# Copyright (C) 2016 all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+#
+# ref: https://cr.yp.to/proto/maildir.html
+# http://wiki2.dovecot.org/MailboxFormat/Maildir
package PublicInbox::WatchMaildir;
use strict;
use warnings;
-use Email::MIME;
+use PublicInbox::MIME;
use Email::MIME::ContentType;
$Email::MIME::ContentType::STRICT_PARAMS = 0; # user input is imperfect
use PublicInbox::Git;
sub new {
my ($class, $config) = @_;
- my (%mdmap, @mdir);
- my $k = 'publicinboxlearn.watchspam';
- if (my $spamdir = $config->{$k}) {
- if ($spamdir =~ s/\Amaildir://) {
- $spamdir =~ s!/+\z!!;
- # skip "new", no MUA has seen it, yet.
- my $cur = "$spamdir/cur";
- push @mdir, $cur;
- $mdmap{$cur} = 'watchspam';
+ my (%mdmap, @mdir, $spamc);
+
+ # "publicinboxwatch" is the documented namespace
+ # "publicinboxlearn" is legacy but may be supported
+ # indefinitely...
+ foreach my $pfx (qw(publicinboxwatch publicinboxlearn)) {
+ my $k = "$pfx.watchspam";
+ if (my $spamdir = $config->{$k}) {
+ if ($spamdir =~ s/\Amaildir://) {
+ $spamdir =~ s!/+\z!!;
+ # skip "new", no MUA has seen it, yet.
+ my $cur = "$spamdir/cur";
+ push @mdir, $cur;
+ $mdmap{$cur} = 'watchspam';
+ } else {
+ warn "unsupported $k=$spamdir\n";
+ }
+ }
+ }
+
+ my $k = 'publicinboxwatch.spamcheck';
+ my $spamcheck = $config->{$k};
+ if ($spamcheck) {
+ if ($spamcheck eq 'spamc') {
+ $spamcheck = 'PublicInbox::Spamcheck::Spamc';
+ }
+ if ($spamcheck =~ /::/) {
+ eval "require $spamcheck";
+ $spamcheck = _spamcheck_cb($spamcheck->new);
} else {
- warn "unsupported $k=$spamdir\n";
+ warn "unsupported $k=$spamcheck\n";
+ $spamcheck = undef;
}
}
foreach $k (keys %$config) {
my $mdre = join('|', map { quotemeta($_) } @mdir);
$mdre = qr!\A($mdre)/!;
bless {
+ spamcheck => $spamcheck,
mdmap => \%mdmap,
mdir => \@mdir,
mdre => $mdre,
+ config => $config,
importers => {},
}, $class;
}
_done_for_now($self);
}
-sub _check_spam {
+sub _remove_spam {
my ($self, $path) = @_;
+ # path must be marked as (S)een
+ $path =~ /:2,[A-R]*S[T-Za-z]*\z/ or return;
my $mime = _path_to_mime($path) or return;
_force_mid($mime);
- foreach my $inbox (values %{$self->{mdmap}}) {
- next unless ref $inbox;
- my $im = _importer_for($self, $inbox);
- $im->remove($mime);
- if (my $scrub = _scrubber_for($inbox)) {
- my $scrubbed = $scrub->scrub($mime) or next;
- $im->remove($scrubbed);
- }
- }
+ $self->{config}->each_inbox(sub {
+ my ($ibx) = @_;
+ eval {
+ my $im = _importer_for($self, $ibx);
+ $im->remove($mime);
+ if (my $scrub = _scrubber_for($ibx)) {
+ my $scrubbed = $scrub->scrub($mime) or return;
+ $im->remove($scrubbed);
+ }
+ };
+ warn "error removing spam at $path from $ibx->{name}\n" if $@;
+ })
}
# used to hash the relevant portions of a message when there are conflicts
sub _try_path {
my ($self, $path) = @_;
+ my @p = split(m!/+!, $path);
+ return if $p[-1] !~ /\A[a-zA-Z0-9][\w:,=\.]+\z/;
+ if ($p[-1] =~ /:2,([A-Z]+)\z/i) {
+ my $flags = $1;
+ return if $flags =~ /[DT]/; # no [D]rafts or [T]rashed mail
+ }
+ return unless -f $path;
if ($path !~ $self->{mdre}) {
warn "unrecognized path: $path\n";
return;
return;
}
if (!ref($inbox) && $inbox eq 'watchspam') {
- return _check_spam($self, $path);
+ return _remove_spam($self, $path);
}
my $im = _importer_for($self, $inbox);
my $mime = _path_to_mime($path) or return;
}
_force_mid($mime);
- $im->add($mime);
+ $im->add($mime, $self->{spamcheck});
}
sub watch {
my $cb = sub { _try_fsn_paths($self, \@_) };
my $mdir = $self->{mdir};
+ # lazy load here, we may support watching via IMAP IDLE
+ # in the future...
require Filesys::Notify::Simple;
my $watcher = Filesys::Notify::Simple->new($mdir);
$watcher->wait($cb) while (1);
next;
}
while (my $fn = readdir($dh)) {
- next unless $fn =~ /\A[a-zA-Z0-9][\w:,=\.]+\z/;
- $fn = "$dir/$fn";
- if (-f $fn) {
- _try_path($self, $fn);
- } else {
- warn "not a file: $fn\n";
- }
+ _try_path($self, "$dir/$fn");
}
closedir $dh;
}
local $/;
my $str = <$fh>;
$str or return;
- return Email::MIME->new(\$str);
+ return PublicInbox::MIME->new(\$str);
} elsif ($!{ENOENT}) {
return;
} else {
my $git = $inbox->git;
my $name = $inbox->{name};
my $addr = $inbox->{-primary_address};
- PublicInbox::Import->new($git, $name, $addr);
+ PublicInbox::Import->new($git, $name, $addr, $inbox);
};
- $self->{importers}->{"$im"} = $im;
+
+ my $importers = $self->{importers};
+ if (scalar(keys(%$importers)) > 2) {
+ delete $importers->{"$im"};
+ _done_for_now($self);
+ }
+
+ $importers->{"$im"} = $im;
}
sub _scrubber_for {
my ($inbox) = @_;
my $f = $inbox->{filter};
if ($f && $f =~ /::/) {
+ my @args;
+ # basic line splitting, only
+ # Perhaps we can have proper quote splitting one day...
+ ($f, @args) = split(/\s+/, $f) if $f =~ /\s+/;
+
eval "require $f";
if ($@) {
warn $@;
} else {
- return $f->new;
+ # e.g: PublicInbox::Filter::Vger->new(@args)
+ return $f->new(@args);
}
}
undef;
}
+sub _spamcheck_cb {
+ my ($sc) = @_;
+ sub {
+ my ($mime) = @_;
+ my $tmp = '';
+ if ($sc->spamcheck($mime, \$tmp)) {
+ return PublicInbox::MIME->new(\$tmp);
+ }
+ warn $mime->header('Message-ID')." failed spam check\n";
+ undef;
+ }
+}
+
1;