Get the dimensions of an HDF5 dataset

Two convenience functions to obtain the dimensions of an HDF5 dataset as well as the dimensions of its chunks.

Usage

h5dim(filepath, name, as.integer=TRUE)

h5chunkdim(filepath, name, adjust=FALSE)

Arguments

filepath

The path (as a single string) to an HDF5 file.

name

The name (as a single string) of a dataset in the HDF5 file.

as.integer

By default h5dim() returns the dimensions of the dataset in an integer vector and will raise an error if any dimension is greater than .Machine$integer.max (= 2^31 - 1).

Use as.integer=FALSE to support datasets with dimensions greater than .Machine$integer.max. In this case the dimensions are returned in a numeric vector.

adjust

By default h5chunkdim() returns the dimensions of the chunks as reported by H5Pget_chunk from the C HDF5 API. Note that the HDF5 specs allow some or all the dimensions of the chunks to be greater than the dimensions of the dataset.

You can use adjust=TRUE to request h5chunkdim() to return adjusted chunk dimensions, that is, dimensions that do not exceed the dimensions of the dataset. The adjusted chunk dimensions are simply obtained by replacing those dimensions in the vector of chunk dimensions that are greater than the corresponding dataset dimension with the latter.

Value

An integer (or numeric) vector of length the number of dimensions of the HDF5 dataset.

Examples

test_h5 <- system.file("extdata", "test.h5", package="h5mread")
h5ls(test_h5)
#> Warning: An open HDF5 file handle exists. If the file has changed on disk meanwhile, the function may not work properly. Run 'h5closeAll()' to close all open HDF5 object handles.
#>           group         name       otype  dclass          dim
#> 0             / .m2_dimnames   H5I_GROUP                     
#> 1 /.m2_dimnames            1 H5I_DATASET  STRING         4000
#> 2 /.m2_dimnames            2 H5I_DATASET  STRING           90
#> 3             /           a3 H5I_DATASET INTEGER 180 x 75 x 4
#> 4             /           m1 H5I_DATASET INTEGER       12 x 5
#> 5             /           m2 H5I_DATASET   FLOAT    4000 x 90
#> 6             /           m4 H5I_DATASET INTEGER    28 x 4000
#> 7             /       rwords H5I_DATASET  STRING        30000

h5dim(test_h5, "m2")
#> [1] 4000   90
h5chunkdim(test_h5, "m2")
#> [1] 50 40

h5dim(test_h5, "a3")
#> [1] 180  75   4
h5chunkdim(test_h5, "a3")
#> [1] 20 20  4