across in R

Applying the same transformation to several columns at once.

library(dplyr)

runs <- tibble(
  lang = c("r", "python"),
  tpm = c(96.4, 104.2),
  accuracy = c(97.2, 99.1)
)

print(mutate(runs, across(where(is.numeric), round, 1)))

print(summarise(runs, across(c(tpm, accuracy),
                            list(mean = mean, max = max),
                            .names = "{.col}_{.fn}")))

print(mutate(runs, across(where(is.character), toupper)))
print(summarise(runs, across(everything(), ~ length(unique(.x)))))

How it works

1. across(where(is.numeric), fn) is the common shape.
2. A named list of functions produces one column per pair.
3. .names controls what the new columns are called.

Keywords and builtins used here

• across
• c
• everything
• length
• library
• list
• mutate
• print
• summarise
• tibble
• unique
• where

The run, in numbers

Lines

16

Characters to type

394

Tokens

129

Three-star pace

95 tpm

At the three-star pace of 95 tokens a minute, this run takes about 81 seconds.

Type this snippet

Step 3 of 4 in New columns, step 12 of 27 in dplyr.

← Previous Next →