unexpected symbol error while trying to rename columns

Question

So I'm new to R and I am trying to make the columns of this dataset more readable:

with the rename() function from dplyr:

  x <- names(data)
  cols <- gsub('\\.', ' ', x)
  
  for (col in ncol(data)) {
    data <- data %>% rename(cols[col] = names(data)[col])
    col
  }

but I get an error, which I can't fix with my knowledge or google:

Error in source("~/Desktop/r/assignment1/best.R") : 
  ~/Desktop/r/assignment1/best.R:9:45: unexpected '='
8:   for (col in ncol(data)) {
9:     outcome <- data %>% rename(cols[col] =

I would be very grateful if someone could tell me where to look for the solution to this.

Thanks a lot!

It is generally not recommended to have spaces in column names. However, you can simply do names(data) <- gsub(...) No loop needed or anything — Sotos
The issue is that while in base R you can do selective reassignment of vectors with vec[ind] <- 1, that does not work in dplyr functions, so rename(cols[col] = ...) or mutate(cols[col] = ...) or similar will not work. There are workarounds (perhaps if_else in some cases), but I believe that the gsub workaround here is more succinct. — r2evans

slava-kohut slava-kohut · Accepted Answer · 2020-07-30T14:38:45

An option is to use dplyr::rename_all. You need to provide a function for renaming. An example would be as follows:

mtcars %>% rename_all(~paste0(.,"_renamed"))

This will add the _renamed postfix to each and every column name. In your case, you can use:

df %>% rename_all(~gsub("\\.", " ", .))

Keep in mind, though, that spaces in column names is BAD idea.

unexpected symbol error while trying to rename columns

2 Answers