new_dvs_bytes()

new_dvs_bytes() constructs a dvs_bytes vector. The size and stored_size columns returned by dvs_add(), dvs_status(), and dvs_get() are dvs_bytes values, so they print as human-readable sizes in a tibble. Values are stored as double to represent sizes past 2 GB without integer overflow.

new_dvs_bytes(x)

🔗Parameters

🔗Construct a value

options(width = 1000)
library(dvs)
b <- new_dvs_bytes(c(512, 1048576, 1073741824))
class(b)

[1] "dvs_bytes" "numeric"  

🔗Printing in a tibble

A bare dvs_bytes vector prints as plain numbers, but as a tibble column it renders via pillar_shaft() and type_sum() as right-aligned, human-readable sizes (the <bytes> column type).

tibble::tibble(file = c("small", "medium", "large"), size = b)

# A tibble: 3 × 2
  file      size
  <chr>  <bytes>
1 small    512 B
2 medium  1.0 MB
3 large   1.0 GB

🔗Arithmetic and summaries

The Ops group generic keeps the dvs_bytes class for + and - (the result is still bytes) and drops it for other operators. The Summary group generic keeps it for sum, min, max, and range. A summed column stays a dvs_bytes value in a tibble.

suppressMessages(library(dplyr))
tibble::tibble(size = b) |> summarise(total = sum(size))

# A tibble: 1 × 1
    total
  <bytes>
1  1.0 GB

🔗See also

• format_byte_size(): the underlying formatter.
• dvs_add(), dvs_status(), dvs_get()