-# Copyright (C) 2015 all contributors <meta@public-inbox.org>
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
#
# Each instance of this represents a NNTP client socket
package PublicInbox::NNTP;
use Email::Simple;
use POSIX qw(strftime);
use Time::HiRes qw(clock_gettime CLOCK_MONOTONIC);
+use Digest::SHA qw(sha1_hex);
+use Time::Local qw(timegm timelocal);
use constant {
r501 => '501 command syntax error',
r221 => '221 Header follows',
qw(:bytes :lines Xref To Cc)) . "\r\n";
# disable commands with easy DoS potential:
-# LISTGROUP could get pretty bad, too...
my %DISABLED; # = map { $_ => 1 } qw(xover list_overview_fmt newnews xhdr);
my $EXPMAP; # fd -> [ idle_time, $self ]
sub process_line ($$) {
my ($self, $l) = @_;
my ($req, @args) = split(/\s+/, $l);
+ return unless defined($req);
$req = lc($req);
$req = eval {
no strict 'refs';
}
$self->{ng} or return '412 no newsgroup selected';
- long_response($self, 0, long_response_limit, sub {
- my ($i) = @_;
- my $nr = $self->{ng}->mm->id_batch($$i, sub {
- my ($ary) = @_;
- more($self, join("\r\n", @$ary));
- });
-
- # -1 to adjust for implicit increment in long_response
- $$i = $nr ? $$i + $nr - 1 : long_response_limit;
+ my $n = 0;
+ long_response($self, sub {
+ my $ary = $self->{ng}->mm->ids_after(\$n);
+ scalar @$ary or return;
+ more($self, join("\r\n", @$ary));
+ 1;
});
}
sub parse_time ($$;$) {
my ($date, $time, $gmt) = @_;
- use Time::Local qw();
my ($hh, $mm, $ss) = unpack('A2A2A2', $time);
if (defined $gmt) {
$gmt =~ /\A(?:UTC|GMT)\z/i or die "GM invalid: $gmt";
($YYYY, $MM, $DD) = unpack('A4A2A2', $date);
} else { # legacy clients send YYMMDD
($YYYY, $MM, $DD) = unpack('A2A2A2', $date);
- if ($YYYY > strftime('%y', @now)) {
- my $cur_year = $now[5] + 1900;
+ my $cur_year = $now[5] + 1900;
+ if ($YYYY > $cur_year) {
$YYYY += int($cur_year / 1000) * 1000 - 100;
}
}
if ($gmt) {
- Time::Local::timegm($ss, $mm, $hh, $DD, $MM - 1, $YYYY);
+ timegm($ss, $mm, $hh, $DD, $MM - 1, $YYYY);
} else {
- Time::Local::timelocal($ss, $mm, $hh, $DD, $MM - 1, $YYYY);
+ timelocal($ss, $mm, $hh, $DD, $MM - 1, $YYYY);
}
}
return $_[0] = qr/.*/ if (!defined $_[0] || $_[0] eq '*');
my %keep;
my $salt = rand;
- use Digest::SHA qw(sha1_hex);
my $tmp = $_[0];
$tmp =~ s#(?<!\\)\[(.+)(?<!\\)\]#
};
return '.' unless @srch;
- $ts .= '..';
- my $opts = { asc => 1, limit => 1000, offset => 0 };
- long_response($self, 0, long_response_limit, sub {
- my ($i) = @_;
+ my $prev = 0;
+ long_response($self, sub {
my $srch = $srch[0];
- my $res = $srch->query($ts, $opts);
- my $msgs = $res->{msgs};
- if (my $nr = scalar @$msgs) {
+ my $msgs = $srch->query_ts($ts, $prev);
+ if (scalar @$msgs) {
more($self, '<' .
join(">\r\n<", map { $_->mid } @$msgs ).
'>');
- $opts->{offset} += $nr;
+ $prev = $msgs->[-1]->{num};
} else {
shift @srch;
if (@srch) { # continue onto next newsgroup
- $opts->{offset} = 0;
+ $prev = 0;
+ return 1;
} else { # break out of the long response.
- $$i = long_response_limit;
+ return;
}
}
});
$hdr->header_set($k, @v, $v);
}
-sub set_nntp_headers {
- my ($hdr, $ng, $n, $mid) = @_;
+sub xref ($$$$) {
+ my ($self, $ng, $n, $mid) = @_;
+ my $ret = "$ng->{domain} $ng->{newsgroup}:$n";
+
+ # num_for is pretty cheap and sometimes we'll lookup the existence
+ # of an article without getting even the OVER info. In other words,
+ # I'm not sure if its worth optimizing by scanning To:/Cc: and
+ # PublicInbox::ExtMsg on the PSGI end is just as expensive
+ foreach my $other (@{$self->{nntpd}->{grouplist}}) {
+ next if $ng eq $other;
+ my $num = eval { $other->mm->num_for($mid) } or next;
+ $ret .= " $other->{newsgroup}:$num";
+ }
+ $ret;
+}
+
+sub set_nntp_headers ($$$$$) {
+ my ($self, $hdr, $ng, $n, $mid) = @_;
# clobber some
- $hdr->header_set('Newsgroups', $ng->{newsgroup});
- $hdr->header_set('Xref', xref($ng, $n));
+ my $xref = xref($self, $ng, $n, $mid);
+ $hdr->header_set('Xref', $xref);
+ $xref =~ s/:\d+//g;
+ $hdr->header_set('Newsgroups', (split(/ /, $xref, 2))[1]);
header_append($hdr, 'List-Post', "<mailto:$ng->{-primary_address}>");
if (my $url = $ng->base_url) {
$mid = mid_escape($mid);
defined $mid or return $err;
}
found:
- my $bytes;
- my $s = eval { $ng->msg_by_mid($mid, \$bytes) } or return $err;
- $s = Email::Simple->new($s);
- my $lines;
+ my $smsg = $ng->search->{over_ro}->get_art($n) or return $err;
+ my $msg = $ng->msg_by_smsg($smsg) or return $err;
+ my $s = Email::Simple->new($msg);
if ($set_headers) {
- set_nntp_headers($s->header_obj, $ng, $n, $mid);
- $lines = $s->body =~ tr!\n!\n!;
+ set_nntp_headers($self, $s->header_obj, $ng, $n, $mid);
# must be last
$s->body_set('') if ($set_headers == 2);
}
- [ $n, $mid, $s, $bytes, $lines, $ng ];
+ [ $n, $mid, $s, $smsg->bytes, $smsg->lines, $ng ];
}
sub simple_body_write ($$) {
[ $beg, $end ];
}
-sub long_response ($$$$) {
- my ($self, $beg, $end, $cb) = @_;
+sub long_response ($$) {
+ my ($self, $cb) = @_;
die "BUG: nested long response" if $self->{long_res};
my $fd = $self->{fd};
$self->watch_read(0);
my $t0 = now();
$self->{long_res} = sub {
- # limit our own running time for fairness with other
- # clients and to avoid buffering too much:
- my $lim = 100;
-
- my $err;
- do {
- eval { $cb->(\$beg) };
- } until (($err = $@) || $self->{closed} ||
- ++$beg > $end || !--$lim || $self->{write_buf_size});
-
- if ($err || $self->{closed}) {
+ my $more = eval { $cb->() };
+ if ($@ || $self->{closed}) {
$self->{long_res} = undef;
- if ($err) {
+ if ($@) {
err($self,
"%s during long response[$fd] - %0.6f",
- $err, now() - $t0);
+ $@, now() - $t0);
}
if ($self->{closed}) {
out($self, " deferred[$fd] aborted - %0.6f",
update_idle_time($self);
$self->watch_read(1);
}
- } elsif (!$lim || $self->{write_buf_size}) {
+ } elsif ($more) { # $self->{write_buf_size}:
# no recursion, schedule another call ASAP
# but only after all pending writes are done
update_idle_time($self);
my $mm = $self->{ng}->mm;
my ($beg, $end) = @$r;
more($self, $xhdr ? r221 : r225);
- long_response($self, $beg, $end, sub {
- my ($i) = @_;
- my $mid = $mm->mid_for($$i);
- more($self, "$$i <$mid>") if defined $mid;
+ long_response($self, sub {
+ my $r = $mm->msg_range(\$beg, $end);
+ @$r or return;
+ more($self, join("\r\n", map {
+ "$_->[0] <$_->[1]>"
+ } @$r));
+ 1;
});
}
}
-sub xref ($$) {
- my ($ng, $n) = @_;
- "$ng->{domain} $ng->{newsgroup}:$n"
-}
-
sub mid_lookup ($$) {
my ($self, $mid) = @_;
my $self_ng = $self->{ng};
my ($self, $xhdr, $range) = @_;
if (defined $range && $range =~ /\A<(.+)>\z/) { # Message-ID
- my ($ng, $n) = mid_lookup($self, $1);
+ my $mid = $1;
+ my ($ng, $n) = mid_lookup($self, $mid);
return r430 unless $n;
- hdr_mid_response($self, $xhdr, $ng, $n, $range, xref($ng, $n));
+ hdr_mid_response($self, $xhdr, $ng, $n, $range,
+ xref($self, $ng, $n, $mid));
} else { # numeric range
$range = $self->{article} unless defined $range;
my $r = get_range($self, $range);
my $mm = $ng->mm;
my ($beg, $end) = @$r;
more($self, $xhdr ? r221 : r225);
- long_response($self, $beg, $end, sub {
- my ($i) = @_;
- my $mid = $mm->mid_for($$i);
- more($self, "$$i ".xref($ng, $$i)) if defined $mid;
+ long_response($self, sub {
+ my $r = $mm->msg_range(\$beg, $end);
+ @$r or return;
+ more($self, join("\r\n", map {
+ my $num = $_->[0];
+ "$num ".xref($self, $ng, $num, $_->[1]);
+ } @$r));
+ 1;
});
}
}
sub search_header_for {
- my ($srch, $mid, $field) = @_;
- my $smsg = $srch->lookup_mail($mid) or return;
- $smsg->$field;
+ my ($srch, $num, $field) = @_;
+ my $smsg = $srch->{over_ro}->get_art($num) or return;
+ return PublicInbox::SearchMsg::date($smsg) if $field eq 'date';
+ $smsg->{$field};
}
sub hdr_searchmsg ($$$$) {
my ($self, $xhdr, $field, $range) = @_;
if (defined $range && $range =~ /\A<(.+)>\z/) { # Message-ID
my ($ng, $n) = mid_lookup($self, $1);
- return r430 unless $n;
- my $v = search_header_for($ng->search, $range, $field);
+ return r430 unless defined $n;
+ my $v = search_header_for($ng->search, $n, $field);
hdr_mid_response($self, $xhdr, $ng, $n, $range, $v);
} else { # numeric range
$range = $self->{article} unless defined $range;
return $r unless ref $r;
my ($beg, $end) = @$r;
more($self, $xhdr ? r221 : r225);
- my $off = 0;
- long_response($self, $beg, $end, sub {
- my ($i) = @_;
- my $res = $srch->query_xover($beg, $end, $off);
- my $msgs = $res->{msgs};
+ my $cur = $beg;
+ long_response($self, sub {
+ my $msgs = $srch->query_xover($cur, $end);
my $nr = scalar @$msgs or return;
- $off += $nr;
my $tmp = '';
foreach my $s (@$msgs) {
- $tmp .= $s->num . ' ' . $s->$field . "\r\n";
+ $tmp .= $s->{num} . ' ' . $s->$field . "\r\n";
}
utf8::encode($tmp);
do_more($self, $tmp);
- # -1 to adjust for implicit increment in long_response
- $$i = $nr ? $$i + $nr - 1 : long_response_limit;
+ $cur = $msgs->[-1]->{num} + 1;
});
}
}
my $mm = $ng->mm;
my $srch = $ng->search;
more($self, '224 Overview information follows');
- long_response($self, $beg, $end, sub {
- my ($i) = @_;
- my $mid = $mm->mid_for($$i) or return;
- my $h = search_header_for($srch, $mid, 'references');
- more($self, "$$i $h");
+
+ long_response($self, sub {
+ my $h = search_header_for($srch, $beg, 'references');
+ more($self, "$beg $h") if defined($h);
+ $beg++ < $end;
});
}
$smsg->{subject},
$smsg->{from},
PublicInbox::SearchMsg::date($smsg),
- '<'.PublicInbox::SearchMsg::mid($smsg).'>',
+ "<$smsg->{mid}>",
$smsg->{references},
- PublicInbox::SearchMsg::bytes($smsg),
- PublicInbox::SearchMsg::lines($smsg));
+ $smsg->{bytes},
+ $smsg->{lines});
utf8::encode($s);
$s
}
my ($self, $range) = @_;
if ($range && $range =~ /\A<(.+)>\z/) {
my ($ng, $n) = mid_lookup($self, $1);
- my $smsg = $ng->search->lookup_mail($range) or
- return '430 No article with that message-id';
+ defined $n or return r430;
+ my $smsg = $ng->search->{over_ro}->get_art($n) or return r430;
more($self, '224 Overview information follows (multi-line)');
# Only set article number column if it's the current group
my ($beg, $end) = @$r;
more($self, "224 Overview information follows for $beg to $end");
my $srch = $self->{ng}->search;
- my $off = 0;
- long_response($self, $beg, $end, sub {
- my ($i) = @_;
- my $res = $srch->query_xover($beg, $end, $off);
- my $msgs = $res->{msgs};
+ my $cur = $beg;
+ long_response($self, sub {
+ my $msgs = $srch->query_xover($cur, $end);
my $nr = scalar @$msgs or return;
- $off += $nr;
# OVERVIEW.FMT
more($self, join("\r\n", map {
- over_line(PublicInbox::SearchMsg::num($_), $_);
+ over_line($_->{num}, $_);
} @$msgs));
-
- # -1 to adjust for implicit increment in long_response
- $$i = $nr ? $$i + $nr - 1 : long_response_limit;
+ $cur = $msgs->[-1]->{num} + 1;
});
}
sub event_read {
my ($self) = @_;
use constant LINE_MAX => 512; # RFC 977 section 2.3
- my $r = 1;
- my $buf = $self->read(LINE_MAX) or return $self->close;
- $self->{rbuf} .= $$buf;
- while ($r > 0 && $self->{rbuf} =~ s/\A\s*([^\r\n]+)\r?\n//) {
+ if (index($self->{rbuf}, "\n") < 0) {
+ my $buf = $self->read(LINE_MAX) or return $self->close;
+ $self->{rbuf} .= $$buf;
+ }
+ my $r = 1;
+ while ($r > 0 && $self->{rbuf} =~ s/\A\s*([^\r\n]*)\r?\n//) {
my $line = $1;
return $self->close if $line =~ /[[:cntrl:]]/s;
my $t0 = now();
sub watch_read {
my ($self, $bool) = @_;
my $rv = $self->SUPER::watch_read($bool);
- if ($bool && $self->{rbuf} ne '') {
+ if ($bool && 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