I heard standard evaluation is not recommended in dplyr, and we can do similar thing with enquo()
and quo()
.
My original code (simplified) is
my_function <- function(data, x="OriginalX", y="OriginalY"){
data %>%
mutate_(CopyX = x, CopyY = y)
}
and it works.
I tried following code
my_function <- function(data, x="OriginalX", y="OriginalY"){
qx <- enquo(x)
qy <- enquo(y)
data %>%
mutate(CopyX = (!!qx), CopyY = (!!qy))
}
Why it does not work? And should we keep using standard evaluation?
The idea behind tidyeval is specifically that you don't need to put your column name between ""
. So this should work:
my_function <- function(data, x= OriginalX , y= OriginalY ){
qx <- enquo(x)
qy <- enquo(y)
data %>%
mutate(CopyX = !!qx,
CopyY = !!qy)
}
Note that the !!qx
and !!qy
don't need to be between parenthesis
my_function(iris, Sepal.Length, Species) %>%
head()
Sepal.Length Sepal.Width Petal.Length Petal.Width Species CopyX CopyY
1 5.1 3.5 1.4 0.2 setosa 5.1 setosa
2 4.9 3.0 1.4 0.2 setosa 4.9 setosa
3 4.7 3.2 1.3 0.2 setosa 4.7 setosa
4 4.6 3.1 1.5 0.2 setosa 4.6 setosa
5 5.0 3.6 1.4 0.2 setosa 5.0 setosa
6 5.4 3.9 1.7 0.4 setosa 5.4 setosa
If you need to use strings in the function parameters, you can use the ensym function to convert them:
my_function <- function(data, x= "OriginalX" , y= "OriginalY" ){
qx <- ensym(x)
qy <- ensym(y)
data %>%
mutate(CopyX = !!qx,
CopyY = !!qy)
}
my_function(iris, "Sepal.Length", "Species") %>%
head()
Sepal.Length Sepal.Width Petal.Length Petal.Width Species CopyX CopyY
1 5.1 3.5 1.4 0.2 setosa 5.1 setosa
2 4.9 3.0 1.4 0.2 setosa 4.9 setosa
3 4.7 3.2 1.3 0.2 setosa 4.7 setosa
4 4.6 3.1 1.5 0.2 setosa 4.6 setosa
5 5.0 3.6 1.4 0.2 setosa 5.0 setosa
6 5.4 3.9 1.7 0.4 setosa 5.4 setosa