Alternatives to deparse(substitute()) to get name of object within a function

In vignette("programming", package = "dplyr") there is the following example code

my_summarise2 <- function(data, expr) {
  data %>% summarise(
    mean = mean({{ expr }}),
    sum = sum({{ expr }}),
    n = n()
  )
}

I want to adapt this so the name of the expr object is included in the output, and I can do that using deparse(substitute(())

my_summarise2 <- function(data, expr) {
  exprname = deparse(substitute(expr))
  data %>% summarise(
    var = exprname,
    mean = mean({{ expr }}),
    sum = sum({{ expr }}),
    n = n()
  )
}

my_summarise2(iris, Petal.Width)

#          var     mean   sum   n
#1 Petal.Width 1.199333 179.9 150

Is there another tidyverse-recommended way to achieve the same result?

We can use rlang::enquo and rlang::as_name:

library(dplyr)

my_summarise3 <- function(data, expr) {
  exprname = rlang::as_name(rlang::enquo(expr))
  data %>% summarise(
    var = exprname,
    mean = mean({{ expr }}),
    sum = sum({{ expr }}),
    n = n()
  )
}

my_summarise3(iris, Petal.Width)
#>           var     mean   sum   n
#> 1 Petal.Width 1.199333 179.9 150

Or using rlang::enexpr and rlang::as_label:

library(dplyr)

my_summarise3 <- function(data, expr) {
  exprname = rlang::as_label(rlang::enexpr(expr))
  data %>% summarise(
    var = exprname,
    mean = mean({{ expr }}),
    sum = sum({{ expr }}),
    n = n()
  )
}

my_summarise3(iris, Petal.Width)
#>           var     mean   sum   n
#> 1 Petal.Width 1.199333 179.9 150

^{Created on 2025-03-13 with reprex v2.1.1}