bdWrite_hdf5_dimnames: Write dimnames to an HDF5 dataset

In BigDataStatMeth: Title: Scalable Statistical Computing with HDF5-Backed Matrices

Write dimnames to an HDF5 dataset

Description

Write row and/or column names metadata for an existing dataset in an HDF5 file. Empty vectors skip the corresponding dimnames.

Usage

bdWrite_hdf5_dimnames(filename, group, dataset, rownames, colnames)

Arguments

filename	Character string. Path to the HDF5 file.
group	Character string. Group containing the dataset.
dataset	Character string. Dataset name inside group.
rownames	Character vector of row names. Use character(0) to skip writing row names. If provided, length must equal nrow.
colnames	Character vector of column names. Use character(0) to skip writing column names. If provided, length must equal ncol.

Details

The dataset group/dataset must already exist. When non-empty, rownames and colnames lengths are validated against the dataset dimensions.

Value

List with components. If an error occurs, all string values are returned as empty strings (""):

fn

Character string with the HDF5 filename

dsrows

Character string with the full dataset path to the row names, stored as ".dataset_dimnames/1" within the specified group

dscols

Character string with the full dataset path to the column names, stored as ".dataset_dimnames/2" within the specified group

Examples

fn <- tempfile(fileext = ".h5")
hdf5_create_matrix(fn, "MGCCA_IN/X",
                   data = matrix(rnorm(5000), 100, 50))

bdWrite_hdf5_dimnames(
  filename = fn,
  group    = "MGCCA_IN",
  dataset  = "X",
  rownames = paste0("r", seq_len(100)),
  colnames = paste0("c", seq_len(50))
)
hdf5_close_all()
unlink(fn)

BigDataStatMeth documentation built on May 15, 2026, 1:07 a.m.