Preserve names when unnest a named list column

Question

Here's my reprex:

make_named_list <- function(x) {
  list(apple = 1, banana = 2, carrot = 3)
}

tibble(x = 1:3, y = 2:4) %>% 
  mutate(z = map(x, make_named_list)) %>% 
  unnest() %>% 
  pull(z)

The output is a list without any names, but I would like to output a named list. I had a look at a related question here, but I'm not sure the solution applies to my (simpler?) scenario.

To be clear, the desired output is:

list(apple = 1, banana = 2, carrot = 3, 
     apple = 1, carrot = 2, banana = 3, 
     apple = 1, banana = 2, carrot = 3)

Hi @rcorty, Could you inform the desired output? I can not identify if you wish each slot of the list to have a name (apple, banana, carrot) or something else. — Adelmo Filho, Jun 20 '19 at 22:20
FYI the `unnest()` functions have had some updates in the development version of **tidyr** that make keeping list names easier. See some info in the [NEWS](https://github.com/tidyverse/tidyr/blob/master/NEWS.md#rectangling). — aosmith, Jun 20 '19 at 22:22
wow, there are a lot of changes coming down the pipe (haha)...thanks for pointing this out — rcorty, Jun 20 '19 at 22:25
I tried it with the development version of tidyr and still no names in the output — rcorty, Jun 20 '19 at 22:33
see if this helps https://stackoverflow.com/questions/50187445/how-to-name-a-dataframe-so-that-i-can-look-for-it-within-a-list — cephalopod, Jun 20 '19 at 22:35
Oh, I see your desired output now; I was picturing it incorrectly. I think the new functions like `unnest_longer()` would put the names in a separate column when unnesting. — aosmith, Jun 20 '19 at 22:43

score 1 · Answer 1 · answered Jun 21 '19 at 00:05

Depending on how your output requirement is you can try one of the following

library(tidyverse)

tibble(x = 1:3, y = 2:4) %>% 
  mutate(z = map(x, make_named_list)) %>%
  pull(z) 

#[[1]]
#[[1]]$apple
#[1] 1

#[[1]]$banana
#[1] 2

#[[1]]$carrot
#[1] 3


#[[2]]
#[[2]]$apple
#[1] 1
#.....

Or

tibble(x = 1:3, y = 2:4) %>% 
   mutate(z = map(x, make_named_list)) %>%
   pull(z) %>% unlist() 

# apple banana carrot  apple banana carrot  apple banana carrot 
#     1      2      3      1      2      3      1      2      3

Or

tibble(x = 1:3, y = 2:4) %>% 
   mutate(z = map(x, make_named_list)) %>%
   pull(z) %>% unlist() %>% as.list()

#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

#$apple
#[1] 1
#...

akrun · Answer 2 · 2019-06-21T01:33:21.727

Looks like we just need rep

rep(make_named_list(3), 3)
#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

Or if we are using tidyverse, then use flatten

tibble(x = 1:3, y = 2:4) %>% 
    transmute(z = map(x, make_named_list)) %>% 
    pull(z) %>% 
    flatten
#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

#$apple
#[1] 1

#$banana
#[1] 2

#$carrot
#[1] 3

score 0 · Answer 3 · answered Oct 15 '20 at 07:13

Set names_repair = tidyr_legacy in unnest().

library(tidyverse)
make_named_list <- function(x) {
  list(apple = 1, banana = 2, carrot = 3)
}

tibble(x = 1:3, y = 2:4) %>% 
  mutate(z = map(x, make_named_list)) %>% 
  unnest(cols = c(z), names_repair = tidyr_legacy) %>% 
  pull(z)

Preserve names when unnest a named list column

3 Answers3