_f_i_l_e is an AAAAFFFFffffiiiilllleeeehhhhaaaannnnddddlllleeee structure, previously created by a call
to aaaaffffOOOOppppeeeennnnFFFFiiiilllleeee(3dm).
_t_r_a_c_k is an integer which identifies an audio track in _f_i_l_e.
_e_x_t_e_n_d_3_t_o_4 is a integer boolean value (i.e., 0 or 1) indicating whether
the routine should extend the value of the frame size for
24-bit uncompressed data out to 32 bits.
Since all currently supported file formats contain exactly
one audio track per file, the constant AAAAFFFF____DDDDEEEEFFFFAAAAUUUULLLLTTTT____TTTTRRRRAAAACCCCKKKK
should always be used here for now.
RRRREEEETTTTUUUURRRRNNNN VVVVAAAALLLLUUUUEEEE
aaaaffffGGGGeeeettttFFFFrrrraaaammmmeeeeSSSSiiiizzzzeeee(((()))) returns the size in bytes of a sample frame in the file
referenced by _t_r_a_c_k. aaaaffffGGGGeeeettttVVVViiiirrrrttttuuuuaaaallllFFFFrrrraaaammmmeeeeSSSSiiiizzzzeeee(((()))) returns size in bytes of a
sample frame in the audio buffer as specified by the current virtual
format. See aaaaffffIIIInnnnttttrrrroooo(3dm) for an explanation of virtual format.
In case of error, -1 is returned.
DDDDEEEESSSSCCCCRRRRIIIIPPPPTTTTIIIIOOOONNNN
aaaaffffGGGGeeeettttFFFFrrrraaaammmmeeeeSSSSiiiizzzzeeee(((()))) returns the size of a sample frame in an audio track in
the file represented by the AAAAFFFFffffiiiilllleeeehhhhaaaannnnddddlllleeee argument.
aaaaffffGGGGeeeettttVVVViiiirrrrttttuuuuaaaallllFFFFrrrraaaammmmeeeeSSSSiiiizzzzeeee(((()))) does the same for the virtual format for the
given _t_r_a_c_k, as set by any combination of calls to
aaaaffffSSSSeeeettttVVVViiiirrrrttttuuuuaaaallllSSSSaaaammmmpppplllleeeeFFFFoooorrrrmmmmaaaatttt(3dm) and aaaaffffSSSSeeeettttVVVViiiirrrrttttuuuuaaaallllCCCChhhhaaaannnnnnnneeeellllssss(3dm).
For monaural data, a sample frame consists of a single audio sample. For
stereo data, a sample frame consists of a <left, right> stereo sample
pair.
The _e_x_t_e_n_d_3_t_o_4 allows the application to choose, in the case where 24-bit
uncompressed data is present in the file associated with _t_r_a_c_k, whether
to return the actual size in bytes of a frame of data in that track (3 *
number_of_channels) or the size of the frames which will be presented to
the user (since this will _a_l_w_a_y_s be 4-byte sign-extended integer). The
former is useful for informational purposes, but the latter is necessary
for determining the size of the audio buffer to be handed to