summarize_across.(
  .df,
  .cols = everything(),
  .fns = NULL,
  ...,
  .by = NULL,
  .names = NULL
)

summarise_across.(
  .df,
  .cols = everything(),
  .fns = NULL,
  ...,
  .by = NULL,
  .names = NULL
)

test_df <- data.table(
  a = 1:3,
  b = 4:6,
  z = c("a", "a", "b")
)

# Pass a single function
test_df %>%
  summarize_across.(c(a, b), mean, na.rm = TRUE)

# Single function using purrr style interface
test_df %>%
  summarize_across.(c(a, b), ~ mean(.x, na.rm = TRUE))

# Passing a list of functions (with .by)
test_df %>%
  summarize_across.(c(a, b), list(mean, max), .by = z)

# Passing a named list of functions (with .by)
test_df %>%
  summarize_across.(c(a, b),
                    list(avg = mean,
                         max = ~ max(.x)),
                    .by = z)

# Use the `.names` argument for more naming control
test_df %>%
  summarize_across.(c(a, b),
                    list(avg = mean,
                         max = ~ max(.x)),
                    .by = z,
                    .names = "{.col}_test_{.fn}")