Create a list of all dataframes/tibbles in the global environment?

Question

How does one create a named list of all dataframes/tibbles in the global environment in R? Is there a way to do this without manually hardcoding all dataframes/tibbles?

I.e. if the global environment contains the dataframes/tibbles df_1, my_data_1, science_1, all_data, how does one create an output that looks like:

files_list <- list(
df_1 = df_1,
my_data_1 = my_data_1,
science_1 = science_1,
all_data = all_data
)

[Use ls() or objects() to get objects of class data.frame](https://stackoverflow.com/questions/5796508/use-ls-or-objects-to-get-objects-of-class-data-frame) — Henrik, Feb 12 '23 at 19:53

GuedesBF · Answer 1 · 2023-02-13T01:34:30.567

3

We can get all objects first, then keep only the data.frames

library(purrr)

mget(ls()) %>% keep(is.data.frame)

edited Feb 13 '23 at 01:34

answered Feb 12 '23 at 21:16

GuedesBF

8,409
5
19
37

score 2 · Answer 2 · answered Feb 12 '23 at 19:08

2

We may Filter the elements that are data.frame or tibble in the environment that we are working on - e.g. in the global env, it can be

Filter(length, eapply(.GlobalEnv, 
     function(x) if(is.data.frame(x)||is_tibble(x)) x))

answered Feb 12 '23 at 19:08

akrun

874,273
37
540
662

Is `is_tibble` necessary here? `is.data.frame(tibble(iris))` returns TRUE anyway – GuedesBF Feb 13 '23 at 01:33
@GuedesBF Not really, it is just to make it explicit. If I have just used `is.data.frame` (which would be the base class), then there would be comments that it is not clear how it includes the tibble case. – akrun Feb 13 '23 at 06:47

score 1 · Accepted Answer · answered Feb 13 '23 at 08:43

1

A base way, combining methods of @GuedesBF and @akrun could be using ls, mget and Filter.

Filter(is.data.frame, mget(ls()))
#Filter(is.data.frame, mget(ls(.GlobalEnv))) #More explicit using globEnv

answered Feb 13 '23 at 08:43

GKi

37,245
2
26
48

score 0 · Answer 4 · answered Feb 12 '23 at 21:18

Please try the below code which will generate a df

naml <- list()

  for (i in seq_along(ls(envir =.GlobalEnv))) {
    j <- ls(envir =.GlobalEnv)[i]
  if (any(class(get(j))=='data.frame')) name <- {j} else name <- NA
  if (any(class(get(j))=='data.frame')) class <- class(get(j))[3] else class <- NA
  if (!is.na(name) & !is.na(class)) {
  df <- data.frame(namex=name,classx=class)
  naml[[j]] <- df
  }
  }

df2 <- do.call(rbind, naml) %>% rownames_to_column('name') %>% 
pivot_wider(names_from = name, values_from = namex)

Create a list of all dataframes/tibbles in the global environment?

4 Answers4