[Top]
[Prev]
[Next]
[Bottom]
DFSDreadslab/dsrslab
intn DFSDreadslab(char *
filename
, int32
start
[], int32
slab_size
[], int32
stride
[], VOIDP
buffer
, int32
buffer_size
[])
filename
IN:
Name of the HDF file
start
IN:
Buffer of size
rank
containing the coordinates for the start of the slab
slab_size
IN:
Buffer of size
rank
containing the size of each dimension in the slab
stride
IN:
Subsampling (not yet implemented)
buffer
OUT:
\Buffer for the returned slab
buffer_size
OUT:
Dimensions of the
buffer
buffer
Purpose
Reads a slab of data from any scientific dataset.
Return value
Returns SUCCEED (or 0) if successful and FAIL (or -1) otherwise.
Description
DFSDreadslab
will access to the scientific dataset following the current one if
DFSDgetdims
or
DFSDgetdata
are not called earlier. The
start
array indices are one-based. The rank of
start
must be the same as the number of dimensions of the specified variable. The elements of
slab_size
must be no larger than the dimensions of the scientific dataset in order. The stride feature is not currently implemented. For now just pass the
start
array as the argument for
stride
where it will be ignored.
To extract a slab of lower dimension than that of the dataset, enter 1 in the slab_size array for each omitted dimension. For example, to extract a two-dimensional slab from a three-dimensional dataset, specify the beginning coordinates in three dimensions and enter a 1 for the missing dimension in the slab_size array. More specifically, to extract a 3 x 4 slab containing the elements (6, 7, 8) through (8, 7, 11) specify the beginning coordinates as {6, 7, 8} and the slab size as {3, 1, 4}.
FORTRAN
integer function dsrslab(filename, start, slab_size, stride, buffer, buffersize)
character* (*) filename, buffer
integer start(*), slab_size(*),
integer stride(*), buffer_size(*)
[Top]
[Prev]
[Next]
[Bottom]
hdfhelp@ncsa.uiuc.edu
HDF User's Reference Manual, Draft 06/09/97, NCSA HDF Development Group.