CRAN tidyr accidentally allows specification of pivot_wider() with empty names_from / values_from

[DavisVaughan]

library(tidyr)

df <- tibble::tribble(
  ~x, ~y, ~z,
  1,  2,  4,
  1,  3,  5
)

pivot_wider(df, values_from = c(x, y, z))
#> Warning: Values are not uniquely identified; output will contain list-cols.
#> * Use `values_fn = list` to suppress this warning.
#> * Use `values_fn = length` to identify where the duplicates arise
#> * Use `values_fn = {summary_fun}` to summarise duplicates

#> Warning: Values are not uniquely identified; output will contain list-cols.
#> * Use `values_fn = list` to suppress this warning.
#> * Use `values_fn = length` to identify where the duplicates arise
#> * Use `values_fn = {summary_fun}` to summarise duplicates

#> Warning: Values are not uniquely identified; output will contain list-cols.
#> * Use `values_fn = list` to suppress this warning.
#> * Use `values_fn = length` to identify where the duplicates arise
#> * Use `values_fn = {summary_fun}` to summarise duplicates
#> # A tibble: 1 × 3
#>   x_        y_        z_       
#>   <list>    <list>    <list>   
#> 1 <dbl [2]> <dbl [2]> <dbl [2]>

This should be a tidyselect error because name doesn't exist in the dataset, but it slips by because name and value are defined in the tidyr namespace to NULL to quiet R CMD Check and eval_select() finds this

tidyr/R/pivot-wide.R

Line 338 in 51ddd56

name <- value <- NULL

Found in #1102