X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=t%2Fmda.t;h=d20cdb92be30fb8b962dac42585b1e7204c627ee;hb=2f8d15b1170b137a0512a4f3163b236a5dcbf0b8;hp=d37c03cd32305c5def7478b03dddb90ed8c59813;hpb=d850bde272affdae7f6d45bc27bdbf7455761300;p=public-inbox.git diff --git a/t/mda.t b/t/mda.t index d37c03cd..d20cdb92 100644 --- a/t/mda.t +++ b/t/mda.t @@ -1,15 +1,15 @@ -# Copyright (C) 2014, Eric Wong and all contributors -# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) +# Copyright (C) 2014-2021 all contributors +# License: AGPL-3.0+ 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 $tmpdir = tempdir(CLEANUP => 1); +use Cwd qw(getcwd); +use PublicInbox::MID qw(mid2path); +use PublicInbox::Git; +use PublicInbox::InboxWritable; +use PublicInbox::TestCommon; +use PublicInbox::Import; +my ($tmpdir, $for_destroy) = tmpdir(); my $home = "$tmpdir/pi-home"; my $pi_home = "$home/.public-inbox"; my $pi_config = "$pi_home/config"; @@ -20,34 +20,70 @@ my $fail_bin = getcwd()."/t/fail-bin"; 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 <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_FAILBOX} = $failbox; + local $ENV{PI_EMERGENCY} = $faildir; local $ENV{HOME} = $home; - local $ENV{RECIPIENT} = $addr; - my $simple = Email::Simple->new(< To: You Cc: $addr @@ -56,43 +92,41 @@ Subject: hihi 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 0 \+0000\n/m, + my $cmt = $git->cat_file($rev); + like($$cmt, qr/^author Me 0 \+0000\n/m, "author info set correctly"); - like($cmt, qr/^committer test /m, + like($$cmt, qr/^committer test /m, "committer info set correctly"); $good_rev = $rev; } # ensure failures work, fail with bad spamc { - ok(!-e $failbox, "nothing in PI_FAILBOX 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_FAILBOX 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 To: You Message-Id: 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 To: You Cc: $addr @@ -102,26 +136,26 @@ Date: Thu, 01 Jan 1970 00:00:00 +0000 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: 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: 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: Subject: hihi - fail_bad_header($good_rev, "bad date", <<""); + $fail_bad_header->($good_rev, "bad date", <<""); To: $addr From: u\@example.com Message-ID: @@ -130,19 +164,152 @@ Date: deadbeef } -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) = ("", ""); +# spam training +{ + local $ENV{PI_EMERGENCY} = $faildir; + local $ENV{HOME} = $home; + local $ENV{ORIGINAL_RECIPIENT} = $addr; 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_FAILBOX is written to ($msg)"); - [ $in, $out, $err ]; + my $mid = 'spam-train@example.com'; + my $in = < +To: You +Cc: $addr +Message-ID: <$mid> +Subject: this message will be trained as spam +Date: Thu, 01 Jan 1970 00:00:00 +0000 + +EOF + { + # deliver the spam message, first + 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'; + 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} = $faildir; + local $ENV{HOME} = $home; + local $ENV{ORIGINAL_RECIPIENT} = $addr; + local $ENV{PATH} = $main_path; + my $mid = 'ham-train@example.com'; + my $in = < +To: You +Cc: $addr +Message-ID: <$mid> +Subject: this message will be trained as spam +Date: Thu, 01 Jan 1970 00:00:00 +0000 + +EOF + # now train it + # these should be overridden + local $ENV{GIT_AUTHOR_EMAIL} = 'trainer@example.com'; + local $ENV{GIT_COMMITTER_EMAIL} = 'trainer@example.com'; + + 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($$msg, qr//i, ' filtered'); + } +} + +# List-ID based delivery +{ + local $ENV{PI_EMERGENCY} = $faildir; + local $ENV{HOME} = $home; + local $ENV{ORIGINAL_RECIPIENT} = undef; + delete $ENV{ORIGINAL_RECIPIENT}; + local $ENV{PATH} = $main_path; + my $list_id = 'foo.example.com'; + my $mid = 'list-id-delivery@example.com'; + my $in = < +To: You +Cc: $addr +Message-ID: <$mid> +List-Id: <$list_id> +Subject: this message will be trained as spam +Date: Thu, 01 Jan 1970 00:00:00 +0000 + +EOF + xsys(qw(git config --file), $pi_config, "$cfgpfx.listid", uc $list_id); + $? == 0 or die "failed to set listid $?"; + ok(run_script(['-mda'], undef, { 0 => \$in }), + 'mda OK with List-Id match'); + my $path = mid2path($mid); + my $msg = $git->cat_file("HEAD:$path"); + like($$msg, qr/\Q$list_id\E/, 'delivered message w/ List-ID matches'); + + # try a message w/o precheck + $in = < +List-Id: <$list_id> + +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 + $in = < +List-ID: <$list_id> +Message-ID: <2lids\@example> +Subject: two List-IDs +From: user +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();