From 1bb63b08bd1137c8054c58a3c719d57ddae263b1 Mon Sep 17 00:00:00 2001
From: Moritz Bunkus <moritz@bunkus.org>
Date: Wed, 20 Jun 2018 20:56:19 +0200
Subject: [PATCH] SRT textsubs: try UTF-8 if no encoding was specified with
 fallback to system's encoding

Part of the implementation of #2246.
---
 NEWS.md                   |  3 +++
 src/output/p_textsubs.cpp | 12 +++++++++---
 src/output/p_textsubs.h   |  1 +
 3 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/NEWS.md b/NEWS.md
index 9e3dad496..067b5df71 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -6,6 +6,9 @@
   complete `gettext`. Together with an additional fix to how translation files
   are located, MKVToolNix can now use all interface languages on macOS,
   too. Fixes #2110, #2307, #2323.
+* mkvmerge: SRT text subtitles: for files for which no encoding has been
+  specified, mkvmerge will try UTF-8 first before falling back to the system's
+  default encoding. Part of the implementation of #2246.
 
 ## Bug fixes
 
diff --git a/src/output/p_textsubs.cpp b/src/output/p_textsubs.cpp
index f413de2c5..988e15e41 100644
--- a/src/output/p_textsubs.cpp
+++ b/src/output/p_textsubs.cpp
@@ -18,6 +18,7 @@
 #include "common/codec.h"
 #include "common/strings/editing.h"
 #include "common/strings/parsing.h"
+#include "common/strings/utf8.h"
 #include "merge/connection_checks.h"
 #include "merge/output_control.h"
 #include "merge/packet_extensions.h"
@@ -32,8 +33,10 @@ textsubs_packetizer_c::textsubs_packetizer_c(generic_reader_c *p_reader,
   : generic_packetizer_c(p_reader, p_ti)
   , m_codec_id{codec_id}
 {
-  if (recode)
-    m_cc_utf8 = charset_converter_c::init(m_ti.m_sub_charset);
+  if (recode) {
+    m_cc_utf8  = charset_converter_c::init(m_ti.m_sub_charset);
+    m_try_utf8 = m_ti.m_sub_charset.empty();
+  }
 
   set_track_type(track_subtitle);
   if (m_codec_id == MKV_S_TEXTUSF)
@@ -88,7 +91,10 @@ textsubs_packetizer_c::process(packet_cptr packet) {
   auto subs = std::string{reinterpret_cast<char *>(packet->data->get_buffer()), packet->data->get_size()};
   subs      = chomp(normalize_line_endings(subs, m_line_ending_style));
 
-  if (m_cc_utf8)
+  if (m_try_utf8 && !mtx::utf8::is_valid(subs))
+    m_try_utf8 = false;
+
+  if (!m_try_utf8 && m_cc_utf8)
     subs = m_cc_utf8->utf8(subs);
 
   packet->data = memory_c::borrow(subs);
diff --git a/src/output/p_textsubs.h b/src/output/p_textsubs.h
index 51ef508a5..f28294b38 100644
--- a/src/output/p_textsubs.h
+++ b/src/output/p_textsubs.h
@@ -23,6 +23,7 @@ protected:
   unsigned int m_packetno{};
   boost::optional<unsigned int> m_force_rerender_track_headers_on_packetno;
   charset_converter_cptr m_cc_utf8;
+  bool m_try_utf8{};
   std::string m_codec_id;
   line_ending_style_e m_line_ending_style{line_ending_style_e::cr_lf};