Modifying a value in a data.table in R

Question

I'm new with data.table, and I'm having an issue with this class. I have a table (data1) with 2 columns: Couple and Ratio. Couple is the Key of the data.table. I'm trying to modify a value in the table.

In the following code, cple is an existing value of Couple. When I run it, I get the results shown in the comments:

data1[cple]$Ratio[1]<-0 # I get more than 50 warnings and it doesn't work
data1$Ratio[1]<-0 # It works perfectly (but it's not the same as the above code)

The error seems to have something to do with Keys but I cannot figure out what it means.

Below is an example:

>data1<-data.table(Couple=c("a","a","b","b"),Ratio=1:4)
>data1
   Couple Ratio
1:      a     1
2:      a     2
3:      b     3
4:      b     4

>setkey(data1,Couple)

>data1["a"]$Ratio[1]<-2 #doesn't work warning message

WARNING:
#In `[<-.data.table`(`*tmp*`, "a", value = list(Couple = c("a", "a" :
 # Coerced 'double' RHS to 'integer' to match the column's type; may have truncated precision. Either change the target column to 'double' first (by creating a new 'double' vector length 4 (nrows of entire table) and assign that; i.e. 'replace' column), or coerce RHS to 'integer' (e.g. 1L, NA_[real|integer]_, as.*, etc) to make your intent clear and for speed. Or, set the column type correctly up front when you create the table and stick to it, please.

>data1$Ratio[1]<-2 #works 
>data1
   Couple Ratio
1:      a     2
2:      a     2
3:      b     3
4:      b     4

So I can change the value when accessing the element by its indices, but not when accessing the value by its value. How can I do this?

Answer 1

you shouldn't use $ with data.table when you ASSIGN data.table is a daughter class of data.frame but it is much better as it can update by reference, without copies. Each time you try to assign with $ like data1$Ratio[1]<-2 it copies the whole table. You should look at the vignette and especially the update := operator. In your case `data1[Couple=='a',Ratio:=c(0L,Ratio[-1])] is what you want.

You might want to read this very good post too.