How to select data.table columns whose name is variable

Question

I want to subset a data.table by selecting specific columns. However, I don't want to specify all column names, I want to select them via a variable.

library(data.table)
dt <- data.table(x = 1:10, y = 11:20, z = 1:10)
col <- "x"

Now, I want to do something like this:

dt2 <- dt[, c(col, "y")]

This is supposed to be the same as

dt[, c("x", "y")]

However, dt2 is a character vector containing the values "x" and "y"

I have also tried subsetting the data.table using a list

dt2 <- dt[, .(get(col), y)]

This does work, but it does not keep the column name and I need to rename that column (and possibly multiple columns if I select more columns via a variable). Is there an elegant way to do this?

Answer 1

Add in the , with = FALSE

dt <- data.table(x = 1:10, y = 11:20, z = 1:10)
col <- "x"
dt[, c(col, "y"), with=FALSE]