R/fars_functions.R

In swethagarimalla/FARS: Analyze Fatality Data from the US National Highway Traffic Safety Administration's Fatality Analysis Reporting System

#' Load CSV File
#' 
#' @description  This is a simple function that loads a csv file (using the \code{filename} 
#' argument) and converts it into a dataframe. An error is thrown if the file specified by 
#' \code{filename} does not exist.
#' 
#' @param filename A character string that specifies a the location and name of the file
#' to be loaded by this function
#' 
#' @return This function returns a dataframe or tibble. 
#' 
#' @importFrom readr read_csv
#' 
#' @importFrom dplyr tbl_df
#' 
#' @examples 
#' \dontrun{
#' accident_2013 <- fars_read("~/data/accident_2013.csv.bz2")
#' accident_2014 <- fars_read("~/data/accident_2014.csv.bz2")
#' accident_2015 <- fars_read("~/data/accident_2015.csv.bz2")
#' }
#' 
#' @export
fars_read <- function(filename) {
        if(!file.exists(filename))
                stop("file '", filename, "' does not exist")
        data <- suppressMessages({
                readr::read_csv(filename, progress = FALSE)
        })
        dplyr::tbl_df(data)
}

#' Creates Filename
#' 
#' @description  This is a simple function that prints "accident_<year>.csv.bz2",
#' defined by the \code{year} argument.
#' 
#' @param year An integer number to denote the year. 
#' 
#' @return Returns a character string in the form of "accident_<year>.csv.bz2"
#' 
#' @examples 
#' \dontrun{
#' make_filename(2013)
#' make_filename(2014)
#' make_filename(2015)
#' }
#' 
#' @export
make_filename <- function(year) {
        year <- as.integer(year)
        file<-sprintf("accident_%d.csv.bz2", year)
        ext_data <- system.file('extdata', file, package = 'FARS')
        ext_data
}


#' Retrieves Month & Year from the Accident Year Files
#' 
#' @description  This is a simple function that takes a numerical list of integers that 
#' denotes \code{years}, accesses the accident file for each \code{year} of the list of
#' \code{years} and return a list of dataframes of \code{MONTH} and \code{year}, one
#' for each accident file, each associated with a year. The accident files for each year 
#' within list must be located within the current working directory and must be named in 
#' the format 'accident_<year>.csv.bz2'
#' 
#' @param years A list of integer numbers, each of which denote a year. 
#' 
#' @return Returns a list of dataframes/tibbles of months (under the column 'MONTH') and 
#' years (under the column 'year'). Each of these tibbles is associated with each accident 
#' file that is associated with each of the \code{year} elements of the list of \code{years}.
#' If a \code{year} within the list does not have an associated file in the current working
#' directory named in the format 'accident_<year>.csv.bz2', then a warning is thrown and returns
#' NULL. 
#' 
#' @examples 
#' \dontrun{
#' fars_read_years(2013)
#' fars_read_years(list(2013, 2014))
#' fars_read_years(2013:2015)
#' 
#' #A not-found warning is thrown and null is returned for the following
#' fars_read_years(list(2013, 2014, 2015, 2016))
#' fars_read_years(2017)
#' }
#' 
#' 
#' 
#' @export
fars_read_years <- function(years) {
        lapply(years, function(year) {
                file <- make_filename(year)
                tryCatch({
                        dat <- fars_read(file)
                        dplyr::mutate(dat, year = ~ year) %>% 
                                dplyr::select(~ MONTH, ~ year)
                }, error = function(e) {
                        warning("invalid year: ", year)
                        return(NULL)
                })
        })
}

#' Count the number of accidents within each month for each year
#' 
#' @description  This is a simple function that takes a list of numerical integers,
#' each element of which denotes a year, and produces a data frame with the number of accidents
#' for each month within each year. 
#' 
#' @param years A list/vector of integer numbers, each of which denote a year.
#' 
#' @return Returns a pivot data frame containing two columns (one for accident count, and 
#' one for each year in the \code{years} list/vector), where each month is a row. As per use
#' of the fars_read_years function, a warning will be returned if an element of \code{years}
#' does not have an associated file.
#' 
#' @examples 
#' \dontrun{
#' fars_summarize_years(2013)
#' fars_summarize_years(2013:2014)
#' fars_summarize_years(list(2013, 2014, 2015))
#' }
#' 
#' @importFrom dplyr bind_rows %>% group_by summarize
#' @importFrom tidyr spread
#' @export
fars_summarize_years <- function(years) {
        dat_list <- fars_read_years(years)
        dplyr::bind_rows(dat_list) %>% 
                dplyr::group_by( ~ year, ~ MONTH) %>% 
                dplyr::summarize(n = ~ n()) %>%
                tidyr::spread(~ year, ~ n)
}


#' Maps accidents onto states
#' 
#' @description  Takes input of a \code{state.num} and \code{year} as arguments and plots the 
#' accidents onto a map of the states. If the state number is invalid, an error is thrown. 
#' If there are no accidents in that state, a message is returned that there are no 
#' accidents to plot.
#' 
#' @param state.num A numerical integer denoting the US state as is shown in the data set
#' @param year A numerical integer denoting the year
#' 
#' @return Returns a plot of states with the number of accidents on each states
#' the accidents based on the \code{year}. Returns an error if \code{state.num} or if 
#' \code{year} do not exist in the data set.
#' 
#' @examples 
#' \dontrun{
#' fars_map_state(20, 2013)
#' fars_map_state(10, 2014)
#' fars_map_state(30, 2016)
#' 
#' fars_map_state(72, 2013) #Error because \code{state.num} doesn't exist
#' fars_map_state(20, 2020) #Error because \code{year} doesn't exist
#' }
#' 
#' @importFrom dplyr filter
#' @importFrom maps map
#' @importFrom graphics points
#' 
#' @export

fars_map_state <- function(state.num, year) {
  filename <- make_filename(year)
  data <- fars_read(filename)
  state.num <- as.integer(state.num)
  
  if(!(state.num %in% unique(data$STATE)))
    stop("invalid STATE number: ", state.num)
  data.sub <- dplyr::filter_(data, ~ STATE == state.num)
  if(nrow(data.sub) == 0L) {
    message("no accidents to plot")
    return(invisible(NULL))
  }
  is.na(data.sub$LONGITUD) <- data.sub$LONGITUD > 900
  is.na(data.sub$LATITUDE) <- data.sub$LATITUDE > 90
  with(data.sub, {
    maps::map("state", ylim = range(LATITUDE, na.rm = TRUE),
              xlim = range(LONGITUD, na.rm = TRUE))
    graphics::points(LONGITUD, LATITUDE, pch = 46)
  })
}