From dfee7a2226a3fb47385fcf664d9c9af431e8636d Mon Sep 17 00:00:00 2001
From: Moritz Bunkus <moritz@bunkus.org>
Date: Sun, 29 Nov 2015 11:03:51 +0100
Subject: [PATCH] Matroska reader: only analyze last 5 MB of a file if no seek
 head is found

---
 ChangeLog                   |  9 +++++++++
 src/common/kax_analyzer.cpp | 14 ++++++++++++++
 src/common/kax_analyzer.h   |  2 ++
 src/input/r_matroska.cpp    | 12 +++++-------
 4 files changed, 30 insertions(+), 7 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 349efe59a..8b3d68367 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,12 @@
+2015-11-29  Moritz Bunkus  <moritz@bunkus.org>
+
+        * mkvmerge: bug fix: the change to do a deeper file analysis if no
+        seek head was found was causing huge increases in file type
+        detection time as popular tools like x264 don't write seek
+        heads. The way elements at the end are searched has been changed
+        to only scan the last 5 MB of the file instead of iterating over
+        every level 1 element from the beginning of the file.
+
 2015-11-28  Moritz Bunkus  <moritz@bunkus.org>
 
         * Released v8.6.0.
diff --git a/src/common/kax_analyzer.cpp b/src/common/kax_analyzer.cpp
index 34165e815..0c462974c 100644
--- a/src/common/kax_analyzer.cpp
+++ b/src/common/kax_analyzer.cpp
@@ -216,6 +216,12 @@ kax_analyzer_c::set_throw_on_error(bool throw_on_error) {
   return *this;
 }
 
+kax_analyzer_c &
+kax_analyzer_c::set_parser_start_position(uint64_t position) {
+  m_parser_start_position.reset(position);
+  return *this;
+}
+
 bool
 kax_analyzer_c::process() {
   try {
@@ -278,6 +284,14 @@ kax_analyzer_c::process_internal() {
   auto segment_end     = m_segment->IsFiniteSize() ? m_segment->GetElementPosition() + m_segment->HeadSize() + m_segment->GetSize() : m_file->get_size();
   EbmlElement *l1      = nullptr;
 
+  // In certain situations the caller doesn't way to have to pay the
+  // price for full analysis. Then it can configure the parser to
+  // start parsing at a certain offset. EbmlStream::FindNextElement()
+  // should take care of re-syncing to a known level 1 element. But
+  // take care not to start before the segment's data start position.
+  if (m_parser_start_position)
+    m_file->setFilePointer(std::max<uint64_t>(*m_parser_start_position, m_segment->GetElementPosition() + m_segment->HeadSize()));
+
   // We've got our segment, so let's find all level 1 elements.
   while (m_file->getFilePointer() < segment_end) {
     if (!l1)
diff --git a/src/common/kax_analyzer.h b/src/common/kax_analyzer.h
index b8c7e9077..fb8e0a7af 100644
--- a/src/common/kax_analyzer.h
+++ b/src/common/kax_analyzer.h
@@ -102,6 +102,7 @@ private:
   parse_mode_e m_parse_mode{parse_mode_full};
   open_mode m_open_mode{MODE_WRITE};
   bool m_throw_on_error{};
+  boost::optional<uint64_t> m_parser_start_position;
 
 public:                         // Static functions
   static bool probe(std::string file_name);
@@ -130,6 +131,7 @@ public:
   virtual kax_analyzer_c &set_parse_mode(parse_mode_e parse_mode);
   virtual kax_analyzer_c &set_open_mode(open_mode mode);
   virtual kax_analyzer_c &set_throw_on_error(bool throw_on_error);
+  virtual kax_analyzer_c &set_parser_start_position(uint64_t position);
 
   virtual bool process();
 
diff --git a/src/input/r_matroska.cpp b/src/input/r_matroska.cpp
index cf6a6e8e0..7d253f3b2 100644
--- a/src/input/r_matroska.cpp
+++ b/src/input/r_matroska.cpp
@@ -1205,15 +1205,13 @@ kax_reader_c::read_headers() {
 
 void
 kax_reader_c::find_level1_elements_via_analyzer() {
-  if (!debugging_c::requested("kax_reader_deep_scan"))
-    return;
-
   try {
-    auto analyzer = std::make_shared<kax_analyzer_c>(m_in.get());
-    auto ok       = analyzer
-      ->set_parse_mode(kax_analyzer_c::parse_mode_fast)
+    auto start_pos = m_in->get_size() - std::min<int64_t>(m_in->get_size(), 5 * 1024 * 1024);
+    auto analyzer  = std::make_shared<kax_analyzer_c>(m_in.get());
+    auto ok        = analyzer
+      ->set_parse_mode(kax_analyzer_c::parse_mode_full)
       .set_open_mode(MODE_READ)
-      .set_throw_on_error(true)
+      .set_parser_start_position(start_pos)
       .process();
 
     if (!ok)