After being convinced that it's best practice to use list to manage multiple data frames in R, I decided to put all my data frames that have the same column names into list.

sample list:

#create sample df
df_1 <- data.frame(item = c("a", "b", "c"), 
                   measure = c(1,2,3)
                   )
df_2 <- data.frame(item = c("x", "y", "z"), 
                   measure = c(4,5,6)
                   )

#use names as my df has names
data_list <- list(df_1 = df_1, df_2 = df_2)

I wanted to do the same operation across these data frames but I did not want them to be combined as one data frame as later on I'll need to save each into separate output.

Then it become a nightmare because I did not know how to manipulate column across all data frame inside a list.

While we can select a specific element and also the specific df in a list, how do we select by column name?

Taking an example, I need to change the value in the item column as Upper Case. In a data frame I'll do

df_1 <- df_1 %>% mutate(item = toupper(item))

I am still learning to write function and using the apply family in R. For this simple task I believe I can just use the existing function inside lapply like this

data_list = lapply(x, toupper)

The question is what is x in here? Is there a way to subset by column? Like data_list$df_1 or data_list[1] can give me the whole df_1.

I hope I can use lapply and function to do column by column approach across data frame in a list.

CodePudding user response：

Another option is to use map from purrr. So, if you have already written what you want to do for one dataframe in your list, then you can put it in as a function into map. You use .x rather than specifying the specific dataframe.

library(tidyverse)

map(data_list, ~ .x %>%
      mutate(item = toupper(item)))

Output

$df_1
  item measure
1    A       1
2    B       2
3    C       3

$df_2
  item measure
1    X       4
2    Y       5
3    Z       6