Why does combination of rename_with() and str_replace() fail?

Question

I am trying to replace "Abundance." in the names of the columns in my tibble. I am using the following approach:

Error when using dplyr rename_at and stringr str_replace together to rename columns

xx <- structure(list(Sequence = c("HSEAATAQR", "GAAAHPDSEEQQQR"), Abundance..126 = c(12.3, 12.2), Abundance..127N = c(17.7, 13.5), Abundance..127C = c(11.9, 11.3)), row.names = 1:2, class = "data.frame") 
rename_with(xx, str_replace(., pattern = "Abundance.", replacement = ""), contains("Abundance."))

I get the error: "Error in vctrs::vec_size_common(string = string, pattern = pattern, replacement = replacement, : object '.' not found"

I understand that it means that the str_replace() function does not receive the string argument that I am trying to pass to it as ".", but it does seem to work in the example mentioned above

Error when using dplyr rename_at and stringr str_replace together to rename columns

Answer 1

You are missing a ~operator. Use the following

xx %>% 
  rename_with(~str_replace(., pattern = "Abundance.", replacement = ""), contains("Abundance."))

The functions argument must be either

• a named function (e.g. a function defined before, see example later)
• an anonymous function (like \(x) str_replace(x, pattern = "Ab", ""))
• or a formula (indicated by ~)

Named Functions

To use rename_with with a named function, you might use the following:

delete_abundance <- function(x) {
  str_replace(x, pattern = "Abundance.", replacement = "")
}

xx %>%
  rename_with(delete_abundance, contains("Abundance."))