my $dst = $lei->{ovv}->{dst};
my $self = bless {}, $cls;
+ my @conflict;
if ($fmt eq 'maildir') {
require PublicInbox::MdirReader;
$self->{base_type} = 'maildir';
$lei->{ovv}->{dst} = $dst .= '/' if substr($dst, -1) ne '/';
} elsif (substr($fmt, 0, 4) eq 'mbox') {
require PublicInbox::MboxReader;
- (-d $dst || (-e _ && !-w _)) and die
- "$dst exists and is not a writable file\n";
$self->can("eml2$fmt") or die "bad mbox format: $fmt\n";
$self->{base_type} = 'mbox';
- } elsif ($fmt =~ /\Aimaps?\z/) { # TODO .onion support
+ } elsif ($fmt =~ /\Aimaps?\z/) {
require PublicInbox::NetWriter;
+ require PublicInbox::URIimap;
my $net = PublicInbox::NetWriter->new;
- $net->add_url($dst);
$net->{quiet} = $lei->{opt}->{quiet};
- my $err = $net->errors($dst);
+ my $uri = PublicInbox::URIimap->new($dst)->canonical;
+ $net->add_url($uri);
+ my $err = $net->errors($lei);
return $lei->fail($err) if $err;
- require PublicInbox::URIimap; # TODO: URI cast early
- $self->{uri} = PublicInbox::URIimap->new($dst);
- $self->{uri}->mailbox or die "No mailbox: $dst";
+ $uri->mailbox or return $lei->fail("No mailbox: $dst");
+ $self->{uri} = $uri;
+ $dst = $lei->{ovv}->{dst} = $$uri; # canonicalized
$lei->{net} = $net;
$self->{base_type} = 'imap';
+ } elsif ($fmt eq 'text') {
+ require PublicInbox::LeiViewText;
+ $lei->{lvt} = PublicInbox::LeiViewText->new($lei);
+ $self->{base_type} = 'text';
+ @conflict = qw(mua save);
+ } elsif ($fmt eq 'v2') {
+ die "--dedupe=oid and v2 are incompatible\n" if
+ ($lei->{opt}->{dedupe}//'') eq 'oid';
+ $self->{base_type} = 'v2';
+ $lei->{opt}->{save} = \1;
+ $dst = $lei->{ovv}->{dst} = $lei->abs_path($dst);
+ @conflict = qw(mua sort);
} else {
die "bad mail --format=$fmt\n";
}
+ if ($self->{base_type} =~ /\A(?:text|mbox)\z/) {
+ (-d $dst || (-e _ && !-w _)) and die
+ "$dst exists and is not a writable file\n";
+ }
+ my @err = map { defined($lei->{opt}->{$_}) ? "--$_" : () } @conflict;
+ die "@err incompatible with $fmt\n" if @err;
$self->{dst} = $dst;
- $lei->{dedupe} = PublicInbox::LeiDedupe->new($lei);
+ $lei->{dedupe} = $lei->{lss} // do {
+ my $dd_cls = 'PublicInbox::'.
+ ($lei->{opt}->{save} ? 'LeiSavedSearch' : 'LeiDedupe');
+ eval "require $dd_cls";
+ die "$dd_cls: $@" if $@;
+ $dd_cls->new($lei);
+ };
$self;
}
File::Path::mkpath($d);
-d $d or die "$d is not a directory";
}
+ # for utime, so no opendir
+ open $self->{poke_dh}, '<', "${dst}cur" or die "open ${dst}cur: $!";
}
sub _do_augment_maildir {
my ($self, $lei) = @_;
+ return if ($lei->{opt}->{save} // 0) < 0;
my $dst = $lei->{ovv}->{dst};
my $lse = $lei->{opt}->{'import-before'} ? $lei->{lse} : undef;
my $mdr = PublicInbox::MdirReader->new;
}
sub _imap_augment_or_delete { # PublicInbox::NetReader::imap_each cb
- my ($url, $uid, $kw, $eml, $lei, $lse, $delete_mic) = @_;
+ my ($uri, $uid, $kw, $eml, $lei, $lse, $delete_mic) = @_;
update_kw_maybe($lei, $lse, $eml, $kw);
if ($delete_mic) {
- $lei->{net}->imap_delete_1($url, $uid, $delete_mic);
+ $lei->{net}->imap_delete_1($uri, $uid, $delete_mic);
} else {
_augment($eml, $lei);
}
sub _do_augment_imap {
my ($self, $lei) = @_;
+ return if ($lei->{opt}->{save} // 0) < 0;
my $net = $lei->{net};
my $lse = $lei->{opt}->{'import-before'} ? $lei->{lse} : undef;
if ($lei->{opt}->{augment}) {
}
}
+sub _pre_augment_text {
+ my ($self, $lei) = @_;
+ my $dst = $lei->{ovv}->{dst};
+ my $out;
+ my $devfd = $lei->path_to_fd($dst) // die "bad $dst";
+ if ($devfd >= 0) {
+ $out = $lei->{$devfd};
+ } else { # normal-looking path
+ if (-p $dst) {
+ open $out, '>', $dst or die "open($dst): $!";
+ } elsif (-f _ || !-e _) {
+ # text allows augment, HTML/Atom won't
+ my $mode = $lei->{opt}->{augment} ? '>>' : '>';
+ open $out, $mode, $dst or die "open($mode, $dst): $!";
+ } else {
+ die "$dst is not a file or FIFO\n";
+ }
+ }
+ $lei->{ovv}->ovv_out_lk_init if !$lei->{ovv}->{lock_path};
+ $lei->{1} = $out;
+ undef;
+}
+
sub _pre_augment_mbox {
my ($self, $lei) = @_;
my $dst = $lei->{ovv}->{dst};
my ($self, $lei) = @_;
return unless $self->{seekable};
my $opt = $lei->{opt};
+ return if ($opt->{save} // 0) < 0;
my $out = $lei->{1};
my ($fmt, $dst) = @{$lei->{ovv}}{qw(fmt dst)};
return unless -s $out;
truncate($out, 0) or die "truncate($dst): $!";
return;
}
- my $zsfx = $self->{zsfx};
- my $rd = $zsfx ? PublicInbox::MboxReader::zsfxcat($out, $zsfx, $lei)
- : dup_src($out);
+ my $rd;
+ if (my $zsfx = $self->{zsfx}) {
+ $rd = PublicInbox::MboxReader::zsfxcat($out, $zsfx, $lei);
+ } else {
+ open($rd, '+>>&', $out) or die "dup: $!";
+ }
my $dedupe;
if ($opt->{augment}) {
$dedupe = $lei->{dedupe};
$dedupe->pause_dedupe if $dedupe;
}
+sub _pre_augment_v2 {
+ my ($self, $lei) = @_;
+ my $dir = $self->{dst};
+ require PublicInbox::InboxWritable;
+ my ($ibx, @creat);
+ if (-d $dir) {
+ my $opt = { -min_inbox_version => 2 };
+ require PublicInbox::Admin;
+ my @ibx = PublicInbox::Admin::resolve_inboxes([ $dir ], $opt);
+ $ibx = $ibx[0] or die "$dir is not a v2 inbox\n";
+ } else {
+ $creat[0] = {};
+ $ibx = PublicInbox::Inbox->new({
+ name => 'lei-result', # XXX configurable
+ inboxdir => $dir,
+ version => 2,
+ address => [ 'lei@example.com' ],
+ });
+ }
+ PublicInbox::InboxWritable->new($ibx, @creat);
+ $ibx->init_inbox if @creat;
+ my $v2w = $lei->{v2w} = $ibx->importer;
+ $v2w->ipc_lock_init("$dir/ipc.lock");
+ $v2w->ipc_worker_spawn("lei/v2w $dir", $lei->oldset, { lei => $lei });
+ return if !$lei->{opt}->{shared};
+ my $d = "$lei->{ale}->{git}->{git_dir}/objects";
+ my $al = "$dir/git/0.git/objects/info/alternates";
+ open my $fh, '+>>', $al or die "open($al): $!";
+ seek($fh, 0, SEEK_SET) or die "seek($al): $!";
+ grep(/\A\Q$d\E\n/, <$fh>) and return;
+ print $fh "$d\n" or die "print($al): $!";
+ close $fh or die "close($al): $!";
+}
+
sub pre_augment { # fast (1 disk seek), runs in same process as post_augment
my ($self, $lei) = @_;
- # _pre_augment_maildir, _pre_augment_mbox
+ # _pre_augment_maildir, _pre_augment_mbox, _pre_augment_v2
my $m = $self->can("_pre_augment_$self->{base_type}") or return;
$m->($self, $lei);
}
sub do_augment { # slow, runs in wq worker
my ($self, $lei) = @_;
# _do_augment_maildir, _do_augment_mbox, or _do_augment_imap
- my $m = "_do_augment_$self->{base_type}";
- $self->$m($lei);
+ my $m = $self->can("_do_augment_$self->{base_type}") or return;
+ $m->($self, $lei);
}
# fast (spawn compressor or mkdir), runs in same process as pre_augment
$m->($self, $lei, @args);
}
+# called by every single l2m worker process
sub do_post_auth {
my ($self) = @_;
my $lei = $self->{lei};
# lei_xsearch can start as soon as all l2m workers get here
- pkt_do($lei->{pkt_op_p}, 'incr_start_query') or
+ $lei->{pkt_op_p}->pkt_do('incr_start_query') or
die "incr_start_query: $!";
my $aug;
- if (lock_free($self)) {
+ if (lock_free($self)) { # all workers do_augment
my $mod = $self->{-wq_nr_workers};
my $shard = $self->{-wq_worker_nr};
if (my $net = $lei->{net}) {
$self->{shard_info} = [ $mod, $shard ];
}
$aug = '+'; # incr_post_augment
- } elsif ($self->{-wq_worker_nr} == 0) {
+ } elsif ($self->{-wq_worker_nr} == 0) { # 1st worker do_augment
$aug = '.'; # do_post_augment
}
if ($aug) {
local $0 = 'do_augment';
eval { do_augment($self, $lei) };
$lei->fail($@) if $@;
- pkt_do($lei->{pkt_op_p}, $aug) == 1 or
+ $lei->{pkt_op_p}->pkt_do($aug) == 1 or
die "do_post_augment trigger: $!";
}
+ # done augmenting, connect the compressor pipe for each worker
if (my $zpipe = delete $lei->{zpipe}) {
$lei->{1} = $zpipe->[1];
close $zpipe->[0];
}
+ my $au_peers = delete $self->{au_peers};
+ if ($au_peers) { # wait for peer l2m to finish augmenting:
+ $au_peers->[1] = undef;
+ sysread($au_peers->[0], my $barrier1, 1);
+ }
$self->{wcb} = $self->write_cb($lei);
+ if ($au_peers) { # wait for peer l2m to set write_cb
+ $au_peers->[3] = undef;
+ sysread($au_peers->[2], my $barrier2, 1);
+ }
}
sub ipc_atfork_child {
my ($self) = @_;
my $lei = $self->{lei};
$lei->_lei_atfork_child;
+ if (my $lse = $lei->{lse}) {
+ $self->{-lms_ro} = $lse->{-lms_ro} //= $lse->lms;
+ }
$lei->{auth}->do_auth_atfork($self) if $lei->{auth};
$SIG{__WARN__} = PublicInbox::Eml::warn_ignore_cb();
$self->SUPER::ipc_atfork_child;
$_[0]->{base_type} =~ /\A(?:maildir|imap|jmap)\z/ ? 1 : 0;
}
+# wakes up the MUA when complete so it can refresh messages list
sub poke_dst {
my ($self) = @_;
if ($self->{base_type} eq 'maildir') {
my $t = time + 1;
- utime($t, $t, $self->{dst} . 'cur');
+ utime($t, $t, $self->{poke_dh}) or warn "futimes: $!";
}
}
sub write_mail { # via ->wq_io_do
- my ($self, $smsg) = @_;
+ my ($self, $smsg, $eml) = @_;
+ return $self->{wcb}->(undef, $smsg, $eml) if $eml;
+ $smsg->{-lms_ro} = $self->{-lms_ro};
$self->{lei}->{ale}->git->cat_async($smsg->{blob}, \&git_to_mail,
[$self->{wcb}, $smsg]);
}
sub wq_atexit_child {
my ($self) = @_;
+ my $lei = $self->{lei};
delete $self->{wcb};
- $self->{lei}->{ale}->git->async_wait_all;
- $SIG{__WARN__} = 'DEFAULT';
+ $lei->{ale}->git->async_wait_all;
+ my $nr = delete($lei->{-nr_write}) or return;
+ return if $lei->{early_mua} || !$lei->{-progress} || !$lei->{pkt_op_p};
+ $lei->{pkt_op_p}->pkt_do('l2m_progress', $nr);
}
# called in top-level lei-daemon when LeiAuth is done
-sub net_merge_complete {
+sub net_merge_all_done {
my ($self) = @_;
$self->wq_broadcast('do_post_auth');
$self->wq_close(1);