CODEWITHSUNDEEP
rdata.table
abhiieor
abhiieor

Reputation: 3554

List of 2 columns as column data.table

I have a data.table like:

foo <- data.table(a = c(1,2,3), b = c(4,5,6))

I want to add another column c which is by row/record a list of a and b. Something like:

data.table(a = c(1,2,3), b = c(4,5,6), c = c(list(1,4), list(2,5), list(3,6)))

but even in the example it only take one element and recycle

   a b c
1: 1 4 1
2: 2 5 4
3: 3 6 2
4: 1 4 5
5: 2 5 3
6: 3 6 6

However I want:

   a b c
1: 1 4 list(1,4)
2: 2 5 list(2,5)
3: 3 6 list(3,6)

I tried also:

foo$c <- foo[, list('a' = a, 'b' = b)]
foo$c <- foo[, list('a' = a, 'b' = b), by = 1:nrow(foo)]
foo$c <- as.list(foo[, list('a' = a, 'b' = b)])

Nothing works.

Upvotes: 1

Views: 513

Answers (2)

akuiper
akuiper

Reputation: 215057

One way is to use Map and list:

foo[, c := Map(list, a, b)][]

#   a b      c
#1: 1 4 <list>
#2: 2 5 <list>
#3: 3 6 <list>

And column c is a list of lists:

dput(foo$c)
# list(list(1, 4), list(2, 5), list(3, 6))

To keep the names:

foo[, c := Map(list, a=a, b=b)][]
#   a b      c
#1: 1 4 <list>
#2: 2 5 <list>
#3: 3 6 <list>

dput(foo$c)
# list(list(a = 1, b = 4), list(a = 2, b = 5), list(a = 3, b = 6))

Upvotes: 2

s_baldur
s_baldur

Reputation: 33548

Using list instead of c seems to work:

data.table(a = c(1,2,3), b = c(4,5,6), c = list(list(1,4), list(2,5), list(3,6))) 
   a b      c
1: 1 4 <list>
2: 2 5 <list>
3: 3 6 <list>

Upvotes: 1

Related Questions