I am experiencing odd behavior with data.table when I change a copied version of dataset. Here is code that reproduces the error for me:
library(data.table)
set.seed(02138)
data1 <- data.table(var1 = 1:10, var2 = sample(letters, 10))
data2 <- data1
data2[var1 == 3, var3 := 1]
data2[var1 != 3, var3 := 0]
I get the following when I print out the original and copied data.tables:
> data1
var1 var2 var3
1: 1 n 0
2: 2 p 0
3: 3 i 1
4: 4 k 0
5: 5 e 0
6: 6 d 0
7: 7 a 0
8: 8 x 0
9: 9 y 0
10: 10 b 0
> data2
var1 var2 var3
1: 1 n 0
2: 2 p 0
3: 3 i 1
4: 4 k 0
5: 5 e 0
6: 6 d 0
7: 7 a 0
8: 8 x 0
9: 9 y 0
10: 10 b 0
Notice that the new variable has been created both in the copied version of the data.table and in the old version. I only want it in the new version. What is going on here?
data.table version: 1.9.4
R version: 3.1.1
OS: Mac OS X Yosemite 10.10
