R/utils.R

In FvD/junr: Access Open Data Through the 'Junar' API

#' Clean Currency Values
#'
#' Often Junar data involving currency data is not clean because for some
#' reason content and presentation of currency values is not separated). To
#' clean up this data quickly the following helper function accepts the
#' position of the character value for the currency and the thousands and
#' decimal delimiters.
#'
#' The currency character position defaults to the first character from the
#' left. And it refers directly to the symbol as present in the data frame
#' column, because some characters (such as the  symbol for the Costa Rican
#' Colon) give are uncommon and lead to multiple encoding and font errors.
#'
#' @param currency_column The column in the data frame that contains the
#'    currency values.
#' @param currency_symbol_pos The position from the left of the currency symbol
#'    used. If any spaces are included, please include them here as well.
#' @param thousand_separator The character value that separates thousands
#'    (defaults to ",")
#' @param decimal_separator The character value that separates decimals
#'    (defaults to ".")
#' @keywords Currency, Data Cleaning, Scrubbing
#' @export
clean_currency <- function(currency_column, 
                           currency_symbol_pos=1, 
                           thousand_separator=",",
                           decimal_separator=".") {
  clean_currency <- gsub(get_currency_character(currency_column[1],
                                                currency_symbol_pos), "",
                                                currency_column)
  clean_currency <- gsub(thousand_separator, "", clean_currency)
  if (decimal_separator != ".") {
  clean_currency <- gsub(decimal_separator, ".", clean_currency)
  }
  clean_currency <- as.numeric(clean_currency)
}


#' Get the currency symbol
#'
#' The currency symbol is not always obvious, depending on the currency you are
#' working with. The symbol for Costa Rican Colon, for instance may not display
#' correctly or the same. It may have different appearances in the same R
#' session, and cannot be included here because they lead to built errors.
#'
#' To have a check in place you can request the first character of the currency
#' strings with this function, and use it either to assign the currency value,
#' or to do a sanity check.
#'
#' @param currency_value A single value from the currency column
#' @param currency_symbol_pos The number of positions from the left for the
#'   currency symbol (include any spaces)
#' @export
get_currency_character <- function(currency_value, currency_symbol_pos){
  currency_character <- substring(currency_value, 1, 1)
  return(currency_character)
}