Inside Macintosh: Sound

Sound Component Data Records

The flow of data from one sound component to another is managed using a sound component data record. This record indicates to other sound components the format of the data that a particular component is generating, together with the location and length of the buffer containing that data. This allows other sound components to access data from that component as needed. A sound component data record is defined by the SoundComponentData data type.



 
typedef struct {

    long                flags;                      /*sound component flags*/

    OSType              format;                     /*data format*/

    short               numChannels;                /*number of channels in data*/

    short               sampleSize;                 /*size of a sample*/

    UnsignedFixed       sampleRate;                 /*sample rate*/

    long                sampleCount;                /*number of samples in buffer*/

    Byte                *buffer;                    /*location of data*/

    long                reserved;                   /*reserved*/

} SoundComponentData, *SoundComponentDataPtr;

flags: A set of bit flags whose meanings are specific to a particular sound component.
format: The format of the data a sound component is producing. The following formats are defined by Apple:



 
                    #define kOffsetBinary                   'raw '

                    #define kTwosComplement                 'twos'

                    #define kMACE3Compression               'MAC3'

                    #define kMACE6Compression               'MAC6'

See "Audio Data Types" for a description of these formats. You can define additional format types, which are currently assumed to be the types of proprietary compression algorithms.
numChannels: The number of channels of sound in the output data stream. If this field contains the value 1, the data is monophonic. If this field contains 2, the data is stereophonic. Stereo data is stored as interleaved samples, in a left-to-right ordering.
sampleSize: The size, in bits, of each sample in the output data stream. Typically this field contains the values 8 or 16. For compressed sound data, this field indicates the size of the samples after the data has been expanded.
sampleRate: The sample rate for the audio data. The sample rate is expressed as an unsigned, fixed-point number in the range 0 to 65536.0 samples per second.
sampleCount: The number of samples in the buffer pointed to by the buffer field. For compressed sounds, this field indicates the number of compressed samples in the sound, not the size of the buffer.
buffer: The location of the buffer that contains the sound data.
reserved: Reserved for future use. You should set this field to 0.