path: root/webrtc/modules/rtp_rtcp/source/rtp_sender_audio.cc

/*
 *  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.
 */

#include "webrtc/modules/rtp_rtcp/source/rtp_sender_audio.h"

#include <string.h>

#include "webrtc/base/trace_event.h"
#include "webrtc/modules/rtp_rtcp/include/rtp_rtcp_defines.h"
#include "webrtc/modules/rtp_rtcp/source/byte_io.h"
#include "webrtc/system_wrappers/include/tick_util.h"

namespace webrtc {

static const int kDtmfFrequencyHz = 8000;

RTPSenderAudio::RTPSenderAudio(Clock* clock,
                               RTPSender* rtpSender,
                               RtpAudioFeedback* audio_feedback)
    : _clock(clock),
      _rtpSender(rtpSender),
      _audioFeedback(audio_feedback),
      _sendAudioCritsect(CriticalSectionWrapper::CreateCriticalSection()),
      _packetSizeSamples(160),
      _dtmfEventIsOn(false),
      _dtmfEventFirstPacketSent(false),
      _dtmfPayloadType(-1),
      _dtmfTimestamp(0),
      _dtmfKey(0),
      _dtmfLengthSamples(0),
      _dtmfLevel(0),
      _dtmfTimeLastSent(0),
      _dtmfTimestampLastSent(0),
      _REDPayloadType(-1),
      _inbandVADactive(false),
      _cngNBPayloadType(-1),
      _cngWBPayloadType(-1),
      _cngSWBPayloadType(-1),
      _cngFBPayloadType(-1),
      _lastPayloadType(-1),
      _audioLevel_dBov(0) {}

RTPSenderAudio::~RTPSenderAudio() {}

int RTPSenderAudio::AudioFrequency() const {
  return kDtmfFrequencyHz;
}

// set audio packet size, used to determine when it's time to send a DTMF packet
// in silence (CNG)
int32_t RTPSenderAudio::SetAudioPacketSize(uint16_t packetSizeSamples) {
  CriticalSectionScoped cs(_sendAudioCritsect.get());

  _packetSizeSamples = packetSizeSamples;
  return 0;
}

int32_t RTPSenderAudio::RegisterAudioPayload(
    const char payloadName[RTP_PAYLOAD_NAME_SIZE],
    const int8_t payloadType,
    const uint32_t frequency,
    const size_t channels,
    const uint32_t rate,
    RtpUtility::Payload** payload) {
  if (RtpUtility::StringCompare(payloadName, "cn", 2)) {
    CriticalSectionScoped cs(_sendAudioCritsect.get());
    //  we can have multiple CNG payload types
    switch (frequency) {
      case 8000:
        _cngNBPayloadType = payloadType;
        break;
      case 16000:
        _cngWBPayloadType = payloadType;
        break;
      case 32000:
        _cngSWBPayloadType = payloadType;
        break;
      case 48000:
        _cngFBPayloadType = payloadType;
        break;
      default:
        return -1;
    }
  } else if (RtpUtility::StringCompare(payloadName, "telephone-event", 15)) {
    CriticalSectionScoped cs(_sendAudioCritsect.get());
    // Don't add it to the list
    // we dont want to allow send with a DTMF payloadtype
    _dtmfPayloadType = payloadType;
    return 0;
    // The default timestamp rate is 8000 Hz, but other rates may be defined.
  }
  *payload = new RtpUtility::Payload;
  (*payload)->typeSpecific.Audio.frequency = frequency;
  (*payload)->typeSpecific.Audio.channels = channels;
  (*payload)->typeSpecific.Audio.rate = rate;
  (*payload)->audio = true;
  (*payload)->name[RTP_PAYLOAD_NAME_SIZE - 1] = '\0';
  strncpy((*payload)->name, payloadName, RTP_PAYLOAD_NAME_SIZE - 1);
  return 0;
}

bool RTPSenderAudio::MarkerBit(FrameType frameType, int8_t payload_type) {
  CriticalSectionScoped cs(_sendAudioCritsect.get());
  // for audio true for first packet in a speech burst
  bool markerBit = false;
  if (_lastPayloadType != payload_type) {
    if (payload_type != -1 && (_cngNBPayloadType == payload_type ||
                               _cngWBPayloadType == payload_type ||
                               _cngSWBPayloadType == payload_type ||
                               _cngFBPayloadType == payload_type)) {
      // Only set a marker bit when we change payload type to a non CNG
      return false;
    }

    // payload_type differ
    if (_lastPayloadType == -1) {
      if (frameType != kAudioFrameCN) {
        // first packet and NOT CNG
        return true;
      } else {
        // first packet and CNG
        _inbandVADactive = true;
        return false;
      }
    }

    // not first packet AND
    // not CNG AND
    // payload_type changed

    // set a marker bit when we change payload type
    markerBit = true;
  }

  // For G.723 G.729, AMR etc we can have inband VAD
  if (frameType == kAudioFrameCN) {
    _inbandVADactive = true;
  } else if (_inbandVADactive) {
    _inbandVADactive = false;
    markerBit = true;
  }
  return markerBit;
}

int32_t RTPSenderAudio::SendAudio(FrameType frameType,
                                  int8_t payloadType,
                                  uint32_t captureTimeStamp,
                                  const uint8_t* payloadData,
                                  size_t dataSize,
                                  const RTPFragmentationHeader* fragmentation) {
  // TODO(pwestin) Breakup function in smaller functions.
  size_t payloadSize = dataSize;
  size_t maxPayloadLength = _rtpSender->MaxPayloadLength();
  bool dtmfToneStarted = false;
  uint16_t dtmfLengthMS = 0;
  uint8_t key = 0;
  int red_payload_type;
  uint8_t audio_level_dbov;
  int8_t dtmf_payload_type;
  uint16_t packet_size_samples;
  {
    CriticalSectionScoped cs(_sendAudioCritsect.get());
    red_payload_type = _REDPayloadType;
    audio_level_dbov = _audioLevel_dBov;
    dtmf_payload_type = _dtmfPayloadType;
    packet_size_samples = _packetSizeSamples;
  }

  // Check if we have pending DTMFs to send
  if (!_dtmfEventIsOn && PendingDTMF()) {
    int64_t delaySinceLastDTMF =
        _clock->TimeInMilliseconds() - _dtmfTimeLastSent;

    if (delaySinceLastDTMF > 100) {
      // New tone to play
      _dtmfTimestamp = captureTimeStamp;
      if (NextDTMF(&key, &dtmfLengthMS, &_dtmfLevel) >= 0) {
        _dtmfEventFirstPacketSent = false;
        _dtmfKey = key;
        _dtmfLengthSamples = (kDtmfFrequencyHz / 1000) * dtmfLengthMS;
        dtmfToneStarted = true;
        _dtmfEventIsOn = true;
      }
    }
  }
  if (dtmfToneStarted) {
    if (_audioFeedback)
      _audioFeedback->OnPlayTelephoneEvent(key, dtmfLengthMS, _dtmfLevel);
  }

  // A source MAY send events and coded audio packets for the same time
  // but we don't support it
  if (_dtmfEventIsOn) {
    if (frameType == kEmptyFrame) {
      // kEmptyFrame is used to drive the DTMF when in CN mode
      // it can be triggered more frequently than we want to send the
      // DTMF packets.
      if (packet_size_samples > (captureTimeStamp - _dtmfTimestampLastSent)) {
        // not time to send yet
        return 0;
      }
    }
    _dtmfTimestampLastSent = captureTimeStamp;
    uint32_t dtmfDurationSamples = captureTimeStamp - _dtmfTimestamp;
    bool ended = false;
    bool send = true;

    if (_dtmfLengthSamples > dtmfDurationSamples) {
      if (dtmfDurationSamples <= 0) {
        // Skip send packet at start, since we shouldn't use duration 0
        send = false;
      }
    } else {
      ended = true;
      _dtmfEventIsOn = false;
      _dtmfTimeLastSent = _clock->TimeInMilliseconds();
    }
    if (send) {
      if (dtmfDurationSamples > 0xffff) {
        // RFC 4733 2.5.2.3 Long-Duration Events
        SendTelephoneEventPacket(ended, dtmf_payload_type, _dtmfTimestamp,
                                 static_cast<uint16_t>(0xffff), false);

        // set new timestap for this segment
        _dtmfTimestamp = captureTimeStamp;
        dtmfDurationSamples -= 0xffff;
        _dtmfLengthSamples -= 0xffff;

        return SendTelephoneEventPacket(
            ended, dtmf_payload_type, _dtmfTimestamp,
            static_cast<uint16_t>(dtmfDurationSamples), false);
      } else {
        if (SendTelephoneEventPacket(ended, dtmf_payload_type, _dtmfTimestamp,
                                     static_cast<uint16_t>(dtmfDurationSamples),
                                     !_dtmfEventFirstPacketSent) != 0) {
          return -1;
        }
        _dtmfEventFirstPacketSent = true;
        return 0;
      }
    }
    return 0;
  }
  if (payloadSize == 0 || payloadData == NULL) {
    if (frameType == kEmptyFrame) {
      // we don't send empty audio RTP packets
      // no error since we use it to drive DTMF when we use VAD
      return 0;
    }
    return -1;
  }
  uint8_t dataBuffer[IP_PACKET_SIZE];
  bool markerBit = MarkerBit(frameType, payloadType);

  int32_t rtpHeaderLength = 0;
  uint16_t timestampOffset = 0;

  if (red_payload_type >= 0 && fragmentation && !markerBit &&
      fragmentation->fragmentationVectorSize > 1) {
    // have we configured RED? use its payload type
    // we need to get the current timestamp to calc the diff
    uint32_t oldTimeStamp = _rtpSender->Timestamp();
    rtpHeaderLength = _rtpSender->BuildRTPheader(dataBuffer, red_payload_type,
                                                 markerBit, captureTimeStamp,
                                                 _clock->TimeInMilliseconds());

    timestampOffset = uint16_t(_rtpSender->Timestamp() - oldTimeStamp);
  } else {
    rtpHeaderLength = _rtpSender->BuildRTPheader(dataBuffer, payloadType,
                                                 markerBit, captureTimeStamp,
                                                 _clock->TimeInMilliseconds());
  }
  if (rtpHeaderLength <= 0) {
    return -1;
  }
  if (maxPayloadLength < (rtpHeaderLength + payloadSize)) {
    // Too large payload buffer.
    return -1;
  }
  if (red_payload_type >= 0 &&  // Have we configured RED?
      fragmentation && fragmentation->fragmentationVectorSize > 1 &&
      !markerBit) {
    if (timestampOffset <= 0x3fff) {
      if (fragmentation->fragmentationVectorSize != 2) {
        // we only support 2 codecs when using RED
        return -1;
      }
      // only 0x80 if we have multiple blocks
      dataBuffer[rtpHeaderLength++] =
          0x80 + fragmentation->fragmentationPlType[1];
      size_t blockLength = fragmentation->fragmentationLength[1];

      // sanity blockLength
      if (blockLength > 0x3ff) {  // block length 10 bits 1023 bytes
        return -1;
      }
      uint32_t REDheader = (timestampOffset << 10) + blockLength;
      ByteWriter<uint32_t>::WriteBigEndian(dataBuffer + rtpHeaderLength,
                                           REDheader);
      rtpHeaderLength += 3;

      dataBuffer[rtpHeaderLength++] = fragmentation->fragmentationPlType[0];
      // copy the RED data
      memcpy(dataBuffer + rtpHeaderLength,
             payloadData + fragmentation->fragmentationOffset[1],
             fragmentation->fragmentationLength[1]);

      // copy the normal data
      memcpy(
          dataBuffer + rtpHeaderLength + fragmentation->fragmentationLength[1],
          payloadData + fragmentation->fragmentationOffset[0],
          fragmentation->fragmentationLength[0]);

      payloadSize = fragmentation->fragmentationLength[0] +
                    fragmentation->fragmentationLength[1];
    } else {
      // silence for too long send only new data
      dataBuffer[rtpHeaderLength++] = fragmentation->fragmentationPlType[0];
      memcpy(dataBuffer + rtpHeaderLength,
             payloadData + fragmentation->fragmentationOffset[0],
             fragmentation->fragmentationLength[0]);

      payloadSize = fragmentation->fragmentationLength[0];
    }
  } else {
    if (fragmentation && fragmentation->fragmentationVectorSize > 0) {
      // use the fragment info if we have one
      dataBuffer[rtpHeaderLength++] = fragmentation->fragmentationPlType[0];
      memcpy(dataBuffer + rtpHeaderLength,
             payloadData + fragmentation->fragmentationOffset[0],
             fragmentation->fragmentationLength[0]);

      payloadSize = fragmentation->fragmentationLength[0];
    } else {
      memcpy(dataBuffer + rtpHeaderLength, payloadData, payloadSize);
    }
  }
  {
    CriticalSectionScoped cs(_sendAudioCritsect.get());
    _lastPayloadType = payloadType;
  }
  // Update audio level extension, if included.
  size_t packetSize = payloadSize + rtpHeaderLength;
  RtpUtility::RtpHeaderParser rtp_parser(dataBuffer, packetSize);
  RTPHeader rtp_header;
  rtp_parser.Parse(&rtp_header);
  _rtpSender->UpdateAudioLevel(dataBuffer, packetSize, rtp_header,
                               (frameType == kAudioFrameSpeech),
                               audio_level_dbov);
  TRACE_EVENT_ASYNC_END2("webrtc", "Audio", captureTimeStamp, "timestamp",
                         _rtpSender->Timestamp(), "seqnum",
                         _rtpSender->SequenceNumber());
  return _rtpSender->SendToNetwork(dataBuffer, payloadSize, rtpHeaderLength,
                                   TickTime::MillisecondTimestamp(),
                                   kAllowRetransmission,
                                   RtpPacketSender::kHighPriority);
}

// Audio level magnitude and voice activity flag are set for each RTP packet
int32_t RTPSenderAudio::SetAudioLevel(uint8_t level_dBov) {
  if (level_dBov > 127) {
    return -1;
  }
  CriticalSectionScoped cs(_sendAudioCritsect.get());
  _audioLevel_dBov = level_dBov;
  return 0;
}

// Set payload type for Redundant Audio Data RFC 2198
int32_t RTPSenderAudio::SetRED(int8_t payloadType) {
  if (payloadType < -1) {
    return -1;
  }
  CriticalSectionScoped cs(_sendAudioCritsect.get());
  _REDPayloadType = payloadType;
  return 0;
}

// Get payload type for Redundant Audio Data RFC 2198
int32_t RTPSenderAudio::RED(int8_t* payloadType) const {
  CriticalSectionScoped cs(_sendAudioCritsect.get());
  if (_REDPayloadType == -1) {
    // not configured
    return -1;
  }
  *payloadType = _REDPayloadType;
  return 0;
}

// Send a TelephoneEvent tone using RFC 2833 (4733)
int32_t RTPSenderAudio::SendTelephoneEvent(uint8_t key,
                                           uint16_t time_ms,
                                           uint8_t level) {
  {
    CriticalSectionScoped lock(_sendAudioCritsect.get());
    if (_dtmfPayloadType < 0) {
      // TelephoneEvent payloadtype not configured
      return -1;
    }
  }
  return AddDTMF(key, time_ms, level);
}

int32_t RTPSenderAudio::SendTelephoneEventPacket(bool ended,
                                                 int8_t dtmf_payload_type,
                                                 uint32_t dtmfTimeStamp,
                                                 uint16_t duration,
                                                 bool markerBit) {
  uint8_t dtmfbuffer[IP_PACKET_SIZE];
  uint8_t sendCount = 1;
  int32_t retVal = 0;

  if (ended) {
    // resend last packet in an event 3 times
    sendCount = 3;
  }
  do {
    // Send DTMF data
    _rtpSender->BuildRTPheader(dtmfbuffer, dtmf_payload_type, markerBit,
                               dtmfTimeStamp, _clock->TimeInMilliseconds());

    // reset CSRC and X bit
    dtmfbuffer[0] &= 0xe0;

    // Create DTMF data
    /*    From RFC 2833:

     0                   1                   2                   3
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |     event     |E|R| volume    |          duration             |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    */
    // R bit always cleared
    uint8_t R = 0x00;
    uint8_t volume = _dtmfLevel;

    // First packet un-ended
    uint8_t E = ended ? 0x80 : 0x00;

    // First byte is Event number, equals key number
    dtmfbuffer[12] = _dtmfKey;
    dtmfbuffer[13] = E | R | volume;
    ByteWriter<uint16_t>::WriteBigEndian(dtmfbuffer + 14, duration);

    TRACE_EVENT_INSTANT2(TRACE_DISABLED_BY_DEFAULT("webrtc_rtp"),
                         "Audio::SendTelephoneEvent", "timestamp",
                         dtmfTimeStamp, "seqnum", _rtpSender->SequenceNumber());
    retVal = _rtpSender->SendToNetwork(
        dtmfbuffer, 4, 12, TickTime::MillisecondTimestamp(),
        kAllowRetransmission, RtpPacketSender::kHighPriority);
    sendCount--;
  } while (sendCount > 0 && retVal == 0);

  return retVal;
}
}  // namespace webrtc