README.md

In krlmlr/scalar: What the Package Does (One Line, Title Case)

scalar

The goal of scalar is to provide a vctrs compatible way to indicate that a vector is intended to have length one.

Installation

You can install the released version of scalar from CRAN with:

install.packages("scalar")

Example

This class is useful to distinguish between scalars and vectors when creating one-row tibbles that contain arbitrary objects, e.g. results of rowwise()-like computations.

library(scalar)

wrap_unless_scalar <- function(x) {
  if (is_scalar(x)) return(x)
  list(x)
}

tibble_row <- function(...) {
  lst <- tibble::lst(...)
  row <- purrr::map(lst, wrap_unless_scalar)
  tibble(!!!row)
}

tibble_row(a = 1, b = 2:3, c = list(4))
#> # A tibble: 1 x 3
#>   a         b         c         
#>   <list>    <list>    <list>    
#> 1 <dbl [1]> <int [2]> <list [1]>
tibble_row(a = scalar(1), b = 2:3, c = scalar(list(4)))
#> # A tibble: 1 x 3
#>   a        b         c        
#>   <scalar> <list>    <scalar> 
#> 1 1        <int [2]> <dbl [1]>

Non-scalars are rejected. Concatenation returns a “bare” object.

scalar(1:2)
#> Error in scalar(1:2): vec_size(x) == 1 is not TRUE
vec_c(scalar(1), scalar(2))
#> [1] 1 2

krlmlr/scalar documentation built on Nov. 4, 2019, 3:57 p.m.