qffmpegaudiorenderer_p.h

Go to the documentation of this file.

// Copyright (C) 2021 The Qt Company Ltd.
// SPDX-License-Identifier: LicenseRef-Qt-Commercial OR LGPL-3.0-only OR GPL-2.0-only OR GPL-3.0-only
#ifndef QFFMPEGAUDIORENDERER_P_H
#define QFFMPEGAUDIORENDERER_P_H
 
//
//  W A R N I N G
//  -------------
//
// This file is not part of the Qt API. It exists purely as an
// implementation detail. This header file may change from version to
// version without notice, or even be removed.
//
// We mean it.
//
 
#include <QtFFmpegMediaPluginImpl/private/qffmpegrenderer_p.h>
 
#include <QtMultimedia/qaudiobuffer.h>
 
QT_BEGIN_NAMESPACE
 
class QAudioOutput;
class QAudioBufferOutput;
class QAudioSink;
class QFFmpegResampler;
 
namespace QFFmpeg {
 
struct AbstractAudioFrameConverter;
 
class AudioRenderer : public Renderer
{
    Q_OBJECT
public:
    AudioRenderer(const PlaybackEngineObjectID &id, const TimeController &tc, QAudioOutput *output,
                  QAudioBufferOutput *bufferOutput, bool pitchCompensation);
 
    void setOutput(QAudioOutput *output);
 
    void setOutput(QAudioBufferOutput *bufferOutput);
 
    void setPitchCompensation(bool enabled);
 
    ~AudioRenderer() override;
 
protected:
    using Microseconds = std::chrono::microseconds;
    struct SynchronizationStamp
    {
        QAudio::State audioSinkState = QAudio::IdleState;
        qsizetype audioSinkBytesFree = 0;
        qsizetype bufferBytesWritten = 0;
        TimePoint timePoint = TimePoint::max();
    };
 
    struct BufferLoadingInfo
    {
        enum Type { Low, Moderate, High };
        Type type = Moderate;
        TimePoint timePoint = TimePoint::max();
        Microseconds delay = Microseconds(0);
    };
 
    struct AudioTimings
    {
        Microseconds actualBufferDuration = Microseconds(0);
        Microseconds maxSoundDelay = Microseconds(0);
        Microseconds minSoundDelay = Microseconds(0);
    };
 
    struct BufferedDataWithOffset
    {
        QAudioBuffer buffer;
        qsizetype offset = 0;
 
        bool isValid() const { return buffer.isValid(); }
        qsizetype size() const { return buffer.byteCount() - offset; }
        const char *data() const { return buffer.constData<char>() + offset; }
    };
 
    void seekInternal() override;
 
    RenderingResult renderInternal(Frame frame) override;
 
    RenderingResult pushFrameToOutput(const Frame &frame);
 
    void pushFrameToBufferOutput(const Frame &frame);
 
    void onPlaybackRateChanged() override;
 
    TimePoint nextTimePoint() const override;
 
    void onPauseChanged() override;
 
    void freeOutput();
 
    void updateOutputs(const Frame &frame);
 
    void initAudioFrameConverter(const Frame &frame);
 
    void onDeviceChanged();
 
    void updateVolume();
 
    void updateSynchronization(const SynchronizationStamp &stamp, const Frame &frame);
 
    Microseconds bufferLoadingTime(const SynchronizationStamp &syncStamp) const;
 
    void onAudioSinkStateChanged(QAudio::State state);
 
    Microseconds durationForBytes(qsizetype bytes) const;
 
private:
    QPointer<QAudioOutput> m_output;
    QPointer<QAudioBufferOutput> m_bufferOutput;
    std::unique_ptr<QAudioSink> m_sink;
    AudioTimings m_timings;
    BufferLoadingInfo m_bufferLoadingInfo;
    std::unique_ptr<QFFmpegResampler> m_bufferOutputResampler;
    QAudioFormat m_sinkFormat;
 
    QPointer<QIODevice> m_ioDevice;
 
    bool m_deviceChanged = false;
    bool m_bufferOutputChanged = false;
 
    bool m_firstFrameToSink = true;
 
    // pitch compensation
    bool m_pitchCompensation = false;
    std::unique_ptr<AbstractAudioFrameConverter> m_audioFrameConverter;
 
    struct SessionContext
    {
        // TODO: add other members after clarifying the best session cleanup strategy
        BufferedDataWithOffset bufferedData;
        bool lastFramePushDone = true;
        bool drained = false;
    };
 
    SessionContext m_sessionCtx;
};
 
} // namespace QFFmpeg
 
QT_END_NAMESPACE
 
#endif // QFFMPEGAUDIORENDERER_P_H