view audioio/AudioGenerator.h @ 154:386b02c926bf

* Merge from one-fftdataserver-per-fftmodel branch.  This bit of
reworking (which is not described very accurately by the title of
the branch) turns the MatrixFile object into something that either
reads or writes, but not both, and separates the FFT file cache
reader and writer implementations separately.  This allows the
FFT data server to have a single thread owning writers and one reader
per "customer" thread, and for all locking to be vastly simplified
and concentrated in the data server alone (because none of the
classes it makes use of is used in more than one thread at a time).
The result is faster and more trustworthy code.

/* -*- c-basic-offset: 4 indent-tabs-mode: nil -*-  vi:set ts=8 sts=4 sw=4: */

/*
    Sonic Visualiser
    An audio file viewer and annotation editor.
    Centre for Digital Music, Queen Mary, University of London.
    This file copyright 2006 Chris Cannam.
    
    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.  See the file
    COPYING included with this distribution for more information.
*/

#ifndef _AUDIO_GENERATOR_H_
#define _AUDIO_GENERATOR_H_

class Model;
class NoteModel;
class DenseTimeValueModel;
class SparseOneDimensionalModel;
class RealTimePluginInstance;
class Playable;

#include <QObject>
#include <QMutex>

#include <set>
#include <map>

class AudioGenerator : public QObject
{
    Q_OBJECT

public:
    AudioGenerator();
    virtual ~AudioGenerator();

    /**
     * Add a data model to be played from and initialise any necessary
     * audio generation code.  Returns true if the model will be
     * played.  The model will be added regardless of the return
     * value.
     */
    virtual bool addModel(Model *model);

    /**
     * Remove a model.
     */
    virtual void removeModel(Model *model);

    /**
     * Remove all models.
     */
    virtual void clearModels();

    /**
     * Reset playback, clearing plugins and the like.
     */
    virtual void reset();

    /**
     * Set the target channel count.  The buffer parameter to mixModel
     * must always point to at least this number of arrays.
     */
    virtual void setTargetChannelCount(size_t channelCount);

    /**
     * Return the internal processing block size.  The frameCount
     * argument to all mixModel calls must be a multiple of this
     * value.
     */
    virtual size_t getBlockSize() const;

    /**
     * Mix a single model into an output buffer.
     */
    virtual size_t mixModel(Model *model, size_t startFrame, size_t frameCount,
			    float **buffer, size_t fadeIn = 0, size_t fadeOut = 0);

    /**
     * Specify that only the given set of models should be played.
     */
    virtual void setSoloModelSet(std::set<Model *>s);

    /**
     * Specify that all models should be played as normal (if not
     * muted).
     */
    virtual void clearSoloModelSet();

protected slots:
    void playPluginIdChanged(const Playable *, QString);
    void playPluginConfigurationChanged(const Playable *, QString);

protected:
    size_t       m_sourceSampleRate;
    size_t       m_targetChannelCount;

    bool m_soloing;
    std::set<Model *> m_soloModelSet;

    struct NoteOff {

	int pitch;
	size_t frame;

	struct Comparator {
	    bool operator()(const NoteOff &n1, const NoteOff &n2) const {
		return n1.frame < n2.frame;
	    }
	};
    };

    typedef std::map<const Model *, RealTimePluginInstance *> PluginMap;

    typedef std::set<NoteOff, NoteOff::Comparator> NoteOffSet;
    typedef std::map<const Model *, NoteOffSet> NoteOffMap;

    QMutex m_mutex;
    PluginMap m_synthMap;
    NoteOffMap m_noteOffs;
    static QString m_sampleDir;

    virtual RealTimePluginInstance *loadPluginFor(const Model *model);
    virtual RealTimePluginInstance *loadPlugin(QString id, QString program);
    static void initialiseSampleDir();
    static void setSampleDir(RealTimePluginInstance *plugin);

    virtual size_t mixDenseTimeValueModel
    (DenseTimeValueModel *model, size_t startFrame, size_t frameCount,
     float **buffer, float gain, float pan, size_t fadeIn, size_t fadeOut);

    virtual size_t mixSparseOneDimensionalModel
    (SparseOneDimensionalModel *model, size_t startFrame, size_t frameCount,
     float **buffer, float gain, float pan, size_t fadeIn, size_t fadeOut);

    virtual size_t mixNoteModel
    (NoteModel *model, size_t startFrame, size_t frameCount,
     float **buffer, float gain, float pan, size_t fadeIn, size_t fadeOut);

    static const size_t m_pluginBlockSize;
};

#endif