Add the ability to read/write to WAV files in FakeAudioDevice

webrtc/call/call_perf_tests.cc

Index: webrtc/call/call_perf_tests.cc
diff --git a/webrtc/call/call_perf_tests.cc b/webrtc/call/call_perf_tests.cc
index 1551cd7b86791f61989196a5f861f60d201ca6ed..a400d408b2060f1a80d57fc8ba5b57fbd259a5ec 100644
--- a/webrtc/call/call_perf_tests.cc
+++ b/webrtc/call/call_perf_tests.cc
@@ -146,7 +146,9 @@ void CallPerfTest::TestAudioVideoSync(FecMode fec,
   metrics::Reset();
   VoiceEngine* voice_engine = VoiceEngine::Create();
   VoEBase* voe_base = VoEBase::GetInterface(voice_engine);
-  FakeAudioDevice fake_audio_device(audio_rtp_speed, 48000, 256);
+  FakeAudioDevice fake_audio_device(
+      FakeAudioDevice::CreatePulsedNoiseCapturer(256, 48000),
+      FakeAudioDevice::CreateDiscarder(48000), audio_rtp_speed);

[kwiberg-webrtc, 2017/03/13 10:18:21]

This looks very nice!

Hmm. You don't have to, but call sites like this one might get even easier to
read if CreateDiscarder was called CreateDiscardRenderer instead---in general,
if the capturers had "Capturer" in their names, and the renderers "Renderer".

[oprypin_webrtc, 2017/03/13 13:55:14]

Changed to CreateDiscardRenderer, but it's hard to do the same for
'CreateWavFileWriter' because if I include 'Renderer' there, it would be
confusing as it does not render WAV files...

[kwiberg-webrtc, 2017/03/13 14:22:26]

Fair point.

   EXPECT_EQ(0, voe_base->Init(&fake_audio_device, nullptr, decoder_factory_));
   VoEBase::ChannelConfig config;
   config.enable_voice_pacing = true;