R/workweek_weeknd_pmPlot.R

In iozeroff/pavisualizR: Purple Air Sensor Data Visualizations

#' @export
#'
#' @title Plots Particulate Matter 2.5 Hour of Day Averages with lines showing
#'   Workweek and Weekend Averages
#'
#' @description Plot the mean PM2.5 value for each hour of the day, with
#'   separate lines for weekdays and weekend values for one or more sensors.
#'
#' @param sensor_list A list of AirSensor objects.
#' @param aqi_country Character vector ISO 3166-1 alpha-2 country code of
#'   country who's Air Quality Index (AQI) should be used for plotting
#'   background. Current options are United States ("US") or India ("IN").
#' @param sensor_colors Character Vector the length of unique sensors in
#'   outliercount_list, of hexadigit colors to use for distinguishing sensor
#'   lines. If NULL, the default, a random palette will be produce. See details
#'   for more information.
#' @param aqi_bar_alpha Numeric between 0 and 1 indicating the opacity of the
#'   AQI background colors. Default is 0.2
#'
#' @return ggplot object.

# TODO: Make sensor_list argument ... to be passed any number of sensors or list.

workweek_weeknd_pmPlot <- function(
  sensor_list,
  aqi_country,
  sensor_colors = NULL,
  aqi_bar_alpha = 0.2
) {

  # TODO: Add docstring.
  # TODO: Add error control system.

  if ( any(!purrr::map_lgl(sensor_list, sensor_isSensor)) ) {
    stop("All elements of sensor_list must be AirSensor objects.")
  }

  if ( is.null(sensor_colors) ) {
    # Making palette of colors for sensors.
    sensor_colors <- RColorBrewer::brewer.pal(n = length(sensor_list), name = "Dark2")
    # Naming the vector using sensor labels.
    names(sensor_colors) <- names(sensor_list)
  }

  # Loading AQI Categorical Index info for plotting.
  aqi <- aqi_info[[aqi_country]]

  # Readying plotting set.
  data <- sensor_list %>%
    purrr::map(.f = sensor_extractData) %>%
    # Rename second column to pm25.
    purrr::map(.f = function(x) dplyr::rename(x, pm25 = 2)) %>%
    # Convert to df.
    dplyr::bind_rows(.id = "sensor") %>%
    dplyr::mutate(
      # Getting day of week for each hour.
      weekday = weekdays(datetime),
      # Extracting hour.
      hour = format(datetime, "%H"),
      # Workweek 1 or 0.
      workweek = factor(if_else(weekday %in% c("Monday", "Tuesday", "Wednesday",
                                               "Thursday", "Friday"),
                                "workweek",
                                "weekend"))
    )

  # Getting Timezone from data.
  timezone <- attr(data$datetime,"tzone")

  # Creating x label with timezone.
  x_label <- paste("Hour of Day ", "(", timezone, ")", sep = "")

  # Summarizing data over workweek or weekend.
  summary_data <-
    data %>%
    dplyr::group_by(hour, workweek, sensor) %>%
    dplyr::summarize(pm25 = mean(pm25, na.rm=TRUE))

  ymax <- max(summary_data$pm25, na.rm = TRUE)

  plot <-
    summary_data %>%
    ggplot2::ggplot(aes(x = hour, y = pm25,
                        group = interaction(sensor, workweek))) +
    ggplot2::geom_line(aes(color = sensor, linetype = workweek), size = 1) +
    ggplot2::annotate("rect",
             ymin = aqi$aqi_pm_mins,
             ymax = aqi$aqi_pm_maxs,
             xmin = -Inf,
             xmax = Inf,
             alpha = aqi_bar_alpha,
             fill = aqi$colors) +
    ggplot2::scale_linetype_manual(values=c("dotted", "solid")) +
    ggplot2::scale_color_manual(values = sensor_colors) +
    ggplot2::coord_cartesian(ylim = c(0, ymax)) +
    ggplot2::labs(
      x = x_label,
      y = "PM 2.5 microgram/m3",
      title = "Average Particulate Matter 2.5 Concentration by Hour of Day Weekday vs. Weekend",
      color = "Sensor Label",
      linetype = "Day Type")

  return(plot)
}