use PublicInbox::EmlContentFoo qw(parse_content_disposition);
use PublicInbox::DS qw(now);
use PublicInbox::Syscall qw(EPOLLIN EPOLLONESHOT);
+use PublicInbox::GitAsyncCat;
use Text::ParseWords qw(parse_line);
use Errno qw(EAGAIN);
+
my $Address;
for my $mod (qw(Email::Address::XS Mail::Address)) {
eval "require $mod" or next;
sub LINE_MAX () { 512 } # does RFC 3501 have a limit like RFC 977?
my %FETCH_NEED_BLOB = ( # for future optimization
- 'BODY.PEEK[HEADER]' => 1,
- 'BODY.PEEK[TEXT]' => 1,
- 'BODY.PEEK[]' => 1,
'BODY[HEADER]' => 1,
'BODY[TEXT]' => 1,
'BODY[]' => 1,
$FETCH_ATT{$att} = \%h;
}
+my $valid_range = '[0-9]+|[0-9]+:[0-9]+|[0-9]+:\*';
+$valid_range = qr/\A(?:$valid_range)(?:,(?:$valid_range))*\z/;
+
sub greet ($) {
my ($self) = @_;
my $capa = capa($self);
login_success($self, $tag);
}
+sub cmd_close ($$) {
+ my ($self, $tag) = @_;
+ delete $self->{ibx} ? "$tag OK Close done\r\n"
+ : "$tag BAD No mailbox\r\n";
+}
+
sub cmd_logout ($$) {
my ($self, $tag) = @_;
delete $self->{logged_in};
join('', @hold);
}
-sub uid_fetch_cb { # called by git->cat_async
+sub dummy_message ($$) {
+ my ($seqno, $ibx) = @_;
+ my $ret = <<EOF;
+From: nobody\@localhost\r
+To: nobody\@localhost\r
+Date: Thu, 01 Jan 1970 00:00:00 +0000\r
+Message-ID: <dummy-$seqno\@$ibx->{newsgroup}>\r
+Subject: dummy message #$seqno\r
+\r
+You're seeing this message because your IMAP client didn't use UIDs.\r
+The message which used to use this sequence number was likely spam\r
+and removed by the administrator.\r
+EOF
+ \$ret;
+}
+
+sub requeue_once ($) {
+ my ($self) = @_;
+ # COMPRESS users all share the same DEFLATE context.
+ # Flush it here to ensure clients don't see
+ # each other's data
+ $self->zflush;
+
+ # no recursion, schedule another call ASAP,
+ # but only after all pending writes are done.
+ # autovivify wbuf:
+ my $new_size = push(@{$self->{wbuf}}, \&long_step);
+
+ # wbuf may be populated by $cb, no need to rearm if so:
+ $self->requeue if $new_size == 1;
+}
+
+sub uid_fetch_cb { # called by git->cat_async via git_async_cat
my ($bref, $oid, $type, $size, $fetch_m_arg) = @_;
- my ($self, undef, $ibx, undef, undef, $msgs, $want) = @$fetch_m_arg;
+ my ($self, undef, $ibx, $msgs, undef, $want) = @$fetch_m_arg;
my $smsg = shift @$msgs or die 'BUG: no smsg';
- $smsg->{blob} eq $oid or die "BUG: $smsg->{blob} != $oid";
+ if (!defined($oid)) {
+ # it's possible to have TOCTOU if an admin runs
+ # public-inbox-(edit|purge), just move onto the next message
+ return requeue_once($self) unless defined $want->{-seqno};
+ $bref = dummy_message($smsg->{num}, $ibx);
+ } else {
+ $smsg->{blob} eq $oid or die "BUG: $smsg->{blob} != $oid";
+ }
+
$$bref =~ s/(?<!\r)\n/\r\n/sg; # make strict clients happy
# fixup old bug from import (pre-a0c07cba0e5d8b6a)
$want->{INTERNALDATE} and
$self->msg_more(' INTERNALDATE "'.$smsg->internaldate.'"');
$want->{FLAGS} and $self->msg_more(' FLAGS ()');
- for ('RFC822', 'BODY[]', 'BODY.PEEK[]') {
- next unless $want->{$_};
+ for ('RFC822', 'BODY[]') {
+ $want->{$_} or next;
$self->msg_more(" $_ {".length($$bref)."}\r\n");
$self->msg_more($$bref);
}
$want->{ENVELOPE} and
$self->msg_more(' ENVELOPE '.eml_envelope($eml));
- for my $f ('RFC822.HEADER', 'BODY[HEADER]', 'BODY.PEEK[HEADER]') {
- next unless $want->{$f};
- $self->msg_more(" $f {".length(${$eml->{hdr}})."}\r\n");
+ for ('RFC822.HEADER', 'BODY[HEADER]') {
+ $want->{$_} or next;
+ $self->msg_more(" $_ {".length(${$eml->{hdr}})."}\r\n");
$self->msg_more(${$eml->{hdr}});
}
- for my $f ('RFC822.TEXT', 'BODY[TEXT]') {
- next unless $want->{$f};
- $self->msg_more(" $f {".length($$bref)."}\r\n");
+ for ('RFC822.TEXT', 'BODY[TEXT]') {
+ $want->{$_} or next;
+ $self->msg_more(" $_ {".length($$bref)."}\r\n");
$self->msg_more($$bref);
}
$want->{BODYSTRUCTURE} and
$self->msg_more(' BODYSTRUCTURE '.fetch_body($eml, 1));
$want->{BODY} and
$self->msg_more(' BODY '.fetch_body($eml));
- if (my $partial = delete $want->{-partial}) {
+ if (my $partial = $want->{-partial}) {
partial_emit($self, $partial, $eml);
}
$self->msg_more(")\r\n");
+ requeue_once($self);
+}
+
+sub range_step ($$) {
+ my ($ibx, $range_csv) = @_;
+ my ($beg, $end, $range);
+ if ($$range_csv =~ s/\A([^,]+),//) {
+ $range = $1;
+ } else {
+ $range = $$range_csv;
+ $$range_csv = undef;
+ }
+ if ($range =~ /\A([0-9]+):([0-9]+)\z/) {
+ ($beg, $end) = ($1, $2);
+ } elsif ($range =~ /\A([0-9]+):\*\z/) {
+ ($beg, $end) = ($1, $ibx->mm->max // 0);
+ } elsif ($range =~ /\A[0-9]+\z/) {
+ $beg = $end = $range;
+ } else {
+ return 'BAD fetch range';
+ }
+ [ $beg, $end, $$range_csv ];
+}
+
+sub refill_range ($$$) {
+ my ($ibx, $msgs, $range_info) = @_;
+ my ($beg, $end, $range_csv) = @$range_info;
+ if (scalar(@$msgs = @{$ibx->over->query_xover($beg, $end)})) {
+ $range_info->[0] = $msgs->[-1]->{num} + 1;
+ return;
+ }
+ return 'OK Fetch done' if !$range_csv;
+ my $next_range = range_step($ibx, \$range_csv);
+ return $next_range if !ref($next_range); # error
+ @$range_info = @$next_range;
+ undef; # keep looping
}
sub uid_fetch_m { # long_response
- my ($self, $tag, $ibx, $beg, $end, $msgs, $want) = @_;
- if (!@$msgs) { # refill
- @$msgs = @{$ibx->over->query_xover($$beg, $end)};
- if (!@$msgs) {
- $self->write(\"$tag OK Fetch done\r\n");
+ my ($self, $tag, $ibx, $msgs, $range_info, $want) = @_;
+ while (!@$msgs) { # rare
+ if (my $end = refill_range($ibx, $msgs, $range_info)) {
+ $self->write(\"$tag $end\r\n");
return;
}
- $$beg = $msgs->[-1]->{num} + 1;
}
- my $git = $ibx->git;
- $git->cat_async_begin; # TODO: actually make async
- $git->cat_async($msgs->[0]->{blob}, \&uid_fetch_cb, \@_);
- $git->cat_async_wait;
- 1;
+ git_async_cat($ibx->git, $msgs->[0]->{blob}, \&uid_fetch_cb, \@_);
}
sub cmd_status ($$$;@) {
\(join('', @$l, "$tag OK List done\r\n"));
}
+sub cmd_lsub ($$$$) {
+ my (undef, $tag) = @_; # same args as cmd_list
+ "$tag OK Lsub done\r\n";
+}
+
sub eml_index_offs_i { # PublicInbox::Eml::each_part callback
my ($p, $all) = @_;
my ($eml, undef, $idx) = @$p;
# BODY[($SECTION_IDX.)?HEADER.FIELDS.NOT ($HDRS)]<$offset.$bytes>
sub partial_hdr_not {
- my ($eml, $section_idx, $hdrs) = @_;
+ my ($eml, $section_idx, $hdrs_re) = @_;
if (defined $section_idx) {
$eml = eml_body_idx($eml, $section_idx) or return;
}
my $str = $eml->header_obj->as_string;
- my $re = hdrs_regexp($hdrs);
- $str =~ s/$re//g;
+ $str =~ s/$hdrs_re//g;
$str .= "\r\n";
}
# BODY[($SECTION_IDX.)?HEADER.FIELDS ($HDRS)]<$offset.$bytes>
sub partial_hdr_get {
- my ($eml, $section_idx, $hdrs) = @_;
+ my ($eml, $section_idx, $hdrs_re) = @_;
if (defined $section_idx) {
$eml = eml_body_idx($eml, $section_idx) or return;
}
my $str = $eml->header_obj->as_string;
- my $re = hdrs_regexp($hdrs);
- join('', ($str =~ m/($re)/g), "\r\n");
+ join('', ($str =~ m/($hdrs_re)/g), "\r\n");
}
sub partial_prepare ($$$) {
# recombine [ "BODY[1.HEADER.FIELDS", "(foo", "bar)]" ]
# back to: "BODY[1.HEADER.FIELDS (foo bar)]"
- return unless $att =~ /\ABODY(?:\.PEEK)?\[/s;
+ return unless $att =~ /\ABODY\[/s;
until (rindex($att, ']') >= 0) {
my $next = shift @$want or return;
$att .= ' ' . uc($next);
}
- if ($att =~ /\ABODY(?:\.PEEK)?\[
- ([0-9]+(?:\.[0-9]+)*)? # 1 - section_idx
- (?:\.(HEADER|MIME|TEXT))? # 2 - section_name
+ if ($att =~ /\ABODY\[([0-9]+(?:\.[0-9]+)*)? # 1 - section_idx
+ (?:\.(HEADER|MIME|TEXT))? # 2 - section_name
\](?:<([0-9]+)(?:\.([0-9]+))?>)?\z/sx) { # 3, 4
$partial->{$att} = [ \&partial_body, $1, $2, $3, $4 ];
- } elsif ($att =~ /\ABODY(?:\.PEEK)?\[
- (?:([0-9]+(?:\.[0-9]+)*)\.)? # 1 - section_idx
+ } elsif ($att =~ /\ABODY\[(?:([0-9]+(?:\.[0-9]+)*)\.)? # 1 - section_idx
(?:HEADER\.FIELDS(\.NOT)?)\x20 # 2
\(([A-Z0-9\-\x20]+)\) # 3 - hdrs
\](?:<([0-9]+)(?:\.([0-9]+))?>)?\z/sx) { # 4 5
- $partial->{$att} = [ $2 ? \&partial_hdr_not
- : \&partial_hdr_get,
- $1, $3, $4, $5 ];
+ my $tmp = $partial->{$att} = [ $2 ? \&partial_hdr_not
+ : \&partial_hdr_get,
+ $1, undef, $4, $5 ];
+ $tmp->[2] = hdrs_regexp($3);
} else {
undef;
}
sub partial_emit ($$$) {
my ($self, $partial, $eml) = @_;
- for my $k (sort keys %$partial) {
- my ($cb, @args) = @{$partial->{$k}};
+ for (@$partial) {
+ my ($k, $cb, @args) = @$_;
my ($offset, $len) = splice(@args, -2);
# $cb is partial_body|partial_hdr_get|partial_hdr_not
my $str = $cb->($eml, @args) // '';
}
}
-sub cmd_uid_fetch ($$$;@) {
- my ($self, $tag, $range, @want) = @_;
+sub fetch_common ($$$$) {
+ my ($self, $tag, $range_csv, $want) = @_;
my $ibx = $self->{ibx} or return "$tag BAD No mailbox selected\r\n";
- if ($want[0] =~ s/\A\(//s) {
- $want[-1] =~ s/\)\z//s or return "$tag BAD no rparen\r\n";
+ if ($want->[0] =~ s/\A\(//s) {
+ $want->[-1] =~ s/\)\z//s or return "$tag BAD no rparen\r\n";
}
my (%partial, %want);
- while (defined(my $att = shift @want)) {
+ while (defined(my $att = shift @$want)) {
$att = uc($att);
+ $att =~ s/\ABODY\.PEEK\[/BODY\[/; # we're read-only
my $x = $FETCH_ATT{$att};
if ($x) {
%want = (%want, %$x);
- } elsif (!partial_prepare(\%partial, \@want, $att)) {
+ } elsif (!partial_prepare(\%partial, $want, $att)) {
return "$tag BAD param: $att\r\n";
}
}
- $want{-partial} = \%partial if scalar keys %partial;
- my ($beg, $end);
- my $msgs = [];
- if ($range =~ /\A([0-9]+):([0-9]+)\z/s) {
- ($beg, $end) = ($1, $2);
- } elsif ($range =~ /\A([0-9]+):\*\z/s) {
- ($beg, $end) = ($1, $ibx->mm->max // 0);
- } elsif ($range =~ /\A[0-9]+\z/) {
- my $smsg = $ibx->over->get_art($range) or
- return "$tag OK Fetch done\r\n"; # really OK(!)
- push @$msgs, $smsg;
- ($beg, $end) = ($range, 0);
- } else {
- return "$tag BAD fetch range\r\n";
+
+ # stabilize partial order for consistency and ease-of-debugging:
+ if (scalar keys %partial) {
+ $want{-partial} = [ map {;
+ [ $_, @{$partial{$_}} ]
+ } sort keys %partial ];
}
- long_response($self, \&uid_fetch_m, $tag, $ibx,
- \$beg, $end, $msgs, \%want);
+ $range_csv = 'bad' if $range_csv !~ $valid_range;
+ my $range_info = range_step($ibx, \$range_csv);
+ return "$tag $range_info\r\n" if !ref($range_info);
+ [ $tag, $ibx, [], $range_info, \%want ];
+}
+
+sub cmd_uid_fetch ($$$;@) {
+ my ($self, $tag, $range_csv, @want) = @_;
+ my $args = fetch_common($self, $tag, $range_csv, \@want);
+ ref($args) eq 'ARRAY' ?
+ long_response($self, \&uid_fetch_m, @$args) :
+ $args; # error
+}
+
+sub seq_fetch_m { # long_response
+ my ($self, $tag, $ibx, $msgs, $range_info, $want) = @_;
+ while (!@$msgs) { # rare
+ if (my $end = refill_range($ibx, $msgs, $range_info)) {
+ $self->write(\"$tag $end\r\n");
+ return;
+ }
+ }
+ my $seq = $want->{-seqno}++;
+ my $cur_num = $msgs->[0]->{num};
+ if ($cur_num == $seq) { # as expected
+ git_async_cat($ibx->git, $msgs->[0]->{blob},
+ \&uid_fetch_cb, \@_);
+ } elsif ($cur_num > $seq) {
+ # send dummy messages until $seq catches up to $cur_num
+ my $smsg = bless { num => $seq, ts => 0 }, 'PublicInbox::Smsg';
+ unshift @$msgs, $smsg;
+ my $bref = dummy_message($seq, $ibx);
+ uid_fetch_cb($bref, undef, undef, undef, \@_);
+ $smsg; # blessed response since uid_fetch_cb requeues
+ } else { # should not happen
+ die "BUG: cur_num=$cur_num < seq=$seq";
+ }
+}
+
+sub cmd_fetch ($$$;@) {
+ my ($self, $tag, $range_csv, @want) = @_;
+ my $args = fetch_common($self, $tag, $range_csv, \@want);
+ ref($args) eq 'ARRAY' ? do {
+ my $want = $args->[-1];
+ $want->{-seqno} = $args->[3]->[0]; # $beg == $range_info->[0];
+ long_response($self, \&seq_fetch_m, @$args)
+ } : $args; # error
}
sub uid_search_all { # long_response
} elsif ($more) { # $self->{wbuf}:
$self->update_idle_time;
- # COMPRESS users all share the same DEFLATE context.
- # Flush it here to ensure clients don't see
- # each other's data
- $self->zflush;
-
- # no recursion, schedule another call ASAP, but only after
- # all pending writes are done. autovivify wbuf:
- my $new_size = push(@{$self->{wbuf}}, \&long_step);
-
- # wbuf may be populated by $cb, no need to rearm if so:
- $self->requeue if $new_size == 1;
+ # control passed to $more may be a GitAsyncCat object
+ requeue_once($self) if !ref($more);
} else { # all done!
delete $self->{long_cb};
my $elapsed = now() - $t0;
sub event_step {
my ($self) = @_;
- return unless $self->flush_write && $self->{sock};
+ return unless $self->flush_write && $self->{sock} && !$self->{long_cb};
$self->update_idle_time;
# only read more requests if we've drained the write buffer,
# otherwise we can be buffering infinitely w/o backpressure
- my $rbuf = $self->{rbuf} // (\(my $x = ''));
- my $r = 1;
-
- if (index($$rbuf, "\n") < 0) {
- my $off = length($$rbuf);
- $r = $self->do_read($rbuf, LINE_MAX, $off) or return;
- }
- while ($r > 0 && $$rbuf =~ s/\A[ \t]*([^\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 $pending = $self->{wbuf} ? ' pending' : '';
- out($self, "[$fd] %s - %0.6f$pending", $line, now() - $t0);
- }
+ my $rbuf = $self->{rbuf} // \(my $x = '');
+ my $line = index($$rbuf, "\n");
+ while ($line < 0) {
+ return $self->close if length($$rbuf) >= LINE_MAX;
+ $self->do_read($rbuf, LINE_MAX, length($$rbuf)) or return;
+ $line = index($$rbuf, "\n");
+ }
+ $line = substr($$rbuf, 0, $line + 1, '');
+ $line =~ s/\r?\n\z//s;
+ return $self->close if $line =~ /[[:cntrl:]]/s;
+ my $t0 = now();
+ my $fd = fileno($self->{sock});
+ my $r = eval { process_line($self, $line) };
+ my $pending = $self->{wbuf} ? ' pending' : '';
+ out($self, "[$fd] %s - %0.6f$pending - $r", $line, now() - $t0);
return $self->close if $r < 0;
- my $len = length($$rbuf);
- return $self->close if ($len >= LINE_MAX);
$self->rbuf_idle($rbuf);
$self->update_idle_time;
# maybe there's more pipelined data, or we'll have
# to register it for socket-readiness notifications
- $self->requeue unless $self->{wbuf};
+ $self->requeue unless $pending;
}
sub compressed { undef }