Use of ! (or any logical operator) with %>% (magrittr) produces unexpected output

Question

I have run across a situation where %>% produces very surprising output when combined with !. Consider the following code:

x <         


        

Answer 1

I suspect that it's an order of operations issue:

!is.na(x) %>% sum

is evaluating to

!(is.na(x) %>% sum)

Which is equivalent to TRUE

Answer 2

You can also use the "not" alias from the magrittr package:

> is.na(1:20) %>% not %>% sum

[1] 20

Answer 3

Although I accepted @C-Z_ 's answer I want to add another to provide context on this. Thanks to @rawr for directing me to ?Syntax.

Basically %>% is considered to be an operator, like %in% and as such it has to obey the order of operations. On the Syntax help page this corresponds to the %any% operator (i.e. any infix operator), as users can define these at will. As it happens, this means that %>% fires before any logical operator and also before arithmetic operators (e.g. * and \). As a result, if you are naively, like I was, thinking that the left side of %>% will complete before the next step in the chain, you can get some surprises. For example:

3+2 %>% '*'(4) %>% `/`(2)

Does not do 3+2=5, 5*4= 20, 20/2=10

instead it does 2*4/2=4, 4+3=7, because the %>% has precedence over +.

If you use the functions in the magrittr package such as:

add(3,2) %>% multiply_by(4) %>% divide_by(2) 

You get 10 as expected. Placing brackets around the 3+2 will also get you 10.

In my original examples, the logical operators, such as ! have a lower precedence than %>%, so they act last, after the sum has competed.

Moral of the story: Be careful mixing %>% with other operators.