watch_maildir: spam removal support

We can support spam removal by watching a special "spam"
Maildir, too.  We can run public-inbox-learn as a separate
step, and that command will be improved to support
auto-learning, too.

1 files changed, 45 insertions, 1 deletions

diff --git a/t/watch_maildir.t b/t/watch_maildir.t
index 6564a866..8a2c934a 100644
--- a/t/watch_maildir.t
+++ b/t/watch_maildir.t
@@ -8,6 +8,7 @@ use PublicInbox::Config;
 my $tmpdir = tempdir('watch_maildir-XXXXXX', TMPDIR => 1, CLEANUP => 1);
 my $git_dir = "$tmpdir/test.git";
 my $maildir = "$tmpdir/md";
+my $spamdir = "$tmpdir/spam";
 use_ok 'PublicInbox::WatchMaildir';
 use_ok 'PublicInbox::Emergency';
 my $cfgpfx = "publicinbox.test";
@@ -21,14 +22,17 @@ Subject: spam
 Message-Id: <a\@b.com>
 Date: Sat, 18 Jun 2016 00:00:00 +0000
 
-msg
+something
 EOF
 PublicInbox::Emergency->new($maildir)->prepare(\$msg);
+my $sem = PublicInbox::Emergency->new($spamdir); # create dirs
 
 my $config = PublicInbox::Config->new({
         "$cfgpfx.address" => $addr,
         "$cfgpfx.mainrepo" => $git_dir,
         "$cfgpfx.watch" => "maildir:$maildir",
+        "$cfgpfx.filter" => 'PublicInbox::Filter::Vger',
+        "publicinboxlearn.watchspam" => "maildir:$spamdir",
 });
 
 PublicInbox::WatchMaildir->new($config)->scan;
@@ -36,4 +40,44 @@ my $git = PublicInbox::Git->new($git_dir);
 my @list = $git->qx(qw(rev-list refs/heads/master));
 is(scalar @list, 1, 'one revision in rev-list');
 
+my $write_spam = sub {
+        is(scalar glob("$spamdir/new/*"), undef, 'no spam existing');
+        $sem->prepare(\$msg);
+        $sem->commit;
+        my @new = glob("$spamdir/new/*");
+        is(scalar @new, 1);
+        my @p = split(m!/+!, $new[0]);
+        ok(link($new[0], "$spamdir/cur/".$p[-1]));
+        is(unlink($new[0]), 1);
+};
+$write_spam->();
+is(unlink(glob("$maildir/new/*")), 1, 'unlinked old spam');
+PublicInbox::WatchMaildir->new($config)->scan;
+@list = $git->qx(qw(rev-list refs/heads/master));
+is(scalar @list, 2, 'two revisions in rev-list');
+@list = $git->qx(qw(ls-tree -r --name-only refs/heads/master));
+is(scalar @list, 0, 'tree is empty');
+
+# check with scrubbing
+{
+        $msg .= qq(--
+To unsubscribe from this list: send the line "unsubscribe git" in
+the body of a message to majordomo\@vger.kernel.org
+More majordomo info at  http://vger.kernel.org/majordomo-info.html\n);
+        PublicInbox::Emergency->new($maildir)->prepare(\$msg);
+        PublicInbox::WatchMaildir->new($config)->scan;
+        @list = $git->qx(qw(ls-tree -r --name-only refs/heads/master));
+        is(scalar @list, 1, 'tree has one file');
+        my $mref = $git->cat_file('HEAD:'.$list[0]);
+        like($$mref, qr/something\n\z/s, 'message scrubbed on import');
+
+        is(unlink(glob("$maildir/new/*")), 1, 'unlinked spam');
+        $write_spam->();
+        PublicInbox::WatchMaildir->new($config)->scan;
+        @list = $git->qx(qw(ls-tree -r --name-only refs/heads/master));
+        is(scalar @list, 0, 'tree is empty');
+        @list = $git->qx(qw(rev-list refs/heads/master));
+        is(scalar @list, 4, 'four revisions in rev-list');
+}
+
 done_testing;