_e_n_c_o_d_e_r DDDDMMMMMMMMPPPPEEEEGGGG1111aaaauuuuddddiiiiooooeeeennnnccccooooddddeeeerrrr structure, created by
ddddmmmmMMMMPPPPEEEEGGGG1111AAAAuuuuddddiiiiooooEEEEnnnnccccooooddddeeeerrrrCCCCrrrreeeeaaaatttteeee((((3333ddddmmmm)))), specifies and stores the signal
processing parameters. It describes the kind of compression to
perform, and holds the between-block state of the compressor.
It is initially generated and processed by calls to
ddddmmmmMMMMPPPPEEEEGGGG1111AAAAuuuuddddiiiiooooEEEEnnnnccccooooddddeeeerrrrCCCCrrrreeeeaaaatttteeee((((3333ddddmmmm)))) and
ddddmmmmMMMMPPPPEEEEGGGG1111AAAAuuuuddddiiiiooooEEEEnnnnccccooooddddeeeerrrrSSSSeeeettttPPPPaaaarrrraaaammmmssss((((3333ddddmmmm)))), and is modified on each
call.
_s_a_m_p_B_u_f pointer to input sample data buffer. The data format is short
(16-bit). The samples are assumed to be two's complement. The
block of audio data is understood to be
DDDDMMMM____AAAAUUUUDDDDIIIIOOOO____CCCCOOOODDDDEEEECCCC____FFFFRRRRAAAAMMMMEEEESSSS____PPPPEEEERRRR____BBBBLLLLOOOOCCCCKKKK sample frames long, where this
is a symbol, which value is returned by
ddddmmmmMMMMPPPPEEEEGGGG1111AAAAuuuuddddiiiiooooEEEEnnnnccccooooddddeeeerrrrGGGGeeeettttPPPPaaaarrrraaaammmmssss((((3333ddddmmmm)))). For each call to
_o_u_t_p_u_t pointer to output data buffer, an array of bytes. The data
format is unsigned char (8-bit). It points to a pre-allocated
byte block into which ddddmmmmMMMMPPPPEEEEGGGG1111AAAAuuuuddddiiiiooooEEEEnnnnccccooooddddeeee((((3333ddddmmmm)))) will write the
compressed data for this block. The block must be at least
DDDDMMMM____AAAAUUUUDDDDIIIIOOOO____CCCCOOOODDDDEEEECCCC____MMMMAAAAXXXX____BBBBYYYYTTTTEEEESSSS____PPPPEEEERRRR____BBBBLLLLOOOOCCCCKKKK bytes long, where this is
pointer to an integer return value; indicates the actual number
of compressed bytes written into the data structure (which will
be less than or equal to DDDDMMMM____AAAAUUUUDDDDIIIIOOOO____CCCCOOOODDDDEEEECCCC____MMMMAAAAXXXX____BBBBYYYYTTTTEEEESSSS____PPPPEEEERRRR____BBBBLLLLOOOOCCCCKKKK).
It is the value returned by the function. In the case of some
error occuring such as an unrecognized encoder structure, this
compression/decompression delay will be caused by filter operations that
convolve a N past input samples with a filter M-coefficient array to
create N+M-1 output samples. The filtered signal contains more samples
than the unfiltered signal. the output signal is offset by
DDDDMMMM____AAAAUUUUDDDDIIIIOOOO____CCCCOOOODDDDEEEECCCC____FFFFIIIILLLLTTTTEEEERRRR____DDDDEEEELLLLAAAAYYYY samples from the first sample thus omits the
last DDDDMMMM____AAAAUUUUDDDDIIIIOOOO____CCCCOOOODDDDEEEECCCC____FFFFIIIILLLLTTTTEEEERRRR____DDDDEEEELLLLAAAAYYYY samples from the output file.