How to seperate a list column to multiple logical columns in R?-CodePudding

I have a data like that looks like the following:

dput(head(my.data[c("name", "categories")],1))

structure(list(name = "among us", categories = list(c("OnlinePvPLAN", 
"PvPOnlineCo-opLAN", "Co-opCross-PlatformMultiplayerRemote", 
"Playon", "PhoneRemotePlay", "onTablet]"))), row.names = 1L, class = "data.frame")

As you can imagine this structure is not easy to deal with. I want to separate the categories column to multiple logical columns like the following:

name | OnlinePvPLAN | PvPOnlineCo-opLAN | MMOOnlinePvPOnline
-----|--------------|-------------------|--------------------     .....
among|    TRUE      |        TRUE       |       FALSE
us

Since there are there are lot's of category columns I decided to write a function to write function that takes a list of categories to separate.

With the following code I can find columns that are in a particular category:

filter(my.data, map_lgl(my.data$categories, ~"OnlinePvPLAN" %in% .))

Using this, I wrote the following function:

compile.category.func <- function(data, category.list){
  lapply(X=category.list, function(category){
    category <- c(category)
    mutate(data, category=ifelse(map_lgl(data$categories, ~category %in% .), TRUE, FALSE))
  })
  data
}
output <- compile.category.func(my.data, c("OnlinePvPLAN","MMOOnlinePvPOnline"))

However this function doesn't work and doesn't generate any new columns.

CodePudding user response：

unnest into long form, use table to create the frequencies, use mutate to convert to TRUE/FALSE, then as.data.frame.matrix to convert that to a data frame and finally add the name column back. Omit the mutate line if 1/0 is wanted instead.

library(dplyr)
library(tidyr)
library(tibble)

DF %>%     
  unnest_longer(categories) %>% 
  table %>% 
  as.data.frame.matrix %>%
  mutate(across(, as.logical)) %>%
  rownames_to_column("name")

giving

      name Co-opCross-PlatformMultiplayerRemote OnlinePvPLAN onTablet]
1 among us                                 TRUE         TRUE      TRUE
  PhoneRemotePlay Playon PvPOnlineCo-opLAN
1            TRUE   TRUE              TRUE