X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FWatchMaildir.pm;h=0e2a6d2c3c5d6f6ea50580b5ad8e25eb9314278f;hb=f9b70eb6ebbf96c2fe79ab2738ea4954c5a124f3;hp=145b363689ee7aa42c292f99c15798a404b7ff24;hpb=4838cef47d8978b1928a89cfdcc4a1e53d5201e6;p=public-inbox.git

diff --git a/lib/PublicInbox/WatchMaildir.pm b/lib/PublicInbox/WatchMaildir.pm
index 145b3636..0e2a6d2c 100644
--- a/lib/PublicInbox/WatchMaildir.pm
+++ b/lib/PublicInbox/WatchMaildir.pm
@@ -6,33 +6,39 @@
 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;
 use PublicInbox::Import;
 use PublicInbox::MDA;
 use PublicInbox::Spawn qw(spawn);
+use File::Temp qw//;
 
 sub new {
 	my ($class, $config) = @_;
-	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://) {
-			$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 (%mdmap, @mdir, $spamc, $spamdir);
+
+	# "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 $dir = $config->{$k}) {
+			if ($dir =~ s/\Amaildir://) {
+				$dir =~ s!/+\z!!;
+				# skip "new", no MUA has seen it, yet.
+				my $cur = "$dir/cur";
+				$spamdir = $cur;
+				push @mdir, $cur;
+				$mdmap{$cur} = 'watchspam';
+			} else {
+				warn "unsupported $k=$dir\n";
+			}
 		}
 	}
 
-	$k = 'publicinboxwatch.spamcheck';
+	my $k = 'publicinboxwatch.spamcheck';
 	my $spamcheck = $config->{$k};
 	if ($spamcheck) {
 		if ($spamcheck eq 'spamc') {
@@ -73,37 +79,66 @@ sub new {
 	$mdre = qr!\A($mdre)/!;
 	bless {
 		spamcheck => $spamcheck,
+		spamdir => $spamdir,
 		mdmap => \%mdmap,
 		mdir => \@mdir,
 		mdre => $mdre,
+		config => $config,
 		importers => {},
+		opendirs => {}, # dirname => dirhandle (in progress scans)
 	}, $class;
 }
 
 sub _done_for_now {
-	$_->done foreach values %{$_[0]->{importers}};
+	my ($self) = @_;
+	my $opendirs = $self->{opendirs};
+
+	# spamdir scanning means every importer remains open
+	my $spamdir = $self->{spamdir};
+	return if defined($spamdir) && $opendirs->{$spamdir};
+
+	foreach my $im (values %{$self->{importers}}) {
+		# not done if we're scanning
+		next if $opendirs->{$im->{git}->{git_dir}};
+		$im->done;
+	}
 }
 
 sub _try_fsn_paths {
-	my ($self, $paths) = @_;
-	_try_path($self, $_->{path}) foreach @$paths;
+	my ($self, $scan_re, $paths) = @_;
+	foreach (@$paths) {
+		my $path = $_->{path};
+		if ($path =~ $scan_re) {
+			scan($self, $path);
+		} else {
+			_try_path($self, $path);
+		}
+	}
 	_done_for_now($self);
 }
 
 sub _remove_spam {
 	my ($self, $path) = @_;
-	$path =~ /:2,[A-R]*S[T-Z]*\z/i or return;
+	# 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;
+				$scrubbed == 100 and return;
+				$im->remove($scrubbed);
+			}
+		};
+		if ($@) {
+			warn "error removing spam at: ", $path,
+				" from ", $ibx->{name}, ': ', $@, "\n";
 		}
-	}
+	})
 }
 
 # used to hash the relevant portions of a message when there are conflicts
@@ -157,40 +192,74 @@ sub _try_path {
 		return unless ($v && $v =~ $wm->[1]);
 	}
 	if (my $scrub = _scrubber_for($inbox)) {
-		$mime = $scrub->scrub($mime) or return;
+		my $ret = $scrub->scrub($mime) or return;
+		$ret == 100 and return;
+		$mime = $ret;
 	}
 
 	_force_mid($mime);
 	$im->add($mime, $self->{spamcheck});
 }
 
+sub quit { $_[0]->{quit} = 1 }
+
 sub watch {
 	my ($self) = @_;
-	my $cb = sub { _try_fsn_paths($self, \@_) };
-	my $mdir = $self->{mdir};
+	my $scan = File::Temp->newdir("public-inbox-watch.$$.scan.XXXXXX",
+					TMPDIR => 1);
+	my $scandir = $self->{scandir} = $scan->dirname;
+	my $re = qr!\A$scandir/!;
+	my $cb = sub { _try_fsn_paths($self, $re, \@_) };
 
 	# 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);
+	my $fsn = Filesys::Notify::Simple->new([@{$self->{mdir}}, $scandir]);
+	$fsn->wait($cb) until ($self->{quit});
+}
+
+sub trigger_scan {
+	my ($self, $base) = @_;
+	my $dir = $self->{scandir} or die "not watch-ing, yet\n";
+	open my $fh, '>', "$dir/$base" or die "open $dir/$base failed: $!\n";
+	close $fh or die "close $dir/$base failed: $!\n";
 }
 
 sub scan {
-	my ($self) = @_;
-	my $mdir = $self->{mdir};
-	foreach my $dir (@$mdir) {
-		my $ok = opendir(my $dh, $dir);
-		unless ($ok) {
-			warn "failed to open $dir: $!\n";
-			next;
-		}
+	my ($self, $path) = @_;
+	my $max = 10;
+	my $opendirs = $self->{opendirs};
+	my @dirnames = keys %$opendirs;
+	foreach my $dir (@dirnames) {
+		my $dh = delete $opendirs->{$dir};
+		my $n = $max;
 		while (my $fn = readdir($dh)) {
 			_try_path($self, "$dir/$fn");
+			last if --$n < 0;
 		}
-		closedir $dh;
+		$opendirs->{$dir} = $dh if $n < 0;
+	}
+	if ($path =~ /full\z/) {
+		foreach my $dir (@{$self->{mdir}}) {
+			next if $opendirs->{$dir}; # already in progress
+			my $ok = opendir(my $dh, $dir);
+			unless ($ok) {
+				warn "failed to open $dir: $!\n";
+				next;
+			}
+			my $n = $max;
+			while (my $fn = readdir($dh)) {
+				_try_path($self, "$dir/$fn");
+				last if --$n < 0;
+			}
+			$opendirs->{$dir} = $dh if $n < 0;
+		}
+	}
+	if (keys %$opendirs) { # do we have more work to do?
+		trigger_scan($self, 'cont');
+	} else {
+		_done_for_now($self);
 	}
-	_done_for_now($self);
 }
 
 sub _path_to_mime {
@@ -199,7 +268,7 @@ sub _path_to_mime {
 		local $/;
 		my $str = <$fh>;
 		$str or return;
-		return Email::MIME->new(\$str);
+		return PublicInbox::MIME->new(\$str);
 	} elsif ($!{ENOENT}) {
 		return;
 	} else {
@@ -214,20 +283,33 @@ sub _importer_for {
 		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 = (-inbox => $inbox);
+		# basic line splitting, only
+		# Perhaps we can have proper quote splitting one day...
+		($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 $f->new(@args);
 		}
 	}
 	undef;
@@ -239,7 +321,7 @@ sub _spamcheck_cb {
 		my ($mime) = @_;
 		my $tmp = '';
 		if ($sc->spamcheck($mime, \$tmp)) {
-			return Email::MIME->new(\$tmp);
+			return PublicInbox::MIME->new(\$tmp);
 		}
 		warn $mime->header('Message-ID')." failed spam check\n";
 		undef;