trying to use select in dplyr in a more compact way

Question

I'm trying to use dplyr and the %>% symbol with multiple selection. What works:

   select(data, b1:b10) / select(data, v1:v10)

What I would like to do:

data %>% select(b1:b10)/select(v1:v10)   

Answer 1

What about

#sample data
data<-data.frame(matrix(runif(20*45), ncol=20, 
    dimnames=list(NULL, paste0(rep(c("b","v"), each=10),1:10))) )

#orig
A<-select(data, b1:b10) / select(data, v1:v10)

#proposed
B<-data %>% function(X) {select(X, b1:b10)/select(X, v1:v10)}

#verify they are the same
all(A==B)

I also have a helper function called withX that i use a lot in situations like this

C <- data %>% withX( select(X, b1:b10) / select(X, v1:v10) )
all(C==A)
# [1] TRUE

Or maybe in this case we might like another helper function for working the the chain and creating a list form different subsets. Consider the function each

each<-function(.data, ...) {
    dots <- substitute(list(...))[-1]
    force(.data)
    e<-environment()
    parent.env(e)<-parent.frame()
    lapply(dots, function(x) {
        eval(bquote(.data %>% .(x)), e)
    })
}

This function will perform the dplyr chaining for each parameter you pass in. So you can do

D <- data %>% each(select(b1:b10), select(v1:v10)) %>% Reduce(`/`, .)
all(A==D)
# [1] TRUE

Now I realize all these alternative methods aren't "compact" in the sense of fewer characters. But you do only have to specify the data.frame name once. So there's that.