R/g_map_covid.R

In fidelmorla/drcovidplots: Plots for COVID19 data from Dominican Republic

#' @title Dominican map with COVID-19 cases by province
#' @aliases g_map_covid
#' @description This function graphs the provinces with the most positive
#'   cases of COVID19 in the DR.
#' @usage g_map_covid(date = "latest", interactive = FALSE, variable = "Cases")
#' @param date Character indecating the date to plot in \code{"yyyy-mm-dd"} format. Default \code{"latest"}
#' @param variable Character. One of the following \code{c("Cases", "Deaths", "Recovered")}
#' @param by_inhabitants Logical. Should take account the province population? Default \code{TRUE}.
#' @param saveplot Logical. Should save the ggplot objet to the \code{.GlobalEnv}? Default \code{FALSE}.
#' @param savepng Logical. Should save a png version of the plot? Default \code{FALSE}.
#'
#' @return Graph the Dominican map with COVID-19 cases by province and saves a
#' copy in png format to the computer at the address defined in \code{setwd()}.
#' @importFrom scales comma
#' @export
#' @examples
#' g_map_covid()
#' g_map_covid(savepng = TRUE)
#' g_map_covid(date = "latest", variable = "Deaths")
#' g_map_covid(date = "2020-04-05", variable = "Recovered")
#' @name g_map_covid

g_map_covid <- function(date = "latest",
                        variable = "Cases",
                        by_inhabitants = FALSE,
                        saveplot = FALSE,
                        savepng = FALSE) {

  if (exists('data_cum') == FALSE) {
    stop("data objects are missing, run load_data_covid_dr()")
  }

  date2 <- date

  `%>%` <- magrittr::`%>%`

  if(date == "latest") {
    data <- data_province %>%
      dplyr::filter(date == max(date))
  } else {
    data <- data_province %>%
      dplyr::filter(date == date2)
  }


  if (variable == "Cases") {
    scale_fill <- colorspace::scale_fill_continuous_sequential(
      p1 = 0.1,
      p2 = 0.4,
      l1 = 40,
      c1 = 100,
      palette = "Blues",
      labels = scales::comma,
      guide = ggplot2::guide_colorbar(barwidth = 12)
    )
  } else if (variable == "Deaths") {
    scale_fill <- colorspace::scale_fill_continuous_sequential(
      p1 = 0.2,
      p2 = 0.4,
      l1 = 40,
      c1 = 185,
      palette = "YlOrRd",
      labels = scales::comma,
      guide = ggplot2::guide_colorbar(barwidth = 12))
  } else if (variable == "Recovered") {
    scale_fill <- colorspace::scale_fill_continuous_sequential(
      p1 = 0.1,
      p2 = 0.4,
      l1 = 40,
      c1 = 150,
      palette = "BuGn",
      labels = scales::comma,
      guide = ggplot2::guide_colorbar(barwidth = 12))
  }


  format_variable <- function(variable) {
    format(sum(data[[variable]], na.rm = TRUE), big.mark = ",")
  }

  format_ratio <- function(variable) {
    ratio = sum(data[[variable]]) / 104.485
    ratio = round(ratio)
    return(ratio)
  }

  if(variable == "Cases" ) {
    title = "DR: Cases of COVID19 by province"
    subtitle = paste0("Total = ", format_variable(variable))
  } else if (variable == "Recovered") {
    title = "DR: Recovered from COVID19 by province"
    subtitle = paste0("Total = ", format_variable(variable))
  } else if (variable == "Deaths") {
    title = "DR: Deaths by COVID19 by province"
    subtitle = paste0("Total = ", format_variable(variable))
  }


  if(by_inhabitants) {

    map_covid <- drcovidplots::map_province %>%
      dplyr::left_join(data_density, by = c("province_short" = "Province")) %>%
      left_join(data_province %>% filter(date == max(date)),
                by = c("province_short" = "Province")) %>%
      select(province_short, province_name, var_toplot = variable, Pop, geometry) %>%
      mutate(
        var_toplot = var_toplot / (Pop / 100000),
        CENTROID = purrr::map(geometry, sf::st_centroid),
        COORDS = purrr::map(CENTROID, sf::st_coordinates),
        COORDS_X = purrr::map_dbl(COORDS, 1),
        COORDS_Y = purrr::map_dbl(COORDS, 2),
        label_short = paste0(province_short, ": ", format(var_toplot, big.mark = ",", digits = 2))
      ) %>%  ggplot2::ggplot() +
      ggplot2::geom_sf(ggplot2::aes(fill = var_toplot)) +
      ggplot2::theme_void() +
      ggplot2::theme(
        legend.position = c(0.5, 0.2),
        legend.direction = "horizontal"
      ) +
      ggplot2::labs(fill = variable) +
      ggrepel::geom_label_repel(ggplot2::aes(COORDS_X, COORDS_Y, label = label_short),
                                size = 3, min.segment.length = 0, point.padding = NA) +
      labs(title = title,
           subtitle = paste0(
             variable,
             " by 100,000 inhabitants = ",
             format_ratio(variable)),
           caption = paste0(
             "Source: @johanRosa_ with the special bulletin #",
             rep_actual,
             " of @SaludPublicaRD, surface and population reports  of @ONERD_")
           ) +
      scale_fill

  } else {
    map_covid <- dplyr::left_join(drcovidplots::map_province,
                                  data,
                                  by = c("province_short" = "Province")) %>%
      select(var_toplot = variable, dplyr::everything()) %>%
      mutate(
        CENTROID = purrr::map(geometry, sf::st_centroid),
        COORDS = purrr::map(CENTROID, sf::st_coordinates),
        COORDS_X = purrr::map_dbl(COORDS, 1),
        COORDS_Y = purrr::map_dbl(COORDS, 2),
        label_short = paste0(province_short, ": ", format(var_toplot, big.mark = ",", digits = 2))
      ) %>%
      ggplot2::ggplot() +
      ggplot2::geom_sf(ggplot2::aes(fill = var_toplot)) +
      ggplot2::theme_void() +
      ggplot2::theme(
        legend.position = c(0.5, 0.2),
        legend.direction = "horizontal"
      ) +
      ggplot2::labs(fill = variable) +
      ggrepel::geom_label_repel(ggplot2::aes(COORDS_X, COORDS_Y, label = label_short),
                                size = 3, min.segment.length = 0, point.padding = NA) +
      labs(
        title = title,
        subtitle = subtitle,
        caption = paste0(
          "Source: @johanRosa_ with the special bulletin #",
          rep_actual,
          " of @SaludPublicaRD and surface reports of @ONERD_")
      ) +
      scale_fill
  }


  print(map_covid) %>% suppressWarnings()

  if (saveplot == TRUE){
    assign('map_covid', map_covid, envir = .GlobalEnv)
  }

  if (savepng == TRUE){

    ggsave(filename = paste0("map_covid_", variable, ".png"),
           plot = map_covid,
           device = "png",
           width = 18.333333333333332 / 1.5,
           height = 10.466666666666667 / 1.5,
           units = "in")
  }
}