Kakadu Hyper-Doc (kdu_stripe_compressor::push

bool push_stripe( kdu_int16 * stripe_bufs[], int stripe_heights[], int * sample_gaps=NULL, int * row_gaps=NULL, int * precisions=NULL, bool * is_signed=NULL, int flush_period=0)

Synopsis

Same as the first form of the overloaded push_stripe function, except in the following respects:

The stripe samples for each image component are provided with a 16-bit representation; as with other forms of the push_stripe function, the actual number of bits of this representation which are used is given by the precisions argument, but all 16 bits may be used (this is the default).
The default representation for each supplied sample value is signed, but the application may explicitly identify whether or not each component has a signed or unsigned representation. Note that there is no required connection between the Ssigned attribute managed by siz_params and the application's decision to supply signed or unsigned data to the present function. If the original data for component c was unsigned, the application may choose to supply signed sample values here, in which case it is responsible for first subtracting 2^{ precisions[c]-1} from each sample of image component c. If the application supplies unsigned data (setting is_signed[c] to true), the present function will subtract 2^{ precisions[c]-1} from the corresponding sample values.

Return Value

True until all samples for all image components have been pushed in, at which point the function returns false.

Arguments

stripe_bufs[] [kdu_int16 *]

stripe_heights[] [int]

See description of the first form of the push_stripe function.

sample_gaps [int *]

See description of the first form of the push_stripe function.

row_gaps [int *]

See description of the first form of the push_stripe function.

precisions [int *]

See description of the first form of the push_stripe function, but note these two changes: the precision for any component may be as large as 16 (this is the default, if precisions is NULL); and the samples all have a nominally signed representation (not the unsigned representation assumed by the first form of the function), unless otherwise indicated by a non-NULL is_signed argument.

is_signed [bool *]

If NULL, the supplied samples for each component, c, are assumed to have a signed representation in the range -2^{ precisions[c]-1} to 2^{ precisions[c]-1}-1. Otherwise, this argument points to an array with one element for each component. If is_signed[c] is true, the default signed representation is assumed for that component; if false, the component samples are assumed to have an unsigned representation in the range 0 to 2^{ precisions[c]}-1. What this means is that the function subtracts 2^{ precisions[c]-1} from the samples of any component for which is_signed[c] is false. It is allowable to have precisions[c]=16 even if is_signed[c] is false, meaning that the input words are treated as though they were 16-bit unsigned words.

flush_period [int]

See description of the first form of the push_stripe function.

kdu_stripe_compressor::push_stripe

Synopsis

Return Value

Arguments

stripe_bufs[] [kdu_int16 *]

stripe_heights[] [int]

sample_gaps [int *]

row_gaps [int *]

precisions [int *]

is_signed [bool *]

flush_period [int]