accessing Y columns with duplicated names in j of X[Y, j] merges

Question

Suppose I have data like this:

set.seed(1)
DT <- data.table(id=rep(1:3,each=3),y=1997L+sample(1:9,9))
DT2<- data.table(id=1:3,y=1997L+sample(1:3,3))

I want to use DT2$y after merging with DT. I see that this column is named y.1 after the merge

setkey(DT,id)
names(DT[DT2])
# [1] "id"  "y"   "y.1"
DT[DT2][,y.1]
# [1] 1998 1998 1998 2000 2000 2000 1999 1999 1999

However, I cannot use it with that name in j:

DT[DT2,y.1]
# Error in `[.data.table`(DT, DT2, y.1) : object 'y.1' not found

What is the secret prefix or postfix that I should be using here?

score 6 · Accepted Answer · answered May 30 '13 at 18:55

6

You can use

DT[DT2, i.y]

and if you find yourself surprised that it's not the same output as DT[DT2][, y.1], see this thread

answered May 30 '13 at 18:55

eddi

Ah, thanks. That's the expected/desired behavior, as far as my uses go. I don't see `.1` or `i.` in the documentation (`?"[.data.table"` data.table 1.8.8), though I think I saw it before. I wonder why they aren't both named `i.`... – Frank May 30 '13 at 19:00

1 Answers1