downloading covid data for JHU.R

library(tidyverse)
library(httr)

latest_two_csv_files <-
  GET("https://api.github.com/repos/CSSEGISandData/COVID-19/git/trees/master?recursive=1") %>% 
  jsonlite::parse_json() %>% 
  pluck("tree") %>% 
  map(pluck("path")) %>% 
  unlist() %>% 
  as_tibble() %>% 
  filter(str_detect(value, "csse_covid_19_data/csse_covid_19_daily_reports/") & str_detect(value, ".csv")) %>% 
  mutate(date = word(value, -1, sep = "/"),
         date = word(date, 1, sep = "\\."),
         date = lubridate::mdy(date)) %>% 
    arrange(desc(date)) %>% 
    head(2) %>% # keep only the latest 2 csv paths
    mutate(full_url = paste0("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/", value)) %>% 
    mutate(downloaded_data = map(full_url, ~read_csv(.x
                                                     ## if you are using older files, sometimes it is better to manually specify the column datatype
                                                     # , col_types = cols(.default = "c") 
                                                     ))) %>% 
    unnest(downloaded_data) %>%    
   
   ## if combining with older files (which used different columns)
   
    # unite(Latitude, Lat, na.rm = TRUE) %>% 
    # unite(Longitude, Long_, na.rm = TRUE) %>% 
    # unite(Last_Update, `Last Update`, na.rm = TRUE) %>% 
    # unite(Province_State, `Province/State`, na.rm = TRUE) %>% 
    # unite(Country_Region, `Country/Region`, na.rm = TRUE) %>% 
   
    janitor::clean_names() %>% 
  
  ## if using cols(.default = "c") above
  
    # mutate(last_update = lubridate::mdy_hm(last_update)) %>% 
    # mutate_at(.vars = c("latitude", "longitude", "confirmed", "deaths", "recovered", "active"),
    #           .funs = as.numeric) %>% 
  
  
    select(-value, -incidence_rate, -case_fatality_ratio)