Change NetEq::GetAudio to use AudioFrame

webrtc/modules/audio_coding/neteq/audio_multi_vector.cc

Index: webrtc/modules/audio_coding/neteq/audio_multi_vector.cc
diff --git a/webrtc/modules/audio_coding/neteq/audio_multi_vector.cc b/webrtc/modules/audio_coding/neteq/audio_multi_vector.cc
index 1381895085904fb731699a41e9f0295f5aada822..bd38c43903a4106985458605426612964164b0b6 100644
--- a/webrtc/modules/audio_coding/neteq/audio_multi_vector.cc
+++ b/webrtc/modules/audio_coding/neteq/audio_multi_vector.cc
@@ -14,6 +14,7 @@
 
 #include <algorithm>
 
+#include "webrtc/base/checks.h"
 #include "webrtc/typedefs.h"
 
 namespace webrtc {
@@ -130,9 +131,7 @@ size_t AudioMultiVector::ReadInterleaved(size_t length,
 size_t AudioMultiVector::ReadInterleavedFromIndex(size_t start_index,
                                                   size_t length,
                                                   int16_t* destination) const {
-  if (!destination) {
-    return 0;
-  }
+  RTC_DCHECK(destination);

[hlundin-webrtc, 2016/03/04 09:55:32]

This change is just to keep consistency with
SyncBuffer::GetNextAudioInterleaved.

   size_t index = 0;  // Number of elements written to |destination| so far.
   assert(start_index <= Size());
   start_index = std::min(start_index, Size());