I've been playing around with data.tables and noticed some strange behaviour & not sure if I'm doing something wrong or not.
If I reassign a data.table to another name then drop a column from the new table it also drops it from the first table. Eg:
a <- data.table(x=rnorm(10,3),y=rnorm(10,3),z=rnorm(10,3))
> dim(a)
[1] 10 3
b <- a
b[,z:=NULL]
> dim(a)
[1] 10 2
> dim(b)
[1] 10 2
Yet if I use the data.frame approach it doesn't impact a. Eg:
> b$z <- NULL
> dim(a)
[1] 10 3
> dim(b)
[1] 10 2
Am I doing something wrong with data.tables or is it just a quirk?
