R: Setting purrr:pmap's .l attribute in advance using the $ operator

Question

I am quite a newbie in R and therefore cannot explain the following behavior.

So, let's assume I have the this data structure.

> use_data
     ID value1             value2             value3 Time               
  <int> <list>             <list>             <list> <dttm>             
1    52 <tibble [96 x 25]> <tibble [59 x 26]> <NULL> 2012-04-25 03:00:00
2    24 <NULL>             <tibble [30 x 26]> <NULL> 2012-07-18 13:45:00

For simplicity, the following example is not very expressive, but it serves as a demonstration of the problem. That is, I want to use the pmap function to iterate over the columns in parallel. At the moment I only return the ID to create the new column target:

fun_example <- function(df) {
  result_df <- df %>% mutate(target = purrr::pmap(
    .l = list(ID, value1, value2),
    .f = function(x, y, z){
      return (x)
    }
  )) %>% unnest(target)
}
(fun_example(use_data))

As intended, this results in

     ID value1             value2             value3 Time                target
  <int> <list>             <list>             <list> <dttm>               <int>
1    52 <tibble [96 x 25]> <tibble [59 x 26]> <NULL> 2012-04-25 03:00:00     52
2    24 <NULL>             <tibble [30 x 26]> <NULL> 2012-07-18 13:45:00     24

Now I want to set the list with data to iterate over in advance by defining cols <-list(df$ID, df$value1, df$value2) and then

fun_example <- function(df) { 
  result_df <- df %>% mutate(target = purrr::pmap(
    .l = cols,
    .f = function(x, y, z){
      return (x)
    }
  )) %>% unnest(target)
}
(fun_example(use_data))

However, this gives me the following error:

Problem while computing `target = purrr::pmap(...)`.
x `target` must be size 1, not 2.
i The error occurred in group 1: ID = 24.

I guess the problem is that pmap somehow no longer gives the desired result. Finally, two questions:

1. Can someone explain what is happening in the example described?
2. Besides defining the data to iterate over directly, is there a way to pass strings of the column like cols <- list("ID", "value1", "value2") ?

Answer 1

library(purrr)
cols <- c("mpg", "cyl", "disp")
mtcars %>% 
    mutate(target = pmap(
        .l = list(!!!rlang::parse_exprs(cols)), 
        .f = function(x, y, z) x + y + z
    ))
                     mpg cyl  disp  hp drat    wt  qsec vs am gear carb target
Mazda RX4           21.0   6 160.0 110 3.90 2.620 16.46  0  1    4    4    187
Mazda RX4 Wag       21.0   6 160.0 110 3.90 2.875 17.02  0  1    4    4    187
Datsun 710          22.8   4 108.0  93 3.85 2.320 18.61  1  1    4    1  134.8
Hornet 4 Drive      21.4   6 258.0 110 3.08 3.215 19.44  1  0    3    1  285.4
Hornet Sportabout   18.7   8 360.0 175 3.15 3.440 17.02  0  0    3    2  386.7
Valiant             18.1   6 225.0 105 2.76 3.460 20.22  1  0    3    1  249.1
Duster 360          14.3   8 360.0 245 3.21 3.570 15.84  0  0    3    4  382.3
Merc 240D           24.4   4 146.7  62 3.69 3.190 20.00  1  0    4    2  175.1
Merc 230            22.8   4 140.8  95 3.92 3.150 22.90  1  0    4    2  167.6

pmap takes lists, while within mutate we pass bare names (mpg not "mpg").

So given a string vector, we parse it using parse_exprs and then create use list and !!! operator to unquote and evaluate these expressions.