Merge branch 'thread-view-skel'

[public-inbox.git] / lib / PublicInbox / WatchMaildir.pm

diff --git a/lib/PublicInbox/WatchMaildir.pm b/lib/PublicInbox/WatchMaildir.pm
index cb64f899fb27e277b8a6032100f18683c35ff5e3..145b363689ee7aa42c292f99c15798a404b7ff24 100644 (file)
--- a/lib/PublicInbox/WatchMaildir.pm
+++ b/lib/PublicInbox/WatchMaildir.pm
@@ -1,5 +1,8 @@
 # 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;
@@ -13,7 +16,9 @@ use PublicInbox::Spawn qw(spawn);
 
 sub new {
        my ($class, $config) = @_;
-       my (%mdmap, @mdir);
+       my (%mdmap, @mdir, $spamc);
+
+       # XXX is "publicinboxlearn" really a good namespace for this?
        my $k = 'publicinboxlearn.watchspam';
        if (my $spamdir = $config->{$k}) {
                if ($spamdir =~ s/\Amaildir://) {
@@ -26,6 +31,21 @@ sub new {
                        warn "unsupported $k=$spamdir\n";
                }
        }
+
+       $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 $k=$spamcheck\n";
+                       $spamcheck = undef;
+               }
+       }
        foreach $k (keys %$config) {
                $k =~ /\Apublicinbox\.([^\.]+)\.watch\z/ or next;
                my $name = $1;
@@ -52,6 +72,7 @@ sub new {
        my $mdre = join('|', map { quotemeta($_) } @mdir);
        $mdre = qr!\A($mdre)/!;
        bless {
+               spamcheck => $spamcheck,
                mdmap => \%mdmap,
                mdir => \@mdir,
                mdre => $mdre,
@@ -69,8 +90,9 @@ sub _try_fsn_paths {
        _done_for_now($self);
 }
 
-sub _check_spam {
+sub _remove_spam {
        my ($self, $path) = @_;
+       $path =~ /:2,[A-R]*S[T-Z]*\z/i or return;
        my $mime = _path_to_mime($path) or return;
        _force_mid($mime);
        foreach my $inbox (values %{$self->{mdmap}}) {
@@ -107,6 +129,13 @@ sub _force_mid {
 
 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: $path\n";
                return;
@@ -117,7 +146,7 @@ sub _try_path {
                return;
        }
        if (!ref($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;
@@ -132,7 +161,7 @@ sub _try_path {
        }
 
        _force_mid($mime);
-       $im->add($mime);
+       $im->add($mime, $self->{spamcheck});
 }
 
 sub watch {
@@ -140,6 +169,8 @@ 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);
@@ -155,13 +186,7 @@ sub scan {
                        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;
        }
@@ -208,4 +233,17 @@ sub _scrubber_for {
        undef;
 }
 
+sub _spamcheck_cb {
+       my ($sc) = @_;
+       sub {
+               my ($mime) = @_;
+               my $tmp = '';
+               if ($sc->spamcheck($mime, \$tmp)) {
+                       return Email::MIME->new(\$tmp);
+               }
+               warn $mime->header('Message-ID')." failed spam check\n";
+               undef;
+       }
+}
+
 1;