plot3.R

########################################################################################
##
##  To perform analysis for this assignment, it is necessary to download dataset from URL:
##  https://d396qusza40orc.cloudfront.net/exdata%2Fdata%2Fhousehold_power_consumption.zip
##
##  The code below checks if the necessary data file exists in the working directory. If not,
##  it will be downloaded from the web and unzipped to your working directory.
##
##  If you want to prevent downloading, make sure that you retrieve and copy a file named
##  "household_power_consumption.txt" to the working directory.
##
########################################################################################

plot3 <- function() {
    ## Load data.table package needed in order to use 'fread' function for fast file reading
    ## If it isn't installed, install the data.table package with install.packages()
    library(data.table)
    
    ## Set the name of data file
    file <- "household_power_consumption.txt"
    
    ## Check if the file exists in the working directory. If not, retrieve it from the web 
    if (!file.exists(file)) {
        fileURL <- "https://d396qusza40orc.cloudfront.net/exdata%2Fdata%2Fhousehold_power_consumption.zip"
        download.file(fileURL, "data.zip", method = "curl") 
        unzip("data.zip")
    }
    
    ## 'fread' function does the same as 'read.table', but faster. 
    ## Here it reads a file in table format where all variables are read in as characters
    alldata <- fread(file, header = TRUE, sep = ";", na.strings = "?", colClasses = "character")
    
    ## Subset data to the required dates for analysis
    data <- alldata[alldata$Date == "1/2/2007" | alldata$Date == "2/2/2007",]
    
    ## Create a POSIXlt/POSIXct class vector of the concatenated values of Date and Time character vectors
    datetime <- strptime(paste(data$Date, data$Time), format = "%d/%m/%Y %H:%M:%S")
    
    ## Set PNG graphics device and create output file
    png("plot3.png", width = 480, height = 480, units = "px")
    
    ## Plot data only for Sub_metering_1 variable
    plot(datetime, as.numeric(data$Sub_metering_1), 
         type = "l",                        ## set the type of plot ("l" for lines)
         xlab = "",                         ## set the label for x-axis (empty as in the reference plot)
         ylab = "Energy sub metering")      ## set the label for y-axis
    
    ## Add data for Sub_metering_2 variable to the plot generated above
    ## color - red, type - lines
    points(datetime, as.numeric(data$Sub_metering_2), col = "red", type = "l")
    
    ## Add data for Sub_metering_3 variable to the plot generated above
    ## color - blue, type - lines
    points(datetime, as.numeric(data$Sub_metering_3), col = "blue", type = "l")
    
    ## Create legend for the plot
    legend("topright",                      ## set the position of the legend in the plot
           lty = 1,                         ## set the simbol to be shown in the legend (lty = 1 for solid line)
           legend = names(data)[7:9],       ## set the character vector to be shown in the legend (names of the plotted variabels)
           col = c("black", "red", "blue"), ## set the color of symbol/lines shown in the legend
           cex = 0.95)                      ## set the size of the legend relative to the default value
                                            ## (slightly reduced to match the reference plot)
          
    ## Close graphics device
    dev.off() 
}