CODEWITHSUNDEEP
rdplyr
john
john

Reputation: 1036

Convert columns to list

I have data in 2 columns (Region and GroupedCateg). Refer the dataframe below. I want to convert it into nested lists. I tried using group_by and do() function of dplyr and then convert into list but it didn't work.

df <- read.table(header = T,
                 text = '
Region  GroupedCateg
Beja    Alentejo
Evora   Alentejo
Portalegre  Alentejo
Faro    Algarve
Aveiro  Central
"Castelo Branco"    Central
Coimbra Central
Leiria  Central
Santarem    Central
Acores  Islands
Madeira Islands
Lisboa  Lisbon
Setubal Lisbon
Braga   North
Braganca    North
"Viana do Castelo"  North
"Vila Real" North
Porto   Porto
')

Desired Output in lists. Region would be in names. Respective GroupedCateg in nested list

list(
  list(
    name =  "Alentejo",
    categories = list("Beja", "Evora", "Portalegre")
  ),
  list(
    name =  "Algarve",
    categories = list("Faro")
  ),
  list(
    name =  "Central",
    categories = list("Aveiro", "Castelo Branco", "Coimbra", "Leiria", "Santarem" )
  ),
  
  list(
    name =  "North",
    categories = list("Braga", "Braganca", "Viana do Castelo", "Vila Real")
  ),
  list(
    name =  "Lisbon",
    categories = list("Lisboa", "Setubal")
  ),
  list(
    name =  "Islands",
    categories = list("Acores", "Madeira")
  ),
  
  list(
    name =  "Porto",
    categories = list("Porto")
  )
  
)

Upvotes: 1

Views: 167

Answers (3)

zx8754
zx8754

Reputation: 56149

Using base split on column, then lapply to re-format as desired:

x <- split(df$Region, df$GroupedCateg)

res <- lapply(names(x), function(i){
  list(name = i,
       categories = as.list(x[[ i ]]))
})

Upvotes: 2

Darren Tsai
Darren Tsai

Reputation: 35554

You can use pmap() in purrr.

library(dplyr)
library(purrr)

x <- df %>%
  group_by(GroupedCateg) %>% 
  summarise(Region = list(Region)) %>%
  pmap(~ list(name = .x, categories = as.list(.y)))

The corresponding base R version:

y <- apply(aggregate(Region ~ GroupedCateg, df, c),
     1, function(y) list(name = y[[1]], categories = as.list(y[[2]])))

all.equal(x, y)
# [1] TRUE

Output

[[1]]
[[1]]$name
[1] "Alentejo"

[[1]]$categories
[[1]]$categories[[1]]
[1] "Beja"

[[1]]$categories[[2]]
[1] "Evora"

[[1]]$categories[[3]]
[1] "Portalegre"



[[2]]
[[2]]$name
[1] "Algarve"

[[2]]$categories
[[2]]$categories[[1]]
[1] "Faro"



[[3]]
[[3]]$name
[1] "Central"

[[3]]$categories
[[3]]$categories[[1]]
[1] "Aveiro"

[[3]]$categories[[2]]
[1] "Castelo Branco"

[[3]]$categories[[3]]
[1] "Coimbra"

[[3]]$categories[[4]]
[1] "Leiria"

[[3]]$categories[[5]]
[1] "Santarem"



[[4]]
[[4]]$name
[1] "Islands"

[[4]]$categories
[[4]]$categories[[1]]
[1] "Acores"

[[4]]$categories[[2]]
[1] "Madeira"



[[5]]
[[5]]$name
[1] "Lisbon"

[[5]]$categories
[[5]]$categories[[1]]
[1] "Lisboa"

[[5]]$categories[[2]]
[1] "Setubal"



[[6]]
[[6]]$name
[1] "North"

[[6]]$categories
[[6]]$categories[[1]]
[1] "Braga"

[[6]]$categories[[2]]
[1] "Braganca"

[[6]]$categories[[3]]
[1] "Viana do Castelo"

[[6]]$categories[[4]]
[1] "Vila Real"



[[7]]
[[7]]$name
[1] "Porto"

[[7]]$categories
[[7]]$categories[[1]]
[1] "Porto"

Upvotes: 3

Daniel O
Daniel O

Reputation: 4358

In Base-R

apply(aggregate(Region~GroupedCateg,df,c),1, function(x) list(name=x[1], category=as.list(x[2]$Region))) 
    
[[1]]
[[1]]$name
[[1]]$name$GroupedCateg
[1] "Alentejo"


[[1]]$category
[[1]]$category[[1]]
[1] "Beja"

[[1]]$category[[2]]
[1] "Evora"

[[1]]$category[[3]]
[1] "Portalegre"



[[2]]
[[2]]$name
[[2]]$name$GroupedCateg
[1] "Algarve"


[[2]]$category
[[2]]$category[[1]]
[1] "Faro"



[[3]]
[[3]]$name
[[3]]$name$GroupedCateg
[1] "Central"


[[3]]$category
[[3]]$category[[1]]
[1] "Aveiro"

[[3]]$category[[2]]
[1] "Castelo Branco"

[[3]]$category[[3]]
[1] "Coimbra"

[[3]]$category[[4]]
[1] "Leiria"

[[3]]$category[[5]]
[1] "Santarem"



[[4]]
[[4]]$name
[[4]]$name$GroupedCateg
[1] "Islands"


[[4]]$category
[[4]]$category[[1]]
[1] "Acores"

[[4]]$category[[2]]
[1] "Madeira"



[[5]]
[[5]]$name
[[5]]$name$GroupedCateg
[1] "Lisbon"


[[5]]$category
[[5]]$category[[1]]
[1] "Lisboa"

[[5]]$category[[2]]
[1] "Setubal"



[[6]]
[[6]]$name
[[6]]$name$GroupedCateg
[1] "North"


[[6]]$category
[[6]]$category[[1]]
[1] "Braga"

[[6]]$category[[2]]
[1] "Braganca"

[[6]]$category[[3]]
[1] "Viana do Castelo"

[[6]]$category[[4]]
[1] "Vila Real"



[[7]]
[[7]]$name
[[7]]$name$GroupedCateg
[1] "Porto"


[[7]]$category
[[7]]$category[[1]]
[1] "Porto"

Upvotes: 1

Related Questions