jamulus/src/client.cpp
2009-08-11 09:10:23 +00:00

510 lines
17 KiB
C++
Executable File

/******************************************************************************\
* Copyright (c) 2004-2009
*
* Author(s):
* Volker Fischer
*
******************************************************************************
*
* This program is free software; you can redistribute it and/or modify it under
* the terms of the GNU General Public License as published by the Free Software
* Foundation; either version 2 of the License, or (at your option) any later
* version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
* FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
* details.
*
* You should have received a copy of the GNU General Public License along with
* this program; if not, write to the Free Software Foundation, Inc.,
* 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
\******************************************************************************/
#include "client.h"
/* Implementation *************************************************************/
CClient::CClient ( const quint16 iPortNumber ) :
Channel ( false ), /* we need a client channel -> "false" */
Sound ( AudioCallback, this ),
Socket ( &Channel, iPortNumber ),
iAudioInFader ( AUD_FADER_IN_MIDDLE ),
iReverbLevel ( 0 ),
bReverbOnLeftChan ( false ),
vstrIPAddress ( MAX_NUM_SERVER_ADDR_ITEMS, "" ), strName ( "" ),
bOpenChatOnNewMessage ( true ),
bDoAutoSockBufSize ( true ),
iSndCrdPrefMonoFrameSizeFactor ( FRAME_SIZE_FACTOR_DEFAULT )
{
// connection for protocol
QObject::connect ( &Channel,
SIGNAL ( MessReadyForSending ( CVector<uint8_t> ) ),
this, SLOT ( OnSendProtMessage ( CVector<uint8_t> ) ) );
QObject::connect ( &Channel, SIGNAL ( ReqJittBufSize() ),
this, SLOT ( OnReqJittBufSize() ) );
QObject::connect ( &Channel,
SIGNAL ( ConClientListMesReceived ( CVector<CChannelShortInfo> ) ),
SIGNAL ( ConClientListMesReceived ( CVector<CChannelShortInfo> ) ) );
QObject::connect ( &Channel, SIGNAL ( NewConnection() ),
this, SLOT ( OnNewConnection() ) );
QObject::connect ( &Channel, SIGNAL ( ChatTextReceived ( QString ) ),
this, SIGNAL ( ChatTextReceived ( QString ) ) );
QObject::connect ( &Channel, SIGNAL ( PingReceived ( int ) ),
this, SLOT ( OnReceivePingMessage ( int ) ) );
QObject::connect ( &Sound, SIGNAL ( ReinitRequest() ),
this, SLOT ( OnSndCrdReinitRequest() ) );
}
void CClient::OnSendProtMessage ( CVector<uint8_t> vecMessage )
{
// the protocol queries me to call the function to send the message
// send it through the network
Socket.SendPacket ( vecMessage, Channel.GetAddress() );
}
void CClient::OnReqJittBufSize()
{
// TODO cant we implement this OnReqJjittBufSize inside the channel object?
Channel.CreateJitBufMes ( Channel.GetSockBufNumFrames() );
}
void CClient::OnNewConnection()
{
// a new connection was successfully initiated, send name and request
// connected clients list
Channel.SetRemoteName ( strName );
// We have to send a connected clients list request since it can happen
// that we just had connected to the server and then disconnected but
// the server still thinks that we are connected (the server is still
// waiting for the channel time-out). If we now connect again, we would
// not get the list because the server does not know about a new connection.
Channel.CreateReqConnClientsList();
}
void CClient::OnReceivePingMessage ( int iMs )
{
// calculate difference between received time in ms and current time in ms,
// take care of wrap arounds (if wrapping, do not use result)
const int iCurDiff = PreciseTime.elapsed() - iMs;
if ( iCurDiff >= 0 )
{
emit PingTimeReceived ( iCurDiff );
}
}
bool CClient::SetServerAddr ( QString strNAddr )
{
QHostAddress InetAddr;
quint16 iNetPort = LLCON_DEFAULT_PORT_NUMBER;
// parse input address for the type [IP address]:[port number]
QString strPort = strNAddr.section ( ":", 1, 1 );
if ( !strPort.isEmpty() )
{
// a colon is present in the address string, try to extract port number
iNetPort = strPort.toInt();
// extract address port before colon (should be actual internet address)
strNAddr = strNAddr.section ( ":", 0, 0 );
}
// first try if this is an IP number an can directly applied to QHostAddress
if ( !InetAddr.setAddress ( strNAddr ) )
{
// it was no vaild IP address, try to get host by name, assuming
// that the string contains a valid host name string
QHostInfo HostInfo = QHostInfo::fromName ( strNAddr );
if ( HostInfo.error() == QHostInfo::NoError )
{
// apply IP address to QT object
if ( !HostInfo.addresses().isEmpty() )
{
// use the first IP address
InetAddr = HostInfo.addresses().first();
}
}
else
{
return false; // invalid address
}
}
// apply address (the server port is fixed and always the same)
Channel.SetAddress ( CHostAddress ( InetAddr, iNetPort ) );
return true;
}
void CClient::SetSndCrdPrefMonoFrameSizeFactor ( const int iNewFactor )
{
// right now we simply set the internal value
if ( ( iNewFactor == FRAME_SIZE_FACTOR_PREFERRED ) ||
( iNewFactor == FRAME_SIZE_FACTOR_DEFAULT ) ||
( iNewFactor == FRAME_SIZE_FACTOR_SAFE ) )
{
iSndCrdPrefMonoFrameSizeFactor = iNewFactor;
// init with new parameter, if client was running then first
// stop it and restart again after new initialization
const bool bWasRunning = Sound.IsRunning();
if ( bWasRunning )
{
Sound.Stop();
}
// init with new block size index parameter
Init();
if ( bWasRunning )
{
Sound.Start();
}
}
}
QString CClient::SetSndCrdDev ( const int iNewDev )
{
// if client was running then first
// stop it and restart again after new initialization
const bool bWasRunning = Sound.IsRunning();
if ( bWasRunning )
{
Sound.Stop();
}
const QString strReturn = Sound.SetDev ( iNewDev ).c_str();
// init again because the sound card actual buffer size might
// be changed on new device
Init();
if ( bWasRunning )
{
Sound.Start();
}
return strReturn;
}
void CClient::OnSndCrdReinitRequest()
{
// if client was running then first
// stop it and restart again after new initialization
const bool bWasRunning = Sound.IsRunning();
if ( bWasRunning )
{
Sound.Stop();
}
// reinit the driver (we use the currently selected driver) and
// init client object, too
Sound.SetDev ( Sound.GetDev() );
Init();
if ( bWasRunning )
{
Sound.Start();
}
}
void CClient::Start()
{
// init object
Init();
// enable channel
Channel.SetEnable ( true );
// start audio interface
Sound.Start();
}
void CClient::Stop()
{
// stop audio interface
Sound.Stop();
// send disconnect message to server (since we disable our protocol
// receive mechanism with the next command, we do not evaluate any
// respond from the server, therefore we just hope that the message
// gets its way to the server, if not, the old behaviour time-out
// disconnects the connection anyway)
Channel.CreateDisconnectionMes();
// disable channel
Channel.SetEnable ( false );
// reset current signal level and LEDs
SignalLevelMeter.Reset();
PostWinMessage ( MS_RESET_ALL, 0 );
}
void CClient::AudioCallback ( CVector<int16_t>& psData, void* arg )
{
// get the pointer to the object
CClient* pMyClientObj = reinterpret_cast<CClient*> ( arg );
// process audio data
pMyClientObj->ProcessAudioData ( psData );
}
void CClient::Init()
{
// translate block size index in actual block size
const int iPrefMonoFrameSize =
iSndCrdPrefMonoFrameSizeFactor * SYSTEM_BLOCK_FRAME_SAMPLES;
// get actual sound card buffer size using preferred size
iMonoBlockSizeSam = Sound.Init ( iPrefMonoFrameSize );
iStereoBlockSizeSam = 2 * iMonoBlockSizeSam;
vecsAudioSndCrdStereo.Init ( iStereoBlockSizeSam );
vecdAudioStereo.Init ( iStereoBlockSizeSam );
// init response time evaluation
CycleTimeVariance.Init ( iMonoBlockSizeSam,
SYSTEM_SAMPLE_RATE, TIME_MOV_AV_RESPONSE );
CycleTimeVariance.Reset();
// init reverberation
AudioReverb.Init ( SYSTEM_SAMPLE_RATE );
// init audio endocder/decoder (mono)
CeltMode = celt_mode_create (
SYSTEM_SAMPLE_RATE, 1, iMonoBlockSizeSam, NULL );
CeltEncoder = celt_encoder_create ( CeltMode );
CeltDecoder = celt_decoder_create ( CeltMode );
// 16: low/normal quality 132 kbsp (128) / 90 kbps (256)
// 40: high quality 204 kbps (128) / 162 kbps (256)
iCeltNumCodedBytes = 16;
vecCeltData.Init ( iCeltNumCodedBytes );
// init network buffers
vecsNetwork.Init ( iMonoBlockSizeSam );
vecbyNetwData.Init ( iCeltNumCodedBytes );
// set the channel network properties
Channel.SetNetwFrameSizeAndFact ( iCeltNumCodedBytes,
iSndCrdPrefMonoFrameSizeFactor );
}
void CClient::ProcessAudioData ( CVector<int16_t>& vecsStereoSndCrd )
{
int i, j;
// update stereo signal level meter
SignalLevelMeter.Update ( vecsStereoSndCrd );
// convert data from short to double
for ( i = 0; i < iStereoBlockSizeSam; i++ )
{
vecdAudioStereo[i] = (double) vecsStereoSndCrd[i];
}
// add reverberation effect if activated
if ( iReverbLevel != 0 )
{
// calculate attenuation amplification factor
const double dRevLev = (double) iReverbLevel / AUD_REVERB_MAX / 2;
if ( bReverbOnLeftChan )
{
for ( i = 0; i < iStereoBlockSizeSam; i += 2 )
{
// left channel
vecdAudioStereo[i] +=
dRevLev * AudioReverb.ProcessSample ( vecdAudioStereo[i] );
}
}
else
{
for ( i = 1; i < iStereoBlockSizeSam; i += 2 )
{
// right channel
vecdAudioStereo[i] +=
dRevLev * AudioReverb.ProcessSample ( vecdAudioStereo[i] );
}
}
}
// mix both signals depending on the fading setting, convert
// from double to short
if ( iAudioInFader == AUD_FADER_IN_MIDDLE )
{
// just mix channels together
for ( i = 0, j = 0; i < iMonoBlockSizeSam; i++, j += 2 )
{
vecsNetwork[i] =
Double2Short ( ( vecdAudioStereo[j] +
vecdAudioStereo[j + 1] ) / 2 );
}
}
else
{
const double dAttFact =
(double) ( AUD_FADER_IN_MIDDLE - abs ( AUD_FADER_IN_MIDDLE - iAudioInFader ) ) /
AUD_FADER_IN_MIDDLE;
if ( iAudioInFader > AUD_FADER_IN_MIDDLE )
{
for ( i = 0, j = 0; i < iMonoBlockSizeSam; i++, j += 2 )
{
// attenuation on right channel
vecsNetwork[i] =
Double2Short ( ( vecdAudioStereo[j] +
dAttFact * vecdAudioStereo[j + 1] ) / 2 );
}
}
else
{
for ( i = 0, j = 0; i < iMonoBlockSizeSam; i++, j += 2 )
{
// attenuation on left channel
vecsNetwork[i] =
Double2Short ( ( vecdAudioStereo[j + 1] +
dAttFact * vecdAudioStereo[j] ) / 2 );
}
}
}
// encode current audio frame with CELT encoder
celt_encode ( CeltEncoder,
&vecsNetwork[0],
NULL,
&vecCeltData[0],
iCeltNumCodedBytes );
// send coded audio through the network
Socket.SendPacket ( Channel.PrepSendPacket ( vecCeltData ),
Channel.GetAddress() );
// receive a new block
const bool bReceiveDataOk =
( Channel.GetData ( vecbyNetwData ) == GS_BUFFER_OK );
if ( bReceiveDataOk )
{
PostWinMessage ( MS_JIT_BUF_GET, MUL_COL_LED_GREEN );
}
else
{
PostWinMessage ( MS_JIT_BUF_GET, MUL_COL_LED_RED );
}
// check if channel is connected
if ( Channel.IsConnected() )
{
CVector<short> vecsAudioSndCrdMono ( iMonoBlockSizeSam );
// CELT decoding
if ( bReceiveDataOk )
{
celt_decode ( CeltDecoder,
&vecbyNetwData[0],
iCeltNumCodedBytes,
&vecsAudioSndCrdMono[0] );
}
else
{
// lost packet
celt_decode ( CeltDecoder,
NULL,
iCeltNumCodedBytes,
&vecsAudioSndCrdMono[0] );
}
// copy mono data in stereo sound card buffer
for ( i = 0, j = 0; i < iMonoBlockSizeSam; i++, j += 2 )
{
vecsStereoSndCrd[j] = vecsStereoSndCrd[j + 1] =
vecsAudioSndCrdMono[i];
}
}
else
{
// if not connected, clear data
vecsStereoSndCrd.Reset ( 0 );
}
// update response time measurement and socket buffer size
CycleTimeVariance.Update();
UpdateSocketBufferSize();
}
void CClient::UpdateSocketBufferSize()
{
// just update the socket buffer size if auto setting is enabled, otherwise
// do nothing
if ( bDoAutoSockBufSize )
{
// We use the time response measurement for the automatic setting.
// Assumptions:
// - the audio interface/network jitter is assumed to be Gaussian
// - the buffer size is set to 3.3 times the standard deviation of
// the jitter (~98% of the jitter should be fit in the
// buffer)
// - introduce a hysteresis to avoid switching the buffer sizes all the
// time in case the time response measurement is close to a bound
// - only use time response measurement results if averaging buffer is
// completely filled
const double dHysteresis = 0.3;
// calculate current buffer setting
const double dAudioBufferDurationMs =
iMonoBlockSizeSam * 1000 / SYSTEM_SAMPLE_RATE;
// accumulate the standard deviations of input network stream and
// internal timer,
// add 0.5 to "round up" -> ceil,
// divide by MIN_SERVER_BLOCK_DURATION_MS because this is the size of
// one block in the jitter buffer
const double dEstCurBufSet = ( dAudioBufferDurationMs +
3.3 * ( Channel.GetTimingStdDev() + CycleTimeVariance.GetStdDev() ) ) /
SYSTEM_BLOCK_DURATION_MS_FLOAT + 0.5;
// upper/lower hysteresis decision
const int iUpperHystDec = LlconMath().round ( dEstCurBufSet - dHysteresis );
const int iLowerHystDec = LlconMath().round ( dEstCurBufSet + dHysteresis );
// if both decisions are equal than use the result
if ( iUpperHystDec == iLowerHystDec )
{
// set the socket buffer via the main window thread since somehow
// it gives a protocol deadlock if we call the SetSocketBufSize()
// function directly
PostWinMessage ( MS_SET_JIT_BUF_SIZE, iUpperHystDec );
}
else
{
// we are in the middle of the decision region, use
// previous setting for determing the new decision
if ( !( ( GetSockBufNumFrames() == iUpperHystDec ) ||
( GetSockBufNumFrames() == iLowerHystDec ) ) )
{
// The old result is not near the new decision,
// use per definition the upper decision.
// Set the socket buffer via the main window thread since somehow
// it gives a protocol deadlock if we call the SetSocketBufSize()
// function directly.
PostWinMessage ( MS_SET_JIT_BUF_SIZE, iUpperHystDec );
}
}
}
}