[Rank a data frame with a grouping variable using entirely base R] #R

rank_by_group.R

#' @title
#' Rank a data frame by grouping variable using base R
#'
#' @description
#' This function ranks a specified column in a data frame by group using entirely base R functions. 
#' The underlying function is `rank()`, where additional arguments can be passed with `...`. 
#' The grouping variable is specified as a string using the argument `group_var`, and the variable to rank is 
#' specified using the argument `rank_var`. The operation is analogous to using `group_by()` followed by 
#' `mutate()` in {dplyr}.
#' See example below using the base dataset `iris`. 
#'
#' @param data data frame to be passed through
#' @param group_var string containing the name of grouping column
#' @param rank_var string containing the name of the column to rank
#' @param ... additional arguments to pass to `rank()`
#'
#' @examples
#' rank_by_group(data = iris, group_var = "Species", rank_var = "Sepal.Length")

rank_by_group <- function(data, group_var, rank_var, ...){

  # Vector of grouping variable (unique)
  group_chr <- unique(data[[group_var]])

  # List of data frames
  l_df <- split(x = data, f = group_chr)

  # Ranked list of data frames
  l_df_ranked <-
    lapply(l_df,
           function(df_x) {
             df_x[[rank_var]] <- rank(df_x[[rank_var]], ...)
             df_x
           })

  # Row bind list of data frames
  Reduce(rbind, l_df_ranked)
}