about summary refs log tree commit homepage
path: root/script/public-inbox-learn
diff options
context:
space:
mode:
authorEric Wong <e@80x24.org>2016-02-27 21:57:57 +0000
committerEric Wong <e@80x24.org>2016-02-27 21:57:57 +0000
commit052f26f3ada1042afa5acadbecc48b487f4e2d52 (patch)
treeb6b30f94cdb20534ffc074359db30084c9858fa6 /script/public-inbox-learn
parent617f35dacbd4e5972bf2d82411b45009bbc79a42 (diff)
downloadpublic-inbox-052f26f3ada1042afa5acadbecc48b487f4e2d52.tar.gz
This seems to match more closely with what is expected of Perl
packages based on how blib is used.  Hopefully makes the top-level
source tree less cluttered and things easier-to-find.
Diffstat (limited to 'script/public-inbox-learn')
-rwxr-xr-xscript/public-inbox-learn89
1 files changed, 89 insertions, 0 deletions
diff --git a/script/public-inbox-learn b/script/public-inbox-learn
new file mode 100755
index 00000000..0c7b4199
--- /dev/null
+++ b/script/public-inbox-learn
@@ -0,0 +1,89 @@
+#!/usr/bin/perl -w
+# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
+# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+#
+# Used for training spam (via SpamAssassin) and removing messages from a
+# public-inbox
+my $usage = "$0 (spam|ham) < /path/to/message";
+use strict;
+use warnings;
+use PublicInbox::Config;
+use Email::MIME;
+use Email::Address;
+use IPC::Run qw/run/;
+my $train = shift or die "usage: $usage\n";
+if ($train !~ /\A(?:ham|spam)\z/) {
+        die "`$train' not recognized.\nusage: $usage\n";
+}
+
+my $pi_config = PublicInbox::Config->new;
+my $mime = Email::MIME->new(eval { local $/; <> });
+
+# get all recipients
+my %dests;
+foreach my $h (qw(Cc To)) {
+        foreach my $recipient (Email::Address->parse($mime->header($h))) {
+                $dests{lc($recipient->address)} = 1;
+        }
+}
+
+my ($name, $email, $date);
+
+if ($train eq "ham") {
+        require PublicInbox::MDA;
+        require PublicInbox::Filter;
+        PublicInbox::Filter->run($mime);
+        ($name, $email, $date) = PublicInbox::MDA->author_info($mime);
+}
+
+my $in = $mime->as_string;
+my $err = 0;
+my @output = qw(> /dev/null > /dev/null);
+
+# n.b. message may be cross-posted to multiple public-inboxes
+foreach my $recipient (keys %dests) {
+        my $dst = $pi_config->lookup($recipient) or next;
+        my $git_dir = $dst->{mainrepo} or next;
+        my ($out, $err) = ("", "");
+
+        # We do not touch GIT_COMMITTER_* env here so we can track
+        # who trained the message.
+        # We will not touch GIT_AUTHOR_* when learning spam messages, either
+        if ($train eq "spam") {
+                # This needs to be idempotent, as my inotify trainer
+                # may train for each cross-posted message, and this
+                # script already learns for every list in
+                # ~/.public-inbox/config
+                if (!run(["ssoma-rm", $git_dir], \$in, \$out, \$err)) {
+                        if ($err !~ /^git cat-file .+ failed: 32768$/) {
+                                $err = 1;
+                        }
+                }
+        } else { # $train eq "ham"
+                # no checking for spam here, we assume the message has
+                # been reviewed by a human at this point:
+                PublicInbox::MDA->set_list_headers($mime, $dst);
+                my $s  = $mime->as_string;
+
+                local $ENV{GIT_AUTHOR_NAME} = $name;
+                local $ENV{GIT_AUTHOR_EMAIL} = $email;
+                local $ENV{GIT_AUTHOR_DATE} = $date;
+
+                # Ham messages are trained when they're marked into
+                # a SEEN state, so this is idempotent:
+                run([PublicInbox::MDA->cmd, $git_dir], \$s, \$out, \$err);
+                if ($err !~ /CONFLICT/) {
+                        $err = 1;
+                }
+        }
+        if (!run([qw(spamc -L), $train], \$in, @output)) {
+                $err = 1;
+        }
+
+        $err or eval {
+                require PublicInbox::SearchIdx;
+                PublicInbox::SearchIdx->new($git_dir, 2)->index_sync;
+        };
+}
+
+exit $err;