如何基于名称的向量（dplyr）从R data.frame中省略变量

Question

I have data.frame and a long list of variables I need to omit. It would be convenient to use the vector of variable names. Additionally, I'd preferably do this with dplyr or tidyr.

I've already tried this:

df <- data.frame(A = runif(10), B = runif(10))

omits <- c("B")

df %>% select_(.dots = -omits)

That did not work though and I get Invalid argument to unary operator error. Therefore I already know that I'm not doing correctly with the negative sign, but I have not found workable alternative either. The example is very simple, but my actual problem more complicated with much longer omits (and selected) vector. I really want to avoid typing them separately.

Answer 1

You can use one_of to select or in this case omit variables provided in a character vector (those included in the vector omits ):

df %>% select(-one_of(omits))

Output:

           A
1  0.9930896
2  0.4075611
3  0.3654101
4  0.2161043
5  0.4034992
6  0.8350059
7  0.4828840
8  0.1134290
9  0.2902616
10 0.4052897

Answer 2

We can use

 df %>% 
    select_(.dots= setdiff(names(.), omits))
#          A
#1  0.5479797
#2  0.7071427
#3  0.9020869
#4  0.4815137
#5  0.8413017
#6  0.5813052
#7  0.1528620
#8  0.4352227
#9  0.8802945
#10 0.9828040