watch_maildir: implement optional spam checking

Mailing lists I watch and mirror may not have the best spam
filtering, and an extra layer should not hurt.

2 files changed, 37 insertions, 3 deletions

diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 5ffc26ef..27f36a7e 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -140,7 +140,7 @@ sub remove {
 
 # returns undef on duplicate
 sub add {
-        my ($self, $mime) = @_; # mime = Email::MIME
+        my ($self, $mime, $check_cb) = @_; # mime = Email::MIME
 
         my $from = $mime->header('From');
         my ($email) = ($from =~ /([^<\s]+\@[^>\s]+)/g);
@@ -170,6 +170,10 @@ sub add {
 
         # kill potentially confusing/misleading headers
         $mime->header_set($_) for qw(bytes lines content-length status);
+        if ($check_cb) {
+                $mime = $check_cb->($mime) or return;
+        }
+
         $mime = $mime->as_string;
         my $blob = $self->{mark}++;
         print $w "blob\nmark :$blob\ndata ", length($mime), "\n" or wfail;
diff --git a/lib/PublicInbox/WatchMaildir.pm b/lib/PublicInbox/WatchMaildir.pm
index c1fe81ec..72bd3d08 100644
--- a/lib/PublicInbox/WatchMaildir.pm
+++ b/lib/PublicInbox/WatchMaildir.pm
@@ -13,7 +13,9 @@ 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 +28,21 @@ 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 +69,7 @@ sub new {
         my $mdre = join('|', map { quotemeta($_) } @mdir);
         $mdre = qr!\A($mdre)/!;
         bless {
+                spamcheck => $spamcheck,
                 mdmap => \%mdmap,
                 mdir => \@mdir,
                 mdre => $mdre,
@@ -136,7 +154,7 @@ sub _try_path {
         }
 
         _force_mid($mime);
-        $im->add($mime);
+        $im->add($mime, $self->{spamcheck});
 }
 
 sub watch {
@@ -208,4 +226,16 @@ sub _scrubber_for {
         undef;
 }
 
+sub _spamcheck_cb {
+        my ($sc) = @_;
+        sub {
+                my ($mime) = @_;
+                my $tmp = '';
+                if ($sc->spamcheck($mime, \$tmp)) {
+                        return Email::MIME->new(\$tmp);
+                }
+                undef;
+        }
+}
+
 1;