-# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2014-2021 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
use strict;
use warnings;
use Test::More;
-use Email::MIME;
-use File::Temp qw/tempdir/;
-use Cwd;
-use IPC::Run qw(run);
-
-my $mda = "blib/script/public-inbox-mda";
-my $learn = "blib/script/public-inbox-learn";
-my $tmpdir = tempdir(CLEANUP => 1);
+use Cwd qw(getcwd);
+use PublicInbox::MID qw(mid2path);
+use PublicInbox::Git;
+use PublicInbox::InboxWritable;
+use PublicInbox::TestCommon;
+use PublicInbox::Import;
+my ($tmpdir, $for_destroy) = tmpdir();
my $home = "$tmpdir/pi-home";
my $pi_home = "$home/.public-inbox";
my $pi_config = "$pi_home/config";
my $fail_path = "$fail_bin:$ENV{PATH}"; # for spamc spam mock
my $addr = 'test-public@example.com';
my $cfgpfx = "publicinbox.test";
-my $failbox = "$home/fail.mbox";
+my $faildir = "$home/faildir/";
+my $git = PublicInbox::Git->new($maindir);
+
+my $fail_bad_header = sub ($$$) {
+ my ($good_rev, $msg, $in) = @_;
+ my @f = glob("$faildir/*/*");
+ unlink @f if @f;
+ my ($out, $err) = ("", "");
+ my $opt = { 0 => \$in, 1 => \$out, 2 => \$err };
+ local $ENV{PATH} = $main_path;
+ ok(run_script(['-mda'], undef, $opt),
+ "no error on undeliverable ($msg)");
+ my $rev = $git->qx(qw(rev-list HEAD));
+ chomp $rev;
+ is($rev, $good_rev, "bad revision not committed ($msg)");
+ @f = glob("$faildir/*/*");
+ is(scalar @f, 1, "faildir written to");
+ [ $in, $out, $err ];
+};
{
ok(-x "$main_bin/spamc",
"spamc ham mock found (run in top of source tree");
ok(-x "$fail_bin/spamc",
"spamc mock found (run in top of source tree");
- ok(-x $mda, "$mda is executable");
is(1, mkdir($home, 0755), "setup ~/ for testing");
is(1, mkdir($pi_home, 0755), "setup ~/.public-inbox");
- is(0, system(qw(git init -q --bare), $maindir), "git init (main)");
-
- my %cfg = (
- "$cfgpfx.address" => $addr,
- "$cfgpfx.mainrepo" => $maindir,
- );
- while (my ($k,$v) = each %cfg) {
- is(0, system(qw(git config --file), $pi_config, $k, $v),
- "setup $k");
- }
+ PublicInbox::Import::init_bare($maindir);
+
+ open my $fh, '>>', $pi_config or die;
+ print $fh <<EOF or die;
+[publicinbox "test"]
+ address = $addr
+ inboxdir = $maindir
+EOF
+ close $fh or die;
}
+local $ENV{GIT_COMMITTER_NAME} = eval {
+ use PublicInbox::MDA;
+ use PublicInbox::Address;
+ use Encode qw/encode/;
+ my $msg = eml_load 't/utf8.eml';
+ my $from = $msg->header('From');
+ my ($author) = PublicInbox::Address::names($from);
+ my ($email) = PublicInbox::Address::emails($from);
+ my $date = $msg->header('Date');
+
+ is('Eléanor',
+ encode('us-ascii', my $tmp = $author, Encode::HTMLCREF),
+ 'HTML conversion is correct');
+ is($email, 'e@example.com', 'email parsed correctly');
+ is($date, 'Thu, 01 Jan 1970 00:00:00 +0000',
+ 'message date parsed correctly');
+ $author;
+};
+die $@ if $@;
+
{
my $good_rev;
- local $ENV{PI_EMERGENCY} = $failbox;
+ local $ENV{PI_EMERGENCY} = $faildir;
local $ENV{HOME} = $home;
local $ENV{ORIGINAL_RECIPIENT} = $addr;
- my $simple = Email::Simple->new(<<EOF);
+ my $in = <<EOF;
From: Me <me\@example.com>
To: You <you\@example.com>
Cc: $addr
Date: Thu, 01 Jan 1970 00:00:00 +0000
EOF
- my $in = $simple->as_string;
-
# ensure successful message delivery
{
local $ENV{PATH} = $main_path;
- run([$mda], \$in);
- local $ENV{GIT_DIR} = $maindir;
- my $rev = `git rev-list HEAD`;
+ ok(run_script(['-mda'], undef, { 0 => \$in }));
+ my $rev = $git->qx(qw(rev-list HEAD));
like($rev, qr/\A[a-f0-9]{40}/, "good revision committed");
chomp $rev;
- my $cmt = `git cat-file commit $rev`;
- like($cmt, qr/^author Me <me\@example\.com> 0 \+0000\n/m,
+ my $cmt = $git->cat_file($rev);
+ like($$cmt, qr/^author Me <me\@example\.com> 0 \+0000\n/m,
"author info set correctly");
- like($cmt, qr/^committer test <test-public\@example\.com>/m,
+ like($$cmt, qr/^committer test <test-public\@example\.com>/m,
"committer info set correctly");
$good_rev = $rev;
}
# ensure failures work, fail with bad spamc
{
- ok(!-e $failbox, "nothing in PI_EMERGENCY before");
+ my @prev = <$faildir/new/*>;
+ is(scalar @prev, 0 , "nothing in PI_EMERGENCY before");
local $ENV{PATH} = $fail_path;
- run([$mda], \$in);
- local $ENV{GIT_DIR} = $maindir;
- my @revs = `git rev-list HEAD`;
+ ok(run_script(['-mda'], undef, { 0 => \$in }));
+ my @revs = $git->qx(qw(rev-list HEAD));
is(scalar @revs, 1, "bad revision not committed");
- ok(-s $failbox > 0, "PI_EMERGENCY is written to");
+ my @new = <$faildir/new/*>;
+ is(scalar @new, 1, "PI_EMERGENCY is written to");
}
- fail_bad_header($good_rev, "bad recipient", <<"");
+ $fail_bad_header->($good_rev, "bad recipient", <<"");
From: Me <me\@example.com>
To: You <you\@example.com>
Message-Id: <bad-recipient\@example.com>
Subject: hihi
Date: Thu, 01 Jan 1970 00:00:00 +0000
- my $fail = fail_bad_header($good_rev, "duplicate Message-ID", <<"");
+ my $fail = $fail_bad_header->($good_rev, "duplicate Message-ID", <<"");
From: Me <me\@example.com>
To: You <you\@example.com>
Cc: $addr
like($fail->[2], qr/CONFLICT/, "duplicate Message-ID message");
- fail_bad_header($good_rev, "missing From:", <<"");
+ $fail_bad_header->($good_rev, "missing From:", <<"");
To: $addr
Message-ID: <missing-from\@example.com>
Subject: hihi
Date: Thu, 01 Jan 1970 00:00:00 +0000
- fail_bad_header($good_rev, "short subject:", <<"");
+ $fail_bad_header->($good_rev, "short subject:", <<"");
To: $addr
From: cat\@example.com
Message-ID: <short-subject\@example.com>
Subject: a
Date: Thu, 01 Jan 1970 00:00:00 +0000
- fail_bad_header($good_rev, "no date", <<"");
+ $fail_bad_header->($good_rev, "no date", <<"");
To: $addr
From: u\@example.com
Message-ID: <no-date\@example.com>
Subject: hihi
- fail_bad_header($good_rev, "bad date", <<"");
+ $fail_bad_header->($good_rev, "bad date", <<"");
To: $addr
From: u\@example.com
Message-ID: <bad-date\@example.com>
# spam training
{
- local $ENV{PI_EMERGENCY} = $failbox;
+ local $ENV{PI_EMERGENCY} = $faildir;
local $ENV{HOME} = $home;
local $ENV{ORIGINAL_RECIPIENT} = $addr;
local $ENV{PATH} = $main_path;
my $mid = 'spam-train@example.com';
- my $simple = Email::Simple->new(<<EOF);
+ my $in = <<EOF;
From: Spammer <spammer\@example.com>
To: You <you\@example.com>
Cc: $addr
Date: Thu, 01 Jan 1970 00:00:00 +0000
EOF
- my $in = $simple->as_string;
-
{
# deliver the spam message, first
- run([$mda], \$in);
- my $msg = `ssoma cat $mid $maindir`;
- like($msg, qr/\Q$mid\E/, "message delivered");
+ ok(run_script(['-mda'], undef, { 0 => \$in }));
+ my $path = mid2path($mid);
+ my $msg = $git->cat_file("HEAD:$path");
+ like($$msg, qr/\Q$mid\E/, "message delivered");
# now train it
local $ENV{GIT_AUTHOR_EMAIL} = 'trainer@example.com';
local $ENV{GIT_COMMITTER_EMAIL} = 'trainer@example.com';
- run([$learn, "spam"], \$msg);
- is($?, 0, "no failure from learning spam");
- run([$learn, "spam"], \$msg);
- is($?, 0, "no failure from learning spam idempotently");
+ local $ENV{GIT_COMMITTER_NAME};
+ delete $ENV{GIT_COMMITTER_NAME};
+ ok(run_script(['-learn', 'spam'], undef, { 0 => $msg }),
+ "no failure from learning spam");
+ ok(run_script(['-learn', 'spam'], undef, { 0 => $msg }),
+ "no failure from learning spam idempotently");
}
}
# train ham message
{
- local $ENV{PI_EMERGENCY} = $failbox;
+ local $ENV{PI_EMERGENCY} = $faildir;
local $ENV{HOME} = $home;
local $ENV{ORIGINAL_RECIPIENT} = $addr;
local $ENV{PATH} = $main_path;
my $mid = 'ham-train@example.com';
- my $simple = Email::Simple->new(<<EOF);
+ my $in = <<EOF;
From: False-positive <hammer\@example.com>
To: You <you\@example.com>
Cc: $addr
Date: Thu, 01 Jan 1970 00:00:00 +0000
EOF
- my $in = $simple->as_string;
-
# now train it
+ # these should be overridden
local $ENV{GIT_AUTHOR_EMAIL} = 'trainer@example.com';
local $ENV{GIT_COMMITTER_EMAIL} = 'trainer@example.com';
- run([$learn, "ham"], \$in);
- is($?, 0, "learned ham without failure");
- my $msg = `ssoma cat $mid $maindir`;
- like($msg, qr/\Q$mid\E/, "ham message delivered");
- run([$learn, "ham"], \$in);
- is($?, 0, "learned ham idempotently ");
+
+ ok(run_script(['-learn', 'ham'], undef, { 0 => \$in }),
+ "learned ham without failure");
+ my $path = mid2path($mid);
+ my $msg = $git->cat_file("HEAD:$path");
+ like($$msg, qr/\Q$mid\E/, "ham message delivered");
+ ok(run_script(['-learn', 'ham'], undef, { 0 => \$in }),
+ "learned ham idempotently ");
+
+ # ensure trained email is filtered, too
+ my $mime = eml_load 't/mda-mime.eml';
+ ($mid) = ($mime->header_raw('message-id') =~ /<([^>]+)>/);
+ {
+ $in = $mime->as_string;
+ ok(run_script(['-learn', 'ham'], undef, { 0 => \$in }),
+ "learned ham without failure");
+ my $path = mid2path($mid);
+ $msg = $git->cat_file("HEAD:$path");
+ like($$msg, qr/<\Q$mid\E>/, "ham message delivered");
+ unlike($$msg, qr/<html>/i, '<html> filtered');
+ }
}
-# faildir - emergency destination is maildir
+# List-ID based delivery
{
- my $faildir= "$home/faildir/";
local $ENV{PI_EMERGENCY} = $faildir;
local $ENV{HOME} = $home;
- local $ENV{ORIGINAL_RECIPIENT} = $addr;
- local $ENV{PATH} = $fail_path;
+ local $ENV{ORIGINAL_RECIPIENT} = undef;
+ delete $ENV{ORIGINAL_RECIPIENT};
+ local $ENV{PATH} = $main_path;
+ my $list_id = 'foo.example.com';
+ my $mid = 'list-id-delivery@example.com';
my $in = <<EOF;
-From: Faildir <faildir\@example.com>
+From: user <user\@example.com>
To: You <you\@example.com>
Cc: $addr
-Message-ID: <faildir\@example.com>
-Subject: faildir subject
+Message-ID: <$mid>
+List-Id: <$list_id>
+Subject: this message will be trained as spam
Date: Thu, 01 Jan 1970 00:00:00 +0000
EOF
- run([$mda], \$in);
- ok(-d $faildir, "emergency exists");
- my @new = glob("$faildir/new/*");
- is(scalar(@new), 1, "message delivered");
-}
+ xsys(qw(git config --file), $pi_config, "$cfgpfx.listid", uc $list_id);
+ $? == 0 or die "failed to set listid $?";
+ ok(run_script(['-mda'], undef, { 0 => \$in }),
+ 'mda OK with List-Id match');
+ my $path = mid2path($mid);
+ my $msg = $git->cat_file("HEAD:$path");
+ like($$msg, qr/\Q$list_id\E/, 'delivered message w/ List-ID matches');
-done_testing();
+ # try a message w/o precheck
+ $in = <<EOF;
+To: You <you\@example.com>
+List-Id: <$list_id>
-sub fail_bad_header {
- my ($good_rev, $msg, $in) = @_;
- open my $fh, '>', $failbox or die "failed to open $failbox: $!\n";
- close $fh or die "failed to close $failbox: $!\n";
- my ($out, $err) = ("", "");
- local $ENV{PATH} = $main_path;
- run([$mda], \$in, \$out, \$err);
- local $ENV{GIT_DIR} = $maindir;
- my $rev = `git rev-list HEAD`;
- chomp $rev;
- is($rev, $good_rev, "bad revision not commited ($msg)");
- ok(-s $failbox > 0, "PI_EMERGENCY is written to ($msg)");
- [ $in, $out, $err ];
+this message would not be accepted without --no-precheck
+EOF
+ my ($out, $err) = ('', '');
+ my $rdr = { 0 => \$in, 1 => \$out, 2 => \$err };
+ ok(run_script(['-mda', '--no-precheck'], undef, $rdr),
+ 'mda OK with List-Id match and --no-precheck');
+ my $cur = $git->qx(qw(diff HEAD~1..HEAD));
+ like($cur, qr/this message would not be accepted without --no-precheck/,
+ '--no-precheck delivered message anyways');
+
+ # try a message with multiple List-ID headers
+ $in = <<EOF;
+List-ID: <foo.bar>
+List-ID: <$list_id>
+Message-ID: <2lids\@example>
+Subject: two List-IDs
+From: user <user\@example.com>
+To: $addr
+Date: Fri, 02 Oct 1993 00:00:00 +0000
+
+EOF
+ ($out, $err) = ('', '');
+ ok(run_script(['-mda'], undef, $rdr),
+ 'mda OK with multiple List-Id matches');
+ $cur = $git->qx(qw(diff HEAD~1..HEAD));
+ like($cur, qr/^\+Message-ID: <2lids\@example>/sm,
+ 'multi List-ID match delivered');
+ like($err, qr/multiple List-ID/, 'warned about multiple List-ID');
+
+ # ensure -learn rm works after inbox address is updated
+ ($out, $err) = ('', '');
+ xsys(qw(git config --file), $pi_config, "$cfgpfx.address",
+ 'updated-address@example.com');
+ ok(run_script(['-learn', 'rm'], undef, $rdr), 'rm-ed via -learn');
+ $cur = $git->qx(qw(diff HEAD~1..HEAD));
+ like($cur, qr/^-Message-ID: <2lids\@example>/sm, 'changed in git');
}
+
+done_testing();