R data.table doing an inner join on a field and operating on another?

Question

I have the following scenario, I first create a data table as shown below

x = data.table(f1 = c('a','b','c','d'))
x = x[,rn := .I]

This yields

> x
   f1 rn
1:  a  1
2:  b  2
3:  c  3
4:  d  4
>

Where rn is simply the row number. Now, I have another data.table y as

y = data.table(f2=c('b','c','f'))

What I would like to be able to do is for elements in y that are in x, I want to subtract 2 from the corresponding values in rn. So the expected data.table is

x
   f1 rn
1: a  1
2: b  0
3: c  1
4: d  4

How does one get to this? x[y] and y[x] don't help at all as they just do joins.

Answer 1

You can use %chin% in i to subset x by the required rows and then run your j expression...

x[ f1 %chin% y$f2 , rn := rn - 2L ]
x
#   f1 rn
#1:  a  1
#2:  b  0
#3:  c  1
#4:  d  4

%chin% is a fast version of the %in% operator specifically for character vectors that comes with data.table. Note that 2 should be 2L to specify an "integer" type, otherwise you will get a warning (obviously don't use this if you are dealing with "numeric" data types).