Convert raw bytes to an R array based on the specified data type information — convert_bytes_to_array • grumpy

This is a replacement for readBin() that can handle the various data types and endianness specified in the .npy file header.

Usage

convert_bytes_to_array(bytes, what, shape, size, endian)

Arguments

bytes: A raw vector containing the bytes to convert
what: A character specifying the base type to convert to (e.g., "float", "int", "string", etc.)
shape: A numeric vector with desired shape of the output array
size: A numeric value with the number of bytes per element for the specified type
endian: The endianness of the data ("little", "big", or NA for single-byte types)

Value

An R array containing the converted data, with the specified shape and data type.

Examples

x <- matrix(c(3L, 6L, 2L, 1L, 12L, 0L), nrow = 2, ncol = 3)
x
#>      [,1] [,2] [,3]
#> [1,]    3    2   12
#> [2,]    6    1    0

y <- writeBin(c(x), raw()) |>
  convert_bytes_to_array("int", shape = c(2L, 3L), size = 4L, endian = "little")
y
#>      [,1] [,2] [,3]
#> [1,]    3    2   12
#> [2,]    6    1    0
dim(y)
#> [1] 2 3
is.array(y)
#> [1] TRUE
storage.mode(y)
#> [1] "integer"