R/get_unique.R

In pcctc/croquet: Miscellaneous Operations for PCCTC

#' Get unique values from a variable in a data frame
#'
#' Values are returned in their respective natural ordering, whether it be numeric,
#' character, or factor. If the input variable is a factor, get_unique()
#' removes factor levels and returns a character vector.
#'
#' @param data data frame
#' @param var variable from which we want unique values
#' @param drop_na whether or not to drop NA from returned values; defaults to TRUE
#' @param keep_fct_levels whether or not to keep all levels of factor (instead
#' of only values present in data); defaults to TRUE
#'
#'
#' @return a vector (either character or numeric, depending on input)
#' @noRd
#'
#' @examples
#' dat <- tibble::tibble(
#'   w = c(NA, 1, "B"),
#'   x = c(1, 1, 2),
#'   y = c("B", "B", "A"),
#'   z = factor(x = c("cat", "cat", "dog"), levels = c("dog", "cat")),
#'   u = factor(x = c("cat", "cat", "dog"), levels = c("dog", "pig", "cat"))
#'   )
#' get_unique(dat, w)
#' get_unique(dat, x)
#' get_unique(dat, y)
#' get_unique(dat, z)
#' get_unique(dat, u)
#'
get_unique <- function(data, var, drop_na = TRUE, keep_fct_levels = TRUE){

  var <-
    broom.helpers::.select_to_varnames({{ var }}, data = data,
                        arg_name = "var", select_single = TRUE)

  # list to rename variables----------------------------------------------------
  lst_name_recode <-
    list(var = var)

  # initial data renaming and trimming -----------------------------------------
  data <- data %>% dplyr::select(!!!lst_name_recode)

  # check to see if variable is a factor and we want to return
  # all factor levels
  if (inherits(data[["var"]], "factor") && isTRUE(keep_fct_levels)) {
    values <- levels(data[["var"]])
    return(values)
  }


  # this should run if variable is not a factor, or if variable
  # is a factor but we only want factor levels observed in data to
  # be returned
  values <- data %>%
    dplyr::distinct(var) %>%
    dplyr::arrange(var) %>%
    {if(drop_na) tidyr::drop_na(., var) else . } %>%
    dplyr::pull(var)

  # if factor and do not wish to keep all factor levels
  if (inherits(values, "factor")) values <- as.character(values)

  return(values)
}