R/LoadCuba.R

In sjbeckett/subregionalcovid19: Processing and mapping COVID-19 case data at subnational scales

#' LoadCuba
#'
#' @description Reads in subnational data for Cuba to calculate most recent estimate of per capita active COVID-19 cases. Use with LoadData() is recommended.
#'
#' @note
#' Data aggregated from local health resources by Covid19CubaData \url{http://covid19cubadata.github.io/}.
#'
#' @return A simple feature returning the date of most recent data (DateReport), a unique region code (geoid), the region name (RegionName) and country name (Country), the number of active cases per capita (pInf) and the regions geometry (geometry).
#'
#' @examples
#' \dontrun{
#' Cuba <- LoadCuba()
#' }
#' @seealso [LoadData()]
#' @export
LoadCuba <- function() {
  # Data aggregated from local health resources by Covid19CubaData http://covid19cubadata.github.io/
  utils::data("geomCuba", envir = environment())
  geomCuba <- sf::st_as_sf(geomCuba)

  # Old dataset -> July 4 2021
  # cuba_data = read.csv("https://covid19cubadata.github.io/data/covid19-casos.csv", encoding = 'UTF-8')
  # cuba_data <- cuba_data %>%
  #    dplyr::count(fecha_confirmacion, provincia) %>%
  #    dplyr::select(date = fecha_confirmacion, region_level_1 = provincia, cases_new = n) %>%
  #    dplyr::filter(!is.na(region_level_1)) %>%
  #    dplyr::mutate(
  #      cases_new = as.numeric(cases_new),
  #      date = lubridate::as_date(lubridate::ymd(date))
  #    )

  # PRO = unique(cuba_data$region_level_1)
  # LATEST = tail(cuba_data$date,1)
  # DiffCases=c()

  # for(aa in 1:(length(PRO)-1)){
  # 	subsetdata = dplyr::filter(cuba_data,region_level_1==PRO[aa])
  # 	IND = which(subsetdata$date>(LATEST-14))
  # 	DiffCases[aa] = sum(subsetdata$cases_new[IND])*10/length(IND)
  # }

  # data from August 4 2021
  CUBADATA2 <- jsonlite::fromJSON("https://github.com/covid19cubadata/covid19cubadata.github.io/raw/master/data/covid19-cuba-2.json")
  dataset <- CUBADATA2$casos$dias
  DATES <- names(dataset)
  TODAY <- c()
  CASES <- c()
  cuba <- c()
  # PRO = unique(dataset[[DATES[1]]]$diagnosticados$provincia_detección) #provinces

  # geomCuba = st_read("https://raw.githubusercontent.com/covid19cubadata/covid19cubadata.github.io/master/data/provincias.geojson")
  geomCuba$pop <- c(2154454, 679314, 400768, 462114, 525729, 1027683, 830645, 1053837, 84263, 768311, 424750, 783708, 506369, 585452, 487339, 371198) # 2015 census in Cuba

  PRO <- geomCuba$micro_name

  for (aa in 1:length(DATES)) {
    subsetdata <- dataset[[DATES[aa]]]
    names(subsetdata$diagnosticados) <- c("id", "pais", "municipio", "provincia", "codemunicipo", "codeprovince", "contagio", "contacto_focal") # diacritics might affect loading. rename columns to avoid this.
    TODAY <- as.character(as.Date(subsetdata$fecha))
    for (bb in 1:length(PRO)) {
      CASES[bb] <- sum(subsetdata$diagnosticados$provincia == PRO[bb])
    }
    cuba <- rbind(cuba, c(TODAY, CASES))
  }
  cuba_data <- data.frame(cuba)
  names(cuba_data) <- c("date", PRO)

  # print(cuba_data)

  LATEST <- utils::tail(cuba_data$date, 1)
  DiffCases <- c()
  IND <- which(cuba_data$date > (as.Date(LATEST) - 14))
  CASES <- matrix(as.numeric(as.matrix(cuba_data[, -1])), nrow = nrow(cuba_data[, -1]))
  DiffCases <- colSums(CASES[IND, ]) * 10 / 14

  cubaTable <- c()
  cubaTable$Pro <- PRO # PRO[1:(length(PRO)-1)]
  cubaTable$DiffCases <- DiffCases
  cubaTable$date <- LATEST
  cubaTable <- as.data.frame(cubaTable)

  # print(cubaTable)


  MapCuba <- dplyr::inner_join(geomCuba, cubaTable, by = c("micro_name" = "Pro"))

  MapCuba$RegionName <- paste(MapCuba$micro_name, MapCuba$country_name, sep = ", ")
  MapCuba$Country <- MapCuba$country_name
  MapCuba$DateReport <- as.character(MapCuba$date)
  MapCuba$pInf <- MapCuba$DiffCases / MapCuba$pop
  CUBA_DATA <- subset(MapCuba, select = c("DateReport", "geoid", "RegionName", "Country", "pInf", "geometry"))
  return(CUBA_DATA)
}