X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=t%2Fmsg_iter.t;h=ae3594da08324f67b7747ce9b0d8a6432cad02ac;hb=4eee5af6011cc8cdefb66c9729952c7eff5c0b0b;hp=f9b586f183d96f7cc54ee605d418ee0788c705b0;hpb=130af18f06ae9b91e07985ff56b4dd90cedbd744;p=public-inbox.git

diff --git a/t/msg_iter.t b/t/msg_iter.t
index f9b586f1..ae3594da 100644
--- a/t/msg_iter.t
+++ b/t/msg_iter.t
@@ -1,16 +1,12 @@
-# Copyright (C) 2016-2019 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2021 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 strict; use v5.10.1; use PublicInbox::TestCommon;
+use PublicInbox::Hval qw(ascii_html);
+use MIME::QuotedPrint 3.05 qw(encode_qp);
 use_ok('PublicInbox::MsgIter');
 
 {
-	my $parts = [ Email::MIME->create(body => "a\n"),
-			Email::MIME->create(body => "b\n") ];
-	my $mime = Email::MIME->create(parts => $parts,
-				header_str => [ From => 'root@localhost' ]);
+	my $mime = eml_load 't/msg_iter-order.eml';
 	my @parts;
 	msg_iter($mime, sub {
 		my ($part, $level, @ex) = @{$_[0]};
@@ -22,13 +18,7 @@ use_ok('PublicInbox::MsgIter');
 }
 
 {
-	my $parts = [ Email::MIME->create(body => 'a'),
-			Email::MIME->create(body => 'b') ];
-	$parts = [ Email::MIME->create(parts => $parts,
-				header_str => [ From => 'sub@localhost' ]),
-			Email::MIME->create(body => 'sig') ];
-	my $mime = Email::MIME->create(parts => $parts,
-				header_str => [ From => 'root@localhost' ]);
+	my $mime = eml_load 't/msg_iter-nested.eml';
 	my @parts;
 	msg_iter($mime, sub {
 		my ($part, $level, @ex) = @{$_[0]};
@@ -36,17 +26,12 @@ use_ok('PublicInbox::MsgIter');
 		$s =~ s/\s+//s;
 		push @parts, [ $s, $level, @ex ];
 	});
-	is_deeply(\@parts, [ [qw(a 2 1 1)], [qw(b 2 1 2)], [qw(sig 1 2)] ],
+	is_deeply(\@parts, [ [qw(a 2 1.1)], [qw(b 2 1.2)], [qw(sig 1 2)] ],
 		'nested part shows up properly');
 }
 
 {
-	my $f = 't/iso-2202-jp.mbox';
-	my $mime = Email::MIME->new(do {
-		open my $fh, '<', $f or die "open($f): $!";
-		local $/;
-		<$fh>;
-	});
+	my $mime = eml_load 't/iso-2202-jp.eml';
 	my $raw = '';
 	msg_iter($mime, sub {
 		my ($part, $level, @ex) = @{$_[0]};
@@ -58,5 +43,105 @@ use_ok('PublicInbox::MsgIter');
 	is(index($raw, '$$$'), -1, 'no unescaped $$$');
 }
 
+{
+	my $mime = eml_load 't/x-unknown-alpine.eml';
+	my $raw = '';
+	msg_iter($mime, sub {
+		my ($part, $level, @ex) = @{$_[0]};
+		my ($s, $err) = msg_part_text($part, 'text/plain');
+		$raw .= $s;
+	});
+	like($raw, qr!^\thttps://!ms, 'tab expanded with X-UNKNOWN');
+	like(ascii_html($raw), qr/&#8226; bullet point/s,
+		'got bullet point when X-UNKNOWN assumes UTF-8');
+}
+
+{ # API not finalized
+	my @warn;
+	local $SIG{__WARN__} = sub { push @warn, [ @_ ] };
+	my $attr = "So and so wrote:\n";
+	my $q = "> hello world\n" x 10;
+	my $nq = "hello world\n" x 10;
+	my @sections = PublicInbox::MsgIter::split_quotes($attr . $q . $nq);
+	is($sections[0], $attr, 'attribution matches');
+	is($sections[1], $q, 'quoted section matches');
+	is($sections[2], $nq, 'non-quoted section matches');
+	is(scalar(@sections), 3, 'only three sections for short message');
+	is_deeply(\@warn, [], 'no warnings');
+
+	$q x= 3300;
+	$nq x= 3300;
+	@sections = PublicInbox::MsgIter::split_quotes($attr . $q . $nq);
+	is_deeply(\@warn, [], 'no warnings on giant message');
+	is(join('', @sections), $attr . $q . $nq, 'result matches expected');
+	is(shift(@sections), $attr, 'attribution is first section');
+	my @check = ('', '');
+	while (defined(my $l = shift @sections)) {
+		next if $l eq '';
+		like($l, qr/\n\z/s, 'section ends with newline');
+		my $idx = ($l =~ /\A>/) ? 0 : 1;
+		$check[$idx] .= $l;
+	}
+	is($check[0], $q, 'long quoted section matches');
+	is($check[1], $nq, 'long quoted section matches');
+}
+
+{
+	open my $fh, '<', 't/utf8.eml' or BAIL_OUT $!;
+	my $expect = do { local $/; <$fh>  };
+	my $qp_patch = encode_qp($expect, "\r\n");
+	my $common = <<EOM;
+Content-Type: multipart/mixed; boundary="DEADBEEF"
+MIME-Version: 1.0
+
+--DEADBEEF
+Content-Transfer-Encoding: quoted-printable
+Content-Type: text/plain;
+	charset=utf-8
+
+blah
+
+--DEADBEEF
+Content-Disposition: attachment;
+	filename=foo.patch
+Content-Type: application/octet-stream;
+	x-unix-mode=0644;
+	name="foo.patch"
+Content-Transfer-Encoding: quoted-printable
+EOM
+	my $eml = PublicInbox::Eml->new(<<EOM);
+$common
+$qp_patch
+--DEADBEEF--
+EOM
+	my @parts;
+	$eml->each_part(sub {
+		my ($part, $level, @ex) = @{$_[0]};
+		my ($s, $err) = msg_part_text($part, $part->content_type);
+		push @parts, $s;
+	});
+	$expect =~ s/\n/\r\n/sg;
+	utf8::decode($expect); # aka "bytes2str"
+	is_deeply(\@parts, [ "blah\r\n", $expect ],
+		'fallback to application/octet-stream as UTF-8 text');
+
+	my $qp_binary = encode_qp("Binary\0crap", "\r\n");
+	$eml = PublicInbox::Eml->new(<<EOM);
+$common
+$qp_binary
+--DEADBEEF--
+EOM
+	@parts = ();
+	my @err;
+	$eml->each_part(sub {
+		my ($part, $level, @ex) = @{$_[0]};
+		my ($s, $err) = msg_part_text($part, $part->content_type);
+		push @parts, $s;
+		push @err, $err;
+	});
+	is_deeply(\@parts, [ "blah\r\n", undef ],
+		'non-text ignored in octet-stream');
+	ok($err[1], 'got error for second element');
+}
+
 done_testing();
-1;