Only run container name detection upon ffmpeg parse failure.

media/filters/ffmpeg_glue.cc

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "media/filters/ffmpeg_glue.h"
 
 #include "base/logging.h"
 #include "base/macros.h"
 #include "base/metrics/histogram_macros.h"
 #include "base/synchronization/lock.h"
@@ skipping 119 matching lines @@
   format_context_->pb = avio_context_.get();
 }
 
 bool FFmpegGlue::OpenContext() {
   DCHECK(!open_called_) << "OpenContext() shouldn't be called twice.";
 
   // If avformat_open_input() is called we have to take a slightly different
   // destruction path to avoid double frees.
   open_called_ = true;
 
-  // Attempt to recognize the container by looking at the first few bytes of the
-  // stream. The stream position is left unchanged.
-  std::unique_ptr<std::vector<uint8_t>> buffer(new std::vector<uint8_t>(8192));
+  // By passing nullptr for the filename (second parameter) we are telling
+  // FFmpeg to use the AVIO context we setup from the AVFormatContext structure.
+  const int ret =
+      avformat_open_input(&format_context_, nullptr, nullptr, nullptr);
 
-  int64_t pos = AVIOSeekOperation(avio_context_.get()->opaque, 0, SEEK_CUR);
-  AVIOSeekOperation(avio_context_.get()->opaque, 0, SEEK_SET);
-  int numRead = AVIOReadOperation(
-      avio_context_.get()->opaque, buffer.get()->data(), buffer.get()->size());
-  AVIOSeekOperation(avio_context_.get()->opaque, pos, SEEK_SET);
-  if (numRead > 0) {
-    // < 0 means Read failed
-    container_names::MediaContainerName container =
-        container_names::DetermineContainer(buffer.get()->data(), numRead);
-    UMA_HISTOGRAM_SPARSE_SLOWLY("Media.DetectedContainer", container);
+  // If FFmpeg can't identify the file, read the first 8k and attempt to guess
+  // at the container type ourselves. This way we can track emergent formats.
+  // Only try on AVERROR_INVALIDDATA to avoid running after I/O errors.
+  if (ret == AVERROR_INVALIDDATA) {
+    std::vector<uint8_t> buffer(8192);
+
+    const int64_t pos = AVIOSeekOperation(avio_context_->opaque, 0, SEEK_SET);
+    if (pos < 0)
+      return false;
+
+    const int num_read =
+        AVIOReadOperation(avio_context_->opaque, buffer.data(), buffer.size());
+    if (num_read < container_names::kMinimumContainerSize)
+      return false;
+
+    UMA_HISTOGRAM_SPARSE_SLOWLY(
+        "Media.DetectedContainer",
+        container_names::DetermineContainer(buffer.data(), num_read));
+    return false;
+  } else if (ret < 0) {
+    return false;
   }
 
-  // By passing nullptr for the filename (second parameter) we are telling
-  // FFmpeg to use the AVIO context we setup from the AVFormatContext structure.
-  return avformat_open_input(&format_context_, nullptr, nullptr, nullptr) == 0;
+  // Rely on ffmpeg's parsing if we're able to succesfully open the file.
+  container_names::MediaContainerName container =
+      container_names::CONTAINER_UNKNOWN;
+  if (strcmp(format_context_->iformat->name, "mov,mp4,m4a,3gp,3g2,mj2") == 0)

[jrummell, 2017/04/17 18:00:15]

Too bad there isn't a better way to compute this without comparing strings.

[DaleCurtis, 2017/04/17 19:08:46]

Yeah, hence all the test cases to make sure we still match, wish they had an
enum instead for this :( Not sure why they decided not to have one for formats
while having one for codecs.

+    container = container_names::CONTAINER_MOV;
+  else if (strcmp(format_context_->iformat->name, "flac") == 0)
+    container = container_names::CONTAINER_FLAC;
+  else if (strcmp(format_context_->iformat->name, "matroska,webm") == 0)
+    container = container_names::CONTAINER_WEBM;
+  else if (strcmp(format_context_->iformat->name, "ogg") == 0)
+    container = container_names::CONTAINER_OGG;
+  else if (strcmp(format_context_->iformat->name, "wav") == 0)
+    container = container_names::CONTAINER_WAV;
+  else if (strcmp(format_context_->iformat->name, "aac") == 0)
+    container = container_names::CONTAINER_AAC;
+  else if (strcmp(format_context_->iformat->name, "mp3") == 0)
+    container = container_names::CONTAINER_MP3;
+  else if (strcmp(format_context_->iformat->name, "amr") == 0)
+    container = container_names::CONTAINER_AMR;
+  else if (strcmp(format_context_->iformat->name, "avi") == 0)
+    container = container_names::CONTAINER_AVI;
+  // TODO(jrummell): Remove GSM detection. http://crbug.com/711774
+  else if (strcmp(format_context_->iformat->name, "gsm") == 0)
+    container = container_names::CONTAINER_GSM;
+
+  DCHECK_NE(container, container_names::CONTAINER_UNKNOWN);
+  UMA_HISTOGRAM_SPARSE_SLOWLY("Media.DetectedContainer", container);
+
+  return true;
 }
 
 FFmpegGlue::~FFmpegGlue() {
   // In the event of avformat_open_input() failure, FFmpeg may sometimes free
   // our AVFormatContext behind the scenes, but leave the buffer alive.  It will
   // helpfully set |format_context_| to nullptr in this case.
   if (!format_context_) {
     av_free(avio_context_->buffer);
     return;
   }
 
   // If avformat_open_input() hasn't been called, we should simply free the
   // AVFormatContext and buffer instead of using avformat_close_input().
   if (!open_called_) {
     avformat_free_context(format_context_);
     av_free(avio_context_->buffer);
     return;
   }
 
   avformat_close_input(&format_context_);
   av_free(avio_context_->buffer);
 }
 
 }  // namespace media