Refactor WVoE DTMF handling #2

talk/media/webrtc/webrtcvoiceengine.h

Index: talk/media/webrtc/webrtcvoiceengine.h
diff --git a/talk/media/webrtc/webrtcvoiceengine.h b/talk/media/webrtc/webrtcvoiceengine.h
index 843ff797de3d8cecad9cc9947c66e077aba88562..89cf25c25f971cf2471f50cee937e138a5a7624e 100644
--- a/talk/media/webrtc/webrtcvoiceengine.h
+++ b/talk/media/webrtc/webrtcvoiceengine.h
@@ -265,7 +265,7 @@ class WebRtcVoiceMediaChannel final : public VoiceMediaChannel,
   bool send_bitrate_setting_ = false;
   int send_bitrate_bps_ = 0;
   AudioOptions options_;
-  bool dtmf_allowed_ = false;
+  rtc::Optional<int> dtmf_payload_type_;

[pthatcher1, 2015/12/02 18:45:37]

Why do we store this?  Why not just look it up from the list of send_codecs_
whenever we need it?  DTMF is sent so infrequently that perf isn't an issue. 
And the code would be cleaner, I think, if it just reads it on demand.

[the sun, 2015/12/02 19:40:23]

Perhaps it would be a little cleaner if we did that, but note that it will need
to be done in two places: CanInsertDtmf() and InsertDtmf() and, going forward,
the payload type will be added to the AudioSendStream::Config so there will
nevertheless be code under SetSendCodecs() which figures out whether to enable
DTMF or not.

If it is ok with you I'd like to keep it as now and revisit when we refactor
codec setting to use RAC.

   bool desired_playout_ = false;
   bool nack_enabled_ = false;
   bool playout_ = false;