如何使用 tibble::tribble() 创建列表列？

Question

When building a tibble from scratch using tribble() , how can I set up a column as a "list-column" so it could include data of different classes?

For example, consider the following code where I have one logical value and one numeric value.

library(tibble)

some_numeric <- 5
some_logical <- TRUE

my_df <-
  tribble(~name, ~value,
        "logical", some_logical,
        "numeric", some_numeric)

Whereas I was expecting my_df to look like:

## # A tibble: 2 x 2
##   name    value    
##  <chr>   <list>   
## 1 logical <lgl [1]>
## 2 numeric <dbl [1]>

I actually got this:

## # A tibble: 2 x 2
##   name    value
##   <chr>   <dbl>
## 1 logical     1
## 2 numeric     5

---

Sure, there's this way using tibble() that works:

tibble(name = c("logical", "numeric"),
       value = list(some_logical, some_numeric))

## # A tibble: 2 x 2
##   name    value    
##   <chr>   <list>   
## 1 logical <lgl [1]>
## 2 numeric <dbl [1]>

But I wanted to create it row-wise using tribble() , and the following doesn't work either:

tribble(~name, ~value,
          "logical", list(some_logical),
          "numeric", list(some_numeric))

## # A tibble: 2 x 2
##   name    value     
##   <chr>   <list>    
## 1 logical <list [1]> ## <--- I don't want nested list but nested *logical* vec
## 2 numeric <list [1]> ## <--- I don't want nested list but nested *double* vec

---

So my question is, is there a simple way to use tribble() to set up a list-column in which each value is a nested vector and not a nested list?

Answer 1

library(tidyverse)


f <- tribble(~name, ~value,
        "logical", as.vector("some_logical", mode = "list"),
        "numeric", as.vector("some_logical", mode = "list"))

f
#> # A tibble: 2 x 2
#>   name    value     
#>   <chr>   <list>    
#> 1 logical <list [1]>
#> 2 numeric <list [1]>

f$value
#> [[1]]
#> [[1]][[1]]
#> [1] "some_logical"
#> 
#> 
#> [[2]]
#> [[2]][[1]]
#> [1] "some_logical"

^{Created on 2021-02-22 by the reprex package (v1.0.0)}