pmap_dfr and shifting a vector in data.table

Question

I am trying to functionalize my R code, but have run into problems because a lot of my work involves the use of shifted vectors in data.table and pmap_dfr simply does not implement data.table's shift inside the function as expected. I am using R 4.2.2 on a MacOS Venture 13.1 anad the same issue appears in Windows 11. I have presented a minimal working example below.

library(data.table)
library(tidytable)

minimal_failing_function <- function(A){
  DT <- data.table(A) 
  DT[ , A:= shift(A, fill = NA, type = "lag", n = 1)] %>% `[`
  return(DT)
}

# works
minimal_failing_function(c(1,2))

# fails
tidytable::pmap_dfr(.l = list(c(1,2)), 
                    .f = minimal_failing_function)

The line tidytable::pmap_dfr should return:

    A
1: NA
2:  1

Instead, I get:

# A tidytable: 2 × 1
      A
  <dbl>
1    NA
2    NA

I am looking for really any solution that will enable me access to pmap functionality.

Answer 1

On the suggestion of stefan, I realized that pmap_dfr wants a list of lists for its input. It automatically unlists the interior list as a vector.

tidytable::pmap_dfr(.l = list(A = list(c(1,2))), .f = minimal_failing_function)

gives the desired result.