X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FNNTP.pm;h=26bc679f2996b67e0008b8f90bfb1a1d9732521f;hb=ecea327e3d4386a22652fc08f71ac7d65b8f9b70;hp=7729399accdd91ff68163cac6a5c89db82cb6434;hpb=dbf0cad365839a99e8582d6e26ce40c02508155d;p=public-inbox.git diff --git a/lib/PublicInbox/NNTP.pm b/lib/PublicInbox/NNTP.pm index 7729399a..26bc679f 100644 --- a/lib/PublicInbox/NNTP.pm +++ b/lib/PublicInbox/NNTP.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2015-2018 all contributors +# Copyright (C) 2015-2019 all contributors # License: AGPL-3.0+ # # Each instance of this represents a NNTP client socket @@ -6,7 +6,7 @@ package PublicInbox::NNTP; use strict; use warnings; use base qw(PublicInbox::DS); -use fields qw(nntpd article rbuf ng long_res); +use fields qw(nntpd article ng); use PublicInbox::Search; use PublicInbox::Msgmap; use PublicInbox::MID qw(mid_escape); @@ -14,7 +14,7 @@ use PublicInbox::Git; require PublicInbox::EvCleanup; use Email::Simple; use POSIX qw(strftime); -use Time::HiRes qw(clock_gettime CLOCK_MONOTONIC); +PublicInbox::DS->import(qw(now msg_more)); use Digest::SHA qw(sha1_hex); use Time::Local qw(timegm timelocal); use constant { @@ -24,8 +24,8 @@ use constant { r225 => '225 Headers follow (multi-line)', r430 => '430 No article with that message-id', }; - -sub now () { clock_gettime(CLOCK_MONOTONIC) }; +use PublicInbox::Syscall qw(EPOLLIN EPOLLONESHOT); +use Errno qw(EAGAIN); my @OVERVIEW = qw(Subject From Date Message-ID References Xref); my $OVERVIEW_FMT = join(":\r\n", @OVERVIEW, qw(Bytes Lines)) . ":\r\n"; @@ -38,37 +38,11 @@ my %DISABLED; # = map { $_ => 1 } qw(xover list_overview_fmt newnews xhdr); my $EXPMAP; # fd -> [ idle_time, $self ] my $expt; our $EXPTIME = 180; # 3 minutes -my $nextt; - -my $nextq = []; -sub next_tick () { - $nextt = undef; - my $q = $nextq; - $nextq = []; - foreach my $nntp (@$q) { - # for request && response protocols, always finish writing - # before finishing reading: - if (my $long_cb = $nntp->{long_res}) { - $nntp->write($long_cb); - } else { - # pipelined request, we bypassed socket-readiness - # checks to get here: - event_read($nntp); - - # maybe there's more pipelined data, or we'll have - # to register it for socket-readiness notifications - if (!$nntp->{long_res} && !$nntp->{closed}) { - check_read($nntp); - } - } - } -} sub update_idle_time ($) { my ($self) = @_; - my $sock = $self->{sock} or return; - my $fd = fileno($sock); - defined $fd and $EXPMAP->{$fd} = [ now(), $self ]; + my $sock = $self->{sock} or return; + $EXPMAP->{fileno($sock)} = [ now(), $self ]; } sub expire_old () { @@ -80,31 +54,38 @@ sub expire_old () { while (my ($fd, $v) = each %$EXPMAP) { my ($idle_time, $nntp) = @$v; if ($idle_time < $old) { - $nntp->close; # idempotent + if (!$nntp->shutdn) { + ++$nr; + $new{$fd} = $v; + } } else { ++$nr; $new{$fd} = $v; } } $EXPMAP = \%new; - if ($nr) { - $expt = PublicInbox::EvCleanup::later(*expire_old); - } else { - $expt = undef; - # noop to kick outselves out of the loop ASAP so descriptors - # really get closed - PublicInbox::EvCleanup::asap(sub {}); - } + $expt = PublicInbox::EvCleanup::later(*expire_old) if $nr; } +sub greet ($) { $_[0]->write($_[0]->{nntpd}->{greet}) }; + sub new ($$$) { my ($class, $sock, $nntpd) = @_; my $self = fields::new($class); - $self->SUPER::new($sock); + my $ev = EPOLLIN; + my $wbuf; + if (ref($sock) eq 'IO::Socket::SSL' && !$sock->accept_SSL) { + return CORE::close($sock) if $! != EAGAIN; + $ev = PublicInbox::TLS::epollbit(); + $wbuf = [ \&PublicInbox::DS::accept_tls_step, \&greet ]; + } + $self->SUPER::new($sock, $ev | EPOLLONESHOT); $self->{nntpd} = $nntpd; - res($self, '201 ' . $nntpd->{servername} . ' ready - post via email'); - $self->{rbuf} = ''; - $self->watch_read(1); + if ($wbuf) { + $self->{wbuf} = $wbuf; + } else { + greet($self); + } update_idle_time($self); $expt ||= PublicInbox::EvCleanup::later(*expire_old); $self; @@ -134,7 +115,7 @@ sub process_line ($$) { my $res = eval { $req->($self, @args) }; my $err = $@; - if ($err && !$self->{closed}) { + if ($err && $self->{sock}) { local $/ = "\n"; chomp($l); err($self, 'error from: %s (%s)', $l, $err); @@ -162,12 +143,12 @@ sub cmd_xgtitle ($;$) { sub list_overview_fmt ($) { my ($self) = @_; - do_more($self, $OVERVIEW_FMT); + msg_more($self, $OVERVIEW_FMT); } sub list_headers ($;$) { my ($self) = @_; - do_more($self, $LIST_HEADERS); + msg_more($self, $LIST_HEADERS); } sub list_active ($;$) { @@ -251,7 +232,7 @@ sub parse_time ($$;$) { } my @now = $gmt ? gmtime : localtime; my ($YYYY, $MM, $DD); - if (length($date) == 8) { # RFC 3977 allows YYYYMMDD + if (bytes::length($date) == 8) { # RFC 3977 allows YYYYMMDD ($YYYY, $MM, $DD) = unpack('A4A2A2', $date); } else { # legacy clients send YYMMDD ($YYYY, $MM, $DD) = unpack('A2A2A2', $date); @@ -403,7 +384,7 @@ sub cmd_post ($) { sub cmd_quit ($) { my ($self) = @_; res($self, '205 closing connection - goodbye!'); - $self->close; + $self->shutdn; undef; } @@ -435,6 +416,26 @@ sub xref ($$$$) { sub set_nntp_headers ($$$$$) { my ($self, $hdr, $ng, $n, $mid) = @_; + # why? leafnode requires a Path: header for some inexplicable + # reason. We'll fake the shortest one possible. + $hdr->header_set('Path', 'y'); + + # leafnode (and maybe other NNTP clients) have trouble dealing + # with v2 messages which have multiple Message-IDs (either due + # to our own content-based dedupe or buggy git-send-email versions). + my @mids = $hdr->header('Message-ID'); + if (scalar(@mids) > 1) { + my $mid0 = "<$mid>"; + $hdr->header_set('Message-ID', $mid0); + my @alt = $hdr->header('X-Alt-Message-ID'); + my %seen = map { $_ => 1 } (@alt, $mid0); + foreach my $m (@mids) { + next if $seen{$m}++; + push @alt, $m; + } + $hdr->header_set('X-Alt-Message-ID', @alt); + } + # clobber some my $xref = xref($self, $ng, $n, $mid); $hdr->header_set('Xref', $xref); @@ -486,24 +487,23 @@ find_mid: found: my $smsg = $ng->over->get_art($n) or return $err; my $msg = $ng->msg_by_smsg($smsg) or return $err; - my $s = Email::Simple->new($msg); - if ($set_headers) { - set_nntp_headers($self, $s->header_obj, $ng, $n, $mid); - # must be last - $s->body_set('') if ($set_headers == 2); - } - [ $n, $mid, $s, $smsg->bytes, $smsg->lines, $ng ]; + # Email::Simple->new will modify $msg in-place as documented + # in its manpage, so what's left is the body and we won't need + # to call Email::Simple::body(), later + my $hdr = Email::Simple->new($msg)->header_obj; + set_nntp_headers($self, $hdr, $ng, $n, $mid) if $set_headers; + [ $n, $mid, $msg, $hdr ]; } -sub simple_body_write ($$) { - my ($self, $s) = @_; - my $body = $s->body; - $s->body_set(''); - $body =~ s/^\./../smg; - $body =~ s/(?{article} = $art if defined $art && $art =~ /\A[0-9]+\z/; } -sub _header ($) { - my $hdr = $_[0]->header_obj->as_string; +sub msg_hdr_write ($$$) { + my ($self, $hdr, $body_follows) = @_; + $hdr = $hdr->as_string; utf8::encode($hdr); - $hdr =~ s/(? article retrieved - head and body follow"); - do_more($self, _header($s)); - do_more($self, "\r\n"); - simple_body_write($self, $s); + msg_hdr_write($self, $hdr, 1); + msg_body_write($self, $msg); } sub cmd_head ($;$) { my ($self, $art) = @_; my $r = art_lookup($self, $art, 2); return $r unless ref $r; - my ($n, $mid, $s) = @$r; + my ($n, $mid, undef, $hdr) = @$r; set_art($self, $art); more($self, "221 $n <$mid> article retrieved - head follows"); - do_more($self, _header($s)); + msg_hdr_write($self, $hdr, 0); '.' } @@ -546,17 +553,17 @@ sub cmd_body ($;$) { my ($self, $art) = @_; my $r = art_lookup($self, $art, 0); return $r unless ref $r; - my ($n, $mid, $s) = @$r; + my ($n, $mid, $msg) = @$r; set_art($self, $art); more($self, "222 $n <$mid> article retrieved - body follows"); - simple_body_write($self, $s); + msg_body_write($self, $msg); } sub cmd_stat ($;$) { my ($self, $art) = @_; my $r = art_lookup($self, $art, 0); return $r unless ref $r; - my ($n, $mid, undef) = @$r; + my ($n, $mid) = @$r; set_art($self, $art); "223 $n <$mid> article retrieved - request text separately"; } @@ -593,48 +600,46 @@ sub get_range ($$) { } sub long_response ($$) { - my ($self, $cb) = @_; - die "BUG: nested long response" if $self->{long_res}; + my ($self, $cb) = @_; # cb returns true if more, false if done my $fd = fileno($self->{sock}); defined $fd or return; # make sure we disable reading during a long response, # clients should not be sending us stuff and making us do more # work while we are stream a response to them - $self->watch_read(0); my $t0 = now(); - $self->{long_res} = sub { + my $long_cb; # DANGER: self-referential + $long_cb = sub { + # wbuf is unset or empty, here; $cb may add to it my $more = eval { $cb->() }; - if ($@ || $self->{closed}) { - $self->{long_res} = undef; - + if ($@ || !$self->{sock}) { # something bad happened... + $long_cb = undef; + my $diff = now() - $t0; if ($@) { err($self, "%s during long response[$fd] - %0.6f", - $@, now() - $t0); - } - if ($self->{closed}) { - out($self, " deferred[$fd] aborted - %0.6f", - now() - $t0); - } else { - update_idle_time($self); - check_read($self); + $@, $diff); } - } elsif ($more) { # scalar @{$self->{wbuf}}: + out($self, " deferred[$fd] aborted - %0.6f", $diff); + $self->close; + } elsif ($more) { # $self->{wbuf}: + update_idle_time($self); + # no recursion, schedule another call ASAP # but only after all pending writes are done - update_idle_time($self); + my $wbuf = $self->{wbuf} ||= []; + push @$wbuf, $long_cb; - push @$nextq, $self; - $nextt ||= PublicInbox::EvCleanup::asap(*next_tick); + # wbuf may be populated by $cb, no need to rearm if so: + $self->requeue if scalar(@$wbuf) == 1; } else { # all done! - $self->{long_res} = undef; - check_read($self); + $long_cb = undef; res($self, '.'); out($self, " deferred[$fd] done - %0.6f", now() - $t0); + $self->requeue unless $self->{wbuf}; } }; - $self->{long_res}->(); # kick off! + $self->write($long_cb); # kick off! undef; } @@ -738,7 +743,7 @@ sub hdr_searchmsg ($$$$) { $tmp .= $s->{num} . ' ' . $s->$field . "\r\n"; } utf8::encode($tmp); - do_more($self, $tmp); + msg_more($self, $tmp); $cur = $msgs->[-1]->{num} + 1; }); } @@ -786,7 +791,7 @@ sub hdr_mid_prefix ($$$$$) { } sub hdr_mid_response ($$$$$$) { - my ($self, $xhdr, $ng, $n, $mid, $v) = @_; # r: art_lookup result + my ($self, $xhdr, $ng, $n, $mid, $v) = @_; my $res = ''; if ($xhdr) { $res .= r221 . "\r\n"; @@ -877,6 +882,19 @@ sub cmd_xover ($;$) { }); } +sub cmd_starttls ($) { + my ($self) = @_; + my $sock = $self->{sock} or return; + # RFC 4642 2.2.1 + (ref($sock) eq 'IO::Socket::SSL') and return '502 Command unavailable'; + my $opt = $self->{nntpd}->{accept_tls} or + return '580 can not initiate TLS negotiation'; + res($self, '382 Continue with TLS negotiation'); + $self->{sock} = IO::Socket::SSL->start_SSL($sock, %$opt); + $self->requeue if PublicInbox::DS::accept_tls_step($self); + undef; +} + sub cmd_xpath ($$) { my ($self, $mid) = @_; return r501 unless $mid =~ /\A<(.+)>\z/; @@ -890,24 +908,14 @@ sub cmd_xpath ($$) { '223 '.join(' ', @paths); } -sub res ($$) { - my ($self, $line) = @_; - do_write($self, $line . "\r\n"); -} +sub res ($$) { do_write($_[0], $_[1] . "\r\n") } -sub more ($$) { - my ($self, $line) = @_; - do_more($self, $line . "\r\n"); -} +sub more ($$) { msg_more($_[0], $_[1] . "\r\n") } sub do_write ($$) { - my ($self, $data) = @_; - my $done = $self->write($data); - return 0 if $self->{closed}; - - # Do not watch for readability if we have data in the queue, - # instead re-enable watching for readability when we can - $self->watch_read(0) if (!$done || $self->{long_res}); + my $self = $_[0]; + my $done = $self->write(\($_[1])); + return 0 unless $self->{sock}; $done; } @@ -922,83 +930,49 @@ sub out ($$;@) { printf { $self->{nntpd}->{out} } $fmt."\n", @args; } -use constant MSG_MORE => ($^O eq 'linux') ? 0x8000 : 0; - -sub do_more ($$) { - my ($self, $data) = @_; - if (MSG_MORE && !scalar(@{$self->{wbuf}})) { - my $n = send($self->{sock}, $data, MSG_MORE); - if (defined $n) { - my $dlen = length($data); - return 1 if $n == $dlen; # all done! - $data = substr($data, $n, $dlen - $n); - } - } - do_write($self, $data); -} - -# callbacks for PublicInbox::DS +# callback used by PublicInbox::DS for any (e)poll (in/out/hup/err) +sub event_step { + my ($self) = @_; -sub event_hup { $_[0]->close } -sub event_err { $_[0]->close } + return unless $self->flush_write && $self->{sock}; -sub event_write { - my ($self) = @_; update_idle_time($self); - # only continue watching for readability when we are done writing: - if ($self->write(undef) == 1 && !$self->{long_res}) { - $self->watch_read(1); - } -} + # only read more requests if we've drained the write buffer, + # otherwise we can be buffering infinitely w/o backpressure -sub event_read { - my ($self) = @_; use constant LINE_MAX => 512; # RFC 977 section 2.3 + my $rbuf = $self->{rbuf} // (\(my $x = '')); + my $r = 1; - if (index($self->{rbuf}, "\n") < 0) { - my $buf = $self->read(LINE_MAX) or return $self->close; - $self->{rbuf} .= $$buf; + if (index($$rbuf, "\n") < 0) { + my $off = bytes::length($$rbuf); + $r = $self->do_read($rbuf, LINE_MAX, $off) or return; } - my $r = 1; - while ($r > 0 && $self->{rbuf} =~ s/\A[ \t\r\n]*([^\r\n]*)\r?\n//) { + while ($r > 0 && $$rbuf =~ s/\A[ \t\r\n]*([^\r\n]*)\r?\n//) { my $line = $1; return $self->close if $line =~ /[[:cntrl:]]/s; my $t0 = now(); my $fd = fileno($self->{sock}); $r = eval { process_line($self, $line) }; - my $d = $self->{long_res} ? - " deferred[$fd]" : ''; - out($self, "[$fd] %s - %0.6f$d", $line, now() - $t0); + my $pending = $self->{wbuf} ? ' pending' : ''; + out($self, "[$fd] %s - %0.6f$pending", $line, now() - $t0); } return $self->close if $r < 0; - my $len = length($self->{rbuf}); + my $len = bytes::length($$rbuf); return $self->close if ($len >= LINE_MAX); + $self->rbuf_idle($rbuf); update_idle_time($self); -} -sub check_read { - my ($self) = @_; - if (index($self->{rbuf}, "\n") >= 0) { - # Force another read if there is a pipelined request. - # We don't know if the socket has anything for us to read, - # and we must double-check again by the time the timer fires - # in case we really did dispatch a read event and started - # another long response. - push @$nextq, $self; - $nextt ||= PublicInbox::EvCleanup::asap(*next_tick); - } else { - # no pipelined requests available, let the kernel know - # to wake us up if there's more - $self->watch_read(1); # PublicInbox::DS::watch_read - } + # maybe there's more pipelined data, or we'll have + # to register it for socket-readiness notifications + $self->requeue unless $self->{wbuf}; } sub not_idle_long ($$) { my ($self, $now) = @_; - my $sock = $self->{sock} or return; - defined(my $fd = fileno($sock)) or return; - my $ary = $EXPMAP->{$fd} or return; + my $sock = $self->{sock} or return; + my $ary = $EXPMAP->{fileno($sock)} or return; my $exp_at = $ary->[0] + $EXPTIME; $exp_at > $now; } @@ -1006,8 +980,7 @@ sub not_idle_long ($$) { # for graceful shutdown in PublicInbox::Daemon: sub busy { my ($self, $now) = @_; - ($self->{rbuf} ne '' || $self->{long_res} || - scalar(@{$self->{wbuf}}) || not_idle_long($self, $now)); + ($self->{rbuf} || $self->{wbuf} || not_idle_long($self, $now)); } 1;