webrtc/modules/audio_coding/main/acm2/audio_coding_module_impl.h

/*
 *  Copyright (c) 2012 The WebRTC project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#ifndef WEBRTC_MODULES_AUDIO_CODING_MAIN_ACM2_AUDIO_CODING_MODULE_IMPL_H_
#define WEBRTC_MODULES_AUDIO_CODING_MAIN_ACM2_AUDIO_CODING_MODULE_IMPL_H_

#include <vector>

#include "webrtc/common_types.h"
#include "webrtc/engine_configurations.h"
#include "webrtc/modules/audio_coding/main/acm2/acm_codec_database.h"
#include "webrtc/modules/audio_coding/main/acm2/acm_receiver.h"
#include "webrtc/modules/audio_coding/main/acm2/acm_resampler.h"
#include "webrtc/system_wrappers/interface/scoped_ptr.h"

namespace webrtc {

class CriticalSectionWrapper;
class RWLockWrapper;

namespace acm2 {

class ACMDTMFDetection;
class ACMGenericCodec;

class AudioCodingModuleImpl : public AudioCodingModule {
 public:
  explicit AudioCodingModuleImpl(const AudioCodingModule::Config& config);
  ~AudioCodingModuleImpl();

  // Change the unique identifier of this object.
  virtual int32_t ChangeUniqueId(const int32_t id);

  // Returns the number of milliseconds until the module want a worker thread
  // to call Process.
  int32_t TimeUntilNextProcess();

  // Process any pending tasks such as timeouts.
  int32_t Process();

  /////////////////////////////////////////
  //   Sender
  //

  // Initialize send codec.
  int InitializeSender();

  // Reset send codec.
  int ResetEncoder();

  // Can be called multiple times for Codec, CNG, RED.
  int RegisterSendCodec(const CodecInst& send_codec);

  // Register Secondary codec for dual-streaming. Dual-streaming is activated
  // right after the secondary codec is registered.
  int RegisterSecondarySendCodec(const CodecInst& send_codec);

  // Unregister the secondary codec. Dual-streaming is deactivated right after
  // deregistering secondary codec.
  void UnregisterSecondarySendCodec();

  // Get the secondary codec.
  int SecondarySendCodec(CodecInst* secondary_codec) const;

  // Get current send codec.
  int SendCodec(CodecInst* current_codec) const;

  // Get current send frequency.
  int SendFrequency() const;

  // Get encode bit-rate.
  // Adaptive rate codecs return their current encode target rate, while other
  // codecs return there long-term average or their fixed rate.
  int SendBitrate() const;

  // Set available bandwidth, inform the encoder about the
  // estimated bandwidth received from the remote party.
  virtual int SetReceivedEstimatedBandwidth(int bw);

  // Register a transport callback which will be
  // called to deliver the encoded buffers.
  int RegisterTransportCallback(AudioPacketizationCallback* transport);

  // Add 10 ms of raw (PCM) audio data to the encoder.
  int Add10MsData(const AudioFrame& audio_frame);

  /////////////////////////////////////////
  // (FEC) Forward Error Correction
  //

  // Configure FEC status i.e on/off.
  int SetFECStatus(bool enable_fec);

  // Get FEC status.
  bool FECStatus() const;

  /////////////////////////////////////////
  //   (VAD) Voice Activity Detection
  //   and
  //   (CNG) Comfort Noise Generation
  //

  int SetVAD(bool enable_dtx = true,
             bool enable_vad = false,
             ACMVADMode mode = VADNormal);

  int VAD(bool* dtx_enabled, bool* vad_enabled, ACMVADMode* mode) const;

  int RegisterVADCallback(ACMVADCallback* vad_callback);

  /////////////////////////////////////////
  //   Receiver
  //

  // Initialize receiver, resets codec database etc.
  int InitializeReceiver();

  // Reset the decoder state.
  int ResetDecoder();

  // Get current receive frequency.
  int ReceiveFrequency() const;

  // Get current playout frequency.
  int PlayoutFrequency() const;

  // Register possible receive codecs, can be called multiple times,
  // for codecs, CNG, DTMF, RED.
  int RegisterReceiveCodec(const CodecInst& receive_codec);

  // Get current received codec.
  int ReceiveCodec(CodecInst* current_codec) const;

  // Incoming packet from network parsed and ready for decode.
  int IncomingPacket(const uint8_t* incoming_payload,
                     int payload_length,
                     const WebRtcRTPHeader& rtp_info);

  // Incoming payloads, without rtp-info, the rtp-info will be created in ACM.
  // One usage for this API is when pre-encoded files are pushed in ACM.
  int IncomingPayload(const uint8_t* incoming_payload,
                      int payload_length,
                      uint8_t payload_type,
                      uint32_t timestamp);

  // Minimum playout delay.
  int SetMinimumPlayoutDelay(int time_ms);

  // Maximum playout delay.
  int SetMaximumPlayoutDelay(int time_ms);

  // Smallest latency NetEq will maintain.
  int LeastRequiredDelayMs() const;

  // Impose an initial delay on playout. ACM plays silence until |delay_ms|
  // audio is accumulated in NetEq buffer, then starts decoding payloads.
  int SetInitialPlayoutDelay(int delay_ms);

  // TODO(turajs): DTMF playout is always activated in NetEq these APIs should
  // be removed, as well as all VoE related APIs and methods.
  //
  // Configure Dtmf playout status i.e on/off playout the incoming outband Dtmf
  // tone.
  int SetDtmfPlayoutStatus(bool enable) { return 0; }

  // Get Dtmf playout status.
  bool DtmfPlayoutStatus() const { return true; }

  // Estimate the Bandwidth based on the incoming stream, needed
  // for one way audio where the RTCP send the BW estimate.
  // This is also done in the RTP module .
  int DecoderEstimatedBandwidth() const;

  // Set playout mode voice, fax.
  int SetPlayoutMode(AudioPlayoutMode mode);

  // Get playout mode voice, fax.
  AudioPlayoutMode PlayoutMode() const;

  // Get playout timestamp.
  int PlayoutTimestamp(uint32_t* timestamp);

  // Get 10 milliseconds of raw audio data to play out, and
  // automatic resample to the requested frequency if > 0.
  int PlayoutData10Ms(int desired_freq_hz, AudioFrame* audio_frame);

  /////////////////////////////////////////
  //   Statistics
  //

  int NetworkStatistics(ACMNetworkStatistics* statistics);

  void DestructEncoderInst(void* inst);

  // GET RED payload for iSAC. The method id called when 'this' ACM is
  // the default ACM.
  int REDPayloadISAC(int isac_rate,
                     int isac_bw_estimate,
                     uint8_t* payload,
                     int16_t* length_bytes);

  int ReplaceInternalDTXWithWebRtc(bool use_webrtc_dtx);

  int IsInternalDTXReplacedWithWebRtc(bool* uses_webrtc_dtx);

  int SetISACMaxRate(int max_bit_per_sec);

  int SetISACMaxPayloadSize(int max_size_bytes);

  int ConfigISACBandwidthEstimator(int frame_size_ms,
                                   int rate_bit_per_sec,
                                   bool enforce_frame_size = false);

  int UnregisterReceiveCodec(uint8_t payload_type);

  int EnableNack(size_t max_nack_list_size);

  void DisableNack();

  std::vector<uint16_t> GetNackList(int round_trip_time_ms) const;

  void GetDecodingCallStatistics(AudioDecodingCallStats* stats) const;

 private:
  int UnregisterReceiveCodecSafe(int payload_type);

  ACMGenericCodec* CreateCodec(const CodecInst& codec);

  int InitializeReceiverSafe();

  bool HaveValidEncoder(const char* caller_name) const;

  // Set VAD/DTX status. This function does not acquire a lock, and it is
  // created to be called only from inside a critical section.
  int SetVADSafe(bool enable_dtx, bool enable_vad, ACMVADMode mode);

  // Process buffered audio when dual-streaming is not enabled (When RED is
  // enabled still this function is used.)
  int ProcessSingleStream();

  // Process buffered audio when dual-streaming is enabled, i.e. secondary send
  // codec is registered.
  int ProcessDualStream();

  // Preprocessing of input audio, including resampling and down-mixing if
  // required, before pushing audio into encoder's buffer.
  //
  // in_frame: input audio-frame
  // ptr_out: pointer to output audio_frame. If no preprocessing is required
  //          |ptr_out| will be pointing to |in_frame|, otherwise pointing to
  //          |preprocess_frame_|.
  //
  // Return value:
  //   -1: if encountering an error.
  //    0: otherwise.
  int PreprocessToAddData(const AudioFrame& in_frame,
                          const AudioFrame** ptr_out);

  // Change required states after starting to receive the codec corresponding
  // to |index|.
  int UpdateUponReceivingCodec(int index);

  int EncodeFragmentation(int fragmentation_index, int payload_type,
                          uint32_t current_timestamp,
                          ACMGenericCodec* encoder,
                          uint8_t* stream);

  void ResetFragmentation(int vector_size);

  // Get a pointer to AudioDecoder of the given codec. For some codecs, e.g.
  // iSAC, encoding and decoding have to be performed on a shared
  // codec-instance. By calling this method, we get the codec-instance that ACM
  // owns, then pass that to NetEq. This way, we perform both encoding and
  // decoding on the same codec-instance. Furthermore, ACM would have control
  // over decoder functionality if required. If |codec| does not share an
  // instance between encoder and decoder, the |*decoder| is set NULL.
  // The field ACMCodecDB::CodecSettings.owns_decoder indicates that if a
  // codec owns the decoder-instance. For such codecs |*decoder| should be a
  // valid pointer, otherwise it will be NULL.
  int GetAudioDecoder(const CodecInst& codec, int codec_id,
                      int mirror_id, AudioDecoder** decoder);

  AudioPacketizationCallback* packetization_callback_;

  int id_;
  uint32_t expected_codec_ts_;
  uint32_t expected_in_ts_;
  CodecInst send_codec_inst_;

  uint8_t cng_nb_pltype_;
  uint8_t cng_wb_pltype_;
  uint8_t cng_swb_pltype_;
  uint8_t cng_fb_pltype_;

  uint8_t red_pltype_;
  bool vad_enabled_;
  bool dtx_enabled_;
  ACMVADMode vad_mode_;
  ACMGenericCodec* codecs_[ACMCodecDB::kMaxNumCodecs];
  int mirror_codec_idx_[ACMCodecDB::kMaxNumCodecs];
  bool stereo_send_;
  int current_send_codec_idx_;
  bool send_codec_registered_;
  ACMResampler resampler_;
  AcmReceiver receiver_;
  CriticalSectionWrapper* acm_crit_sect_;
  ACMVADCallback* vad_callback_;

  // RED/FEC.
  bool is_first_red_;
  bool fec_enabled_;

  // TODO(turajs): |red_buffer_| is allocated in constructor, why having them
  // as pointers and not an array. If concerned about the memory, then make a
  // set-up function to allocate them only when they are going to be used, i.e.
  // FEC or Dual-streaming is enabled.
  uint8_t* red_buffer_;

  // TODO(turajs): we actually don't need |fragmentation_| as a member variable.
  // It is sufficient to keep the length & payload type of previous payload in
  // member variables.
  RTPFragmentationHeader fragmentation_;
  uint32_t last_fec_timestamp_;

  // This is to keep track of CN instances where we can send DTMFs.
  uint8_t previous_pltype_;

  // Used when payloads are pushed into ACM without any RTP info
  // One example is when pre-encoded bit-stream is pushed from
  // a file.
  // IMPORTANT: this variable is only used in IncomingPayload(), therefore,
  // no lock acquired when interacting with this variable. If it is going to
  // be used in other methods, locks need to be taken.
  WebRtcRTPHeader* aux_rtp_header_;

  bool receiver_initialized_;

  CriticalSectionWrapper* callback_crit_sect_;

  AudioFrame preprocess_frame_;
  CodecInst secondary_send_codec_inst_;
  scoped_ptr<ACMGenericCodec> secondary_encoder_;
  uint32_t codec_timestamp_;
  bool first_10ms_data_;
};

}  // namespace acm2

}  // namespace webrtc

#endif  // WEBRTC_MODULES_AUDIO_CODING_MAIN_ACM2_AUDIO_CODING_MODULE_IMPL_H_