global.R

# ---- Loading libraries ----
library("shiny")
library("shinydashboard")
library("tidyverse")
library("leaflet")
library("plotly")
library("DT")
library("fs")
library("wbstats")

source("utils.R", local = T)

downloadGithubData <- function() {
  download.file(
    url      = "https://github.com/CSSEGISandData/COVID-19/archive/master.zip",
    destfile = "data/covid19_data.zip"
  )
  
  data_path <- "COVID-19-master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_"
  unzip(
    zipfile   = "data/covid19_data.zip",  
    files     = paste0(data_path, c("confirmed_global.csv", "deaths_global.csv", "recovered_global.csv")),
    exdir     = "data",
    junkpaths = T
  )
}


updateData <- function() {
  # Download data from Johns Hopkins (https://github.com/CSSEGISandData/COVID-19) if the data is older than 0.5h
  if (!dir_exists("data")) {
    dir.create('data')
    downloadGithubData()
  } else if ((!file.exists("data/covid19_data.zip")) || (as.double(Sys.time() - file_info("data/covid19_data.zip")$change_time, units = "hours") > 0.5)) {
    downloadGithubData()
  }
}

# Update with start of app
updateData()

# TODO: Still throws a warning but works for now
data_confirmed <- read_csv("data/time_series_covid19_confirmed_global.csv")
data_deceased  <- read_csv("data/time_series_covid19_deaths_global.csv")
# data_recovered <- read_csv("data/time_series_covid19_recovered_global.csv")

# Get latest data
current_date <- as.Date(names(data_confirmed)[ncol(data_confirmed)], format = "%m/%d/%y")
changed_date <- file_info("data/covid19_data.zip")$change_time

# Get evolution data by country
data_confirmed_sub <- data_confirmed %>%
  pivot_longer(names_to = "date", cols = 5:ncol(data_confirmed)) %>%
  group_by(`Province/State`, `Country/Region`, date, Lat, Long) %>%
  summarise("confirmed" = sum(value, na.rm = T))

# data_recovered_sub <- data_recovered %>%
#   pivot_longer(names_to = "date", cols = 5:ncol(data_recovered)) %>%
#   group_by(`Province/State`, `Country/Region`, date, Lat, Long) %>%
#   summarise("recovered" = sum(value, na.rm = T))

data_deceased_sub <- data_deceased %>%
  pivot_longer(names_to = "date", cols = 5:ncol(data_deceased)) %>%
  group_by(`Province/State`, `Country/Region`, date, Lat, Long) %>%
  summarise("deceased" = sum(value, na.rm = T))

data_evolution <- data_confirmed_sub %>%
  full_join(data_deceased_sub) %>%
  ungroup() %>%
  mutate(date = as.Date(date, "%m/%d/%y")) %>%
  arrange(date) %>%
  group_by(`Province/State`, `Country/Region`, Lat, Long) %>%
  mutate(
    recovered = lag(confirmed, 14, default = 0) - deceased,
    recovered = ifelse(recovered > 0, recovered, 0),
    active = confirmed - recovered - deceased
  ) %>%
  pivot_longer(names_to = "var", cols = c(confirmed, recovered, deceased, active)) %>%
  ungroup()

# Calculating new cases
data_evolution <- data_evolution %>%
  group_by(`Province/State`, `Country/Region`) %>%
  mutate(value_new = value - lag(value, 4, default = 0)) %>%
  ungroup()

rm(data_confirmed, data_confirmed_sub, data_recovered, data_recovered_sub, data_deceased, data_deceased_sub)































# ---- Download population data ----
population                                                            <- wb(country = "countries_only", indicator = "SP.POP.TOTL", startdate = 2018, enddate = 2020) %>%
  select(country, value) %>%
  rename(population = value)
countryNamesPop                                                       <- c("Brunei Darussalam", "Congo, Dem. Rep.", "Congo, Rep.", "Czech Republic",
                                                                           "Egypt, Arab Rep.", "Iran, Islamic Rep.", "Korea, Rep.", "St. Lucia", "West Bank and Gaza", "Russian Federation",
                                                                           "Slovak Republic", "United States", "St. Vincent and the Grenadines", "Venezuela, RB")
countryNamesDat                                                       <- c("Brunei", "Congo (Kinshasa)", "Congo (Brazzaville)", "Czechia", "Egypt", "Iran", "Korea, South",
                                                                           "Saint Lucia", "occupied Palestinian territory", "Russia", "Slovakia", "US", "Saint Vincent and the Grenadines", "Venezuela")
population[which(population$country %in% countryNamesPop), "country"] <- countryNamesDat


# Data from wikipedia
noDataCountries <- data.frame(
  country    = c("Cruise Ship", "Guadeloupe", "Guernsey", "Holy See", "Jersey", "Martinique", "Reunion", "Taiwan*"),
  population = c(3700, 395700, 63026, 800, 106800, 376480, 859959, 23780452)
)
population      <- bind_rows(population, noDataCountries)

data_evolution <- data_evolution %>%
  left_join(population, by = c("Country/Region" = "country"))
rm(population, countryNamesPop, countryNamesDat, noDataCountries)

data_atDate <- function(inputDate) {
  data_evolution[which(data_evolution$date == inputDate),] %>%
    distinct() %>%
    pivot_wider(id_cols = c("Province/State", "Country/Region", "date", "Lat", "Long", "population"), names_from = var, values_from = value) %>%
    filter(confirmed > 0 |
             recovered > 0 |
             deceased > 0 |
             active > 0)
}

data_latest <- data_atDate(max(data_evolution$date))

top5_countries <- data_evolution %>%
  filter(var == "active", date == current_date) %>%
  group_by(`Country/Region`) %>%
  summarise(value = sum(value, na.rm = T)) %>%
  arrange(desc(value)) %>%
  top_n(5) %>%
  select(`Country/Region`) %>%
  pull()