R apply a function on a column if column exist in dataframe

Question

I wonder how can I first check if the column exist in a dataframe, and if yes, how can I modify this column. This should be part of the larger function.

Working example:

# working example
dd <- data.frame(a = c(1,2),
                 b = c(2,3),
                 c = c("a", "f"))
# Check if the "a" filed exist, if yes, change all values the whole column
if("a" %in% colnames(dd))
{
  print(dd$a)
  Encoding(dd$a) <- "UTF-8"
}

This brings me error:

Error in `Encoding<-`(`*tmp*`, value = "UTF-8") : 
  a character vector argument expected 

I feel that there is something wrong in this logic, but I can't figure out the correct application?

Answer 1

To print the relevant column if it exists in the dataframe:

as.character(dd[,"a"]) 

to modify it, if it exists:

 Encoding(as.character(dd[,"a"]))

Also, in R version 4.0, strings are no longer converted into factors automatically, so you are probably using an earlier version of R.

Answer 2

The problem here is that the character vector was converted to a factor variable, which is the default option of data.frame() . The solution here simply is to use stringsAsFactors = FALSE :

  dd <- data.frame(a = c(1,2),
                   b = c(2,3),
                   c = c("a", "Maur\xC3\xADcio"), stringsAsFactors = FALSE)

Encoding(dd$c)
#> [1] "unknown" "unknown"

# Check if the "c" field exists. If yes, convert the encoding of the variable
if("c" %in% colnames(dd)) {
  print(dd$c)
  Encoding(dd$c) <- "UTF-8"

}
#> [1] "a"        "Maurício"

Encoding(dd$c)
#> [1] "unknown" "UTF-8"

^{Created on 2020-05-22 by the reprex package (v0.3.0)}

Note: As the encoding for the Latin alphabets cannot be checked, I've changed the example a bit.