tolower on dataframe

[fcas80]

Hi. tolower works fine on a vector, but produces strange results (but not an error) on a dataframe. What is it doing with the dataframe?

x = c("Bug","Cat","dOG","emU","FroG")
tolower(x)

y <- c("a","B","C","d","E")
df <- data.frame(x, y)
tolower(df)

[1] "bug" "cat" "dog" "emu" "frog"
[1] "c("bug", "cat", "dog", "emu", "frog")"
[2] "c("a", "b", "c", "d", "e")"

The last two lines above are not displaying " c ( slash "a slash " etc.

[cmeuli07]

Using across() within the mutate() function will ensure that you apply tolower to each column of the df individually

require(dplyr)


x = c("Bug","Cat","dOG","emU","FroG")
y <- c("a","B","C","d","E")

df <- data.frame(x, y)

df %>% 
  mutate(., across(.cols = everything(), tolower))

[fcas80]

cmeuli07,

Thank you for your solution.

But why does tolower applied to a dataframe produce such strange results?

[FactOREO]

Because, according to the documentation, it can only be applied to character vectors or such objects which can be coerced to character. Hence, using tolower() on a data.frame coerces the data.frame into a chr vector first and then does the lowering of letters. But since there is no way back, you end up with a coerced object which has been transformed to lower letters only.

[nirgrahamuk]

because toLower will attempt to convert anything that is not a character to a character and go from there

as.character(df)

going further than that, I understand there to be a relationship to the behaviour of

c(df)

i.e. the stripping of the data.frame class from the underlying list of the contents of the data.frame