R/fars_functions.R

In FunWithForest/Rpackage_fars: Visualization of Accidents in the USA

#Install and load the needed packages:
# install.packages("devtools")
# install.packages("testthat")
# install.packages("roxygen2")
# install.packages("dplyr")
# install.packages("readr")
# install.packages("tidyr")
# install.packages("maps")

# library(roxygen2)
# library(dplyr)
# library(readr)
# library(tidyr)
# library(maps)
# library(devtools)
# library(testthat)

#-------------------------------------------------------------------------------------------------

#Declaring global variables:
globalVariables(names = c('STATE','MONTH','year'), package = 'fars', add = TRUE)

#QPDF implementation:
Sys.setenv(R_QPDF= "F:/Files/Programming/Coursera_Building_R_Packages/qpdf/bin/qpdf.exe")


#' Function \code{fars_read}
#'
#' This is a function that tries to read in a specified csv file and outputs it as a table
#' if the file has been found or exits with a warning message.
#' Within the data object the content of the csv file is written if filename is found.
#' The option progress = FALSE prevents from displaying a progress bar.
#'
#' @param filename It is a string that should contain the filename
#'
#' @return This function returns a data table created by the dplyr package.
#' The package dplyr and readr are needed in order to make this package work.
#'
#' @examples
#' fars_read(system.file("extdata", "accident_2015.csv.bz2", package = "fars"))
#'
#' @importFrom readr read_csv
#' @importFrom dplyr tbl_df
#'
#' @export
fars_read <- function(filename) {
        if(!file.exists(filename))
                stop("file '", filename, "' does not existX")
                data <- suppressMessages({
                readr::read_csv(filename, progress = FALSE)
                })
        dplyr::tbl_df(data)
}


#' Function \code{make_filename}
#'
#' This function takes a year value as input and adds this value into a textstring for the zipped filename.
#' As this is the last performed operation this is also the return value.
#' sprintf is a generic C function that pastes a text string (coming from the base package).
#' The output is created as accident_ followed by the value of the year.
#' @param year Input of a numeric integer value for the year.
#'
#' @return A textstring is returned.
#'
#' @examples
#' make_filename(2015)
#'
#' @export
make_filename <- function(year) {
        year <- as.integer(year)
        #changes due to package implementation of raw data:
        #sprintf("accident_%d.csv.bz2", year)
        file <- sprintf("accident_%d.csv.bz2", year)
        system.file("extdata", file, package = "fars")
}


#' Function \code{fars_read_years}
#'
#' This function takes a vector of years as input and selects values for month and year row by row.
#' @param years A vector of years.
#' file takes the textstring from the \code{make_filename} function.
#' lapply loops over all years and with the tryCatch statement further changes are made (see below).
#' dat takes the data of the files from the next filename.
#' mutate is a function of the dplyr package and in this case it selects month and year.
#' Importing this function from the dplyr package is needed.
#'
#' @return If an invalid year is taken from the selection an error will occur and nothing is returned.
#' Else the values month and year are returned row by row.
#'
#' @examples
#' fars_read_years(2013:2015)
#' fars_read_years(2015)
#'
#' @importFrom dplyr mutate
#' @importFrom dplyr select
#'
#' @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)
                })
        })
}


#' Function \code{fars_summarize_years}
#'
#' This function takes the returned values of other functions and summarizes the values by year and month.
#' @param years In the function call one ore more years are expected to be submitted.

#' The dat_list object takes the input of the \code{fars_read_years} function and delivers the years value
#' to it with the call.
#' bin_rows comes from the dplyr package and takes the dat_list object and groups it by year and month.
#' spread comes from the tidyr package and reorganizes the data according to year and the summarized value of n.
#'
#' @return A table with the columns for the months and each year.
#'
#' @examples
#' fars_summarize_years(2014:2015)
#' fars_summarize_years(2015)
#'
#' @importFrom dplyr bind_rows
#' @importFrom dplyr group_by
#' @importFrom dplyr 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)
}

#' Function \code{fars_map_state}
#'
#' This function takes the state number and a specific year as arguments and prints out a map of accidents in the selected state.

#' @param state.num This is a single integer value with the state number from the input.
#' @param year This value is an input value that gives out either a single year or a range of several years.
#'
#' @return Either a message is given out when there are no accidents or a map of the state and its accidents is shown.
#'
#' @examples
#' fars_map_state(1,2015)
#'
#' @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)
        })
}

#Coursera Week 4 PGA:
#use_vignette("function_details")

#devtools::load_all()
#devtools::check()