CODEWITHSUNDEEP
rlistgroup-bydata.tablerange
LucasMation
LucasMation

Reputation: 2489

create column with union of ranges by group

I am trying to write code, preferably using data.table, to group data by id from a list column with ranges (my_range) into a new aggregate dataset with a column (union_of_my_range) containing the union of my_range for each id.

library(data.table)

#Input
d <- data.table(id=c(1,1,2,2),
                my_range=list(1:5,2:6,3:7,9:9)
                )

#Desired output:
g <- data.table(id=c(1,2),
                union_of_my_range=list(c(1:5,6),c(3:7,9))
                )

Preferably looking for a data.table solution,

This is what I have thus far:

d[, .(my_range_union = SOMETING unon my_range ) ,id]

but I am not able to keep track of all the lists interacting in the aggregation step

Upvotes: 1

Views: 138

Answers (2)

akrun
akrun

Reputation: 887391

We can use Reduce with union

d[, .(my_range_union = list(Reduce(union, my_range))), id]
#   id my_range_union
#1:  1    1,2,3,4,5,6
#2:  2    3,4,5,6,7,9

Upvotes: 1

Ronak Shah
Ronak Shah

Reputation: 389065

You can unlist my_range values, gather unique values and put it in a new list.

library(data.table)
d[, .(my_range_union = list(unique(unlist(my_range)))) ,id]

#   id my_range_union
#1:  1    1,2,3,4,5,6
#2:  2    3,4,5,6,7,9

Upvotes: 1

Related Questions