Skip to contents

Get R Class of an HDF5 Object

Source: R/info.r
h5_class.Rd

Inspects an HDF5 object and returns the R class that h5_read() would produce.

Usage

h5_class(file, name, attrs = FALSE)

Arguments

file

Path to the HDF5 file.

name

The full path of the object (group or dataset) to check.

attrs

Controls attribute checking. If TRUE or a character vector containing "class", the function will check for a "class" HDF5 attribute on the object.

Value

A character string representing the R class (e.g., "numeric", "complex", "character", "factor", "raw", "list", "NULL"). Returns NA_character_ for HDF5 types that h5lite cannot read.

Details

This function determines the resulting R class by inspecting the object's metadata.

  • Groups are reported as "list".

  • Datasets (integers, floats) are reported as "numeric" (since h5_read always returns double).

  • String datasets are reported as "character".

  • Complex datasets are reported as "complex".

  • Enum datasets are reported as "factor".

  • 1-byte Opaque datasets are reported as "raw".

  • Compound datasets are reported as "data.frame".

  • Null datasets (with a null dataspace) are reported as "NULL".

If attrs is set to TRUE or is a character vector containing "class", this function will first check for an HDF5 attribute on the object named "class". If a string attribute with this name exists, its value (e.g., "data.frame") will be returned, taking precedence over the object's type.

See also

h5_class_attr(), h5_typeof(), h5_read()

Examples

file <- tempfile(fileext = ".h5")

# Write various object types
h5_write(file, "integers", 1:5)
h5_write(file, "doubles", c(1.1, 2.2))
h5_write(file, "text", "hello")
h5_create_group(file, "my_group")

# Write a data.frame, which becomes a compound dataset
h5_write(file, "my_df", data.frame(a = 1:2, b = c("x", "y")))

# Check R classes
h5_class(file, "integers")      # "numeric"
#> [1] "numeric"
h5_class(file, "doubles")       # "numeric"
#> [1] "numeric"
h5_class(file, "text")          # "character"
#> [1] "character"
h5_class(file, "my_group")      # "list"
#> [1] "list"

# Check the data.frame
h5_class(file, "my_df") # "data.frame"
#> [1] "data.frame"
h5_class(file, "my_df", attrs = TRUE)  # "data.frame"
#> [1] "data.frame"

unlink(file)