no applicable method of tibble when passed to a function-CodePudding

I'm trying to pass a tibble to a function call and then perform some dplyr transformations on it. However, when I pass the tibble to the function it tells me that I cannot perform mutate on an object of class character.

For example, say I have this data and this code:

dataframe_1 <- tibble(var1 = c('a','b','c','d'), var2 = c('e','f','g','h'))
dataframe_2 <- tibble(var1 = c(1,2,3,4), var2 = c(5,6,7,8))
dataframe_3 <- tibble(var1 = c('i','j','k','l'), var2 = c('m','n','o','p'))
dataframe_4 <- tibble(var1 = c(9,10,11,12), var2 = c(13,14,15,16)) 


function_name <- function(.x) {
  .x %>%
  mutate(var3 = .x)
}
  
print_table <- map_dfr(
  .x = dataframe_1,
  .f = function_name
)

I get the error Error in UseMethod("mutate") : no applicable method for 'mutate' applied to an object of class "character".

But if I run this:

function_name <- function(.x) {
  return(.x)
}

print_table <- map_dfr(
  .x = dataframe_1,
  .f = function_name
)

I get

print_table
# A tibble: 4 × 2
  var1  var2 
  <chr> <chr>
1 a     e    
2 b     f    
3 c     g    
4 d     h

So R is able to tell that dataframe_1 is in fact a tibble, and I know that it's possible to pipe tibbles into a mutate() command.

Can anyone see where i'm going wrong here?

CodePudding user response：

If we are creating the object name as a column, get all the objects (mget) into a named list based on the pattern in the object names created in ls, then loop over the list with imap, and create the column from the names using .y

library(dplyr)
library(purrr)
mget(ls(pattern = 'dataframe_\\d ')) %>% 
    imap(~ .x %>%
                 mutate(var3 = .y))