Strip disallowed characters from subtitles before they get into the XML.

3 files changed, 57 insertions, 0 deletions

diff --git a/test/srt_subtitle_test.cc b/test/srt_subtitle_test.cc
index 935462867..63d508b76 100644
--- a/test/srt_subtitle_test.cc
+++ b/test/srt_subtitle_test.cc
@@ -248,6 +248,30 @@ BOOST_AUTO_TEST_CASE(srt_subtitle_entity)
 }
 
 
+/** A control code in a .srt file should not make it into the XML */
+BOOST_AUTO_TEST_CASE(srt_subtitle_control_code)
+{
+	std::ofstream srt("build/test/srt_subtitle_control_code.srt");
+	srt << "1\n";
+	srt << "00:00:01,000 -> 00:00:10,000\n";
+	srt << "Hello \x0c world\n";
+	srt.close();
+
+	auto content = make_shared<StringTextFileContent>("build/test/srt_subtitle_control_code.srt");
+	auto film = new_test_film2("srt_subtitle_control_code", { content });
+	film->set_interop(false);
+	content->only_text()->set_use(true);
+	content->only_text()->set_burn(false);
+	make_and_verify_dcp (
+		film,
+		{
+			dcp::VerificationNote::Code::MISSING_SUBTITLE_LANGUAGE,
+			dcp::VerificationNote::Code::INVALID_SUBTITLE_FIRST_TEXT_TIME,
+			dcp::VerificationNote::Code::MISSING_CPL_METADATA,
+		});
+}
+
+
 #if 0
 /* XXX: this is disabled; there is some difference in font rendering
    between the test machine and others.
diff --git a/test/text_decoder_test.cc b/test/text_decoder_test.cc
new file mode 100644
index 000000000..d6cbd4ce1
--- /dev/null
+++ b/test/text_decoder_test.cc
@@ -0,0 +1,32 @@
+/*
+    Copyright (C) 2023 Carl Hetherington <cth@carlh.net>
+
+    This file is part of DCP-o-matic.
+
+    DCP-o-matic is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    DCP-o-matic is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with DCP-o-matic.  If not, see <http://www.gnu.org/licenses/>.
+
+*/
+
+
+#include "lib/text_decoder.h"
+#include <boost/test/unit_test.hpp>
+
+
+BOOST_AUTO_TEST_CASE(strip_invalid_characters_for_xml_test)
+{
+	BOOST_CHECK_EQUAL(TextDecoder::remove_invalid_characters_for_xml("hello world"), "hello world");
+	BOOST_CHECK_EQUAL(TextDecoder::remove_invalid_characters_for_xml("hello\x0cworld"), "helloworld");
+	BOOST_CHECK_EQUAL(TextDecoder::remove_invalid_characters_for_xml("𒀖hello\x02worl𒁝d"), "𒀖helloworl𒁝d");
+	BOOST_CHECK_EQUAL(TextDecoder::remove_invalid_characters_for_xml("😀œ´®†¥¨ˆø\x09π¬˚∆\x1a˙©ƒ∂ßåΩ≈ç√∫\x02˜µ≤ユーザーコードa"), "😀œ´®†¥¨ˆø\x09π¬˚∆˙©ƒ∂ßåΩ≈ç√∫˜µ≤ユーザーコードa");
+}
diff --git a/test/wscript b/test/wscript
index 5e06ed5e2..949f69019 100644
--- a/test/wscript
+++ b/test/wscript
@@ -149,6 +149,7 @@ def build(bld):
                  subtitle_timing_test.cc
                  subtitle_trim_test.cc
                  test.cc
+                 text_decoder_test.cc
                  threed_test.cc
                  time_calculation_test.cc
                  torture_test.cc