R/mymap_NA.R

In borishejblum/COVID19_France_shiny: Visualizations of the Covid19 epidemics in France and Nouvelle Aquitaine

#' Animated maps of Nouvelle Aquitaine
#' 
#' 
myanim_NA <- function(xlsx_file, export_gif=FALSE){
  mfr <- ggplot2::map_data("france")
  na_dpts <- c("Gironde", "Pyrenees-Atlantiques", "Landes", "Lot-et-Garonne", 
               "Dordogne", "Correze", "Creuse", "Haute-Vienne", "Charente", "Vienne",
               "Deux-Sevres", "Charente-Maritime")
  
  rna <- mfr %>% 
    filter(region %in% na_dpts)
  
  cas_NA_ARS <- suppressWarnings(read_xlsx("xlsx_file", 
                                           skip=2, col_types = c("date", "text", "numeric", "text"), ))
  colnames(cas_NA_ARS)[1] <- "Date"
  cas_NA_ARS <- cas_NA_ARS %>% 
    select(1:3) %>% 
    filter(!is.na(Date))
  dpt <- read.csv("data/departements-region.csv")
  
  Unaccent <- function(text) {
    text <- gsub("['`^~\"]", " ", text)
    text <- iconv(text, to="ASCII//TRANSLIT//IGNORE")
    text <- gsub("['`^~\"]", "", text)
    text <- gsub(" ", "", text)
    return(text)
  }
  
  
  dpt$dep_name <- Unaccent(dpt$dep_name)
  cas_NA_ARS_dpt <- merge(cas_NA_ARS, dpt, by.x="Dpt", by.y="num_dep")
  cas_NA_ARS_dpt <- cas_NA_ARS_dpt %>% 
    arrange(Date) %>% 
    group_by(Dpt) %>% 
    #mutate(CumNb = cumsum(Nombre)) %>% 
    mutate(Date = ymd(Date)) 
  
  rna_full <- list()
  all_days <- seq(from = min(unique(cas_NA_ARS_dpt$Date)), 
                  to = max(unique(cas_NA_ARS_dpt$Date)),
                  by=1)
  for (d in as.character(all_days)){
    rna_full[[d]] <- rna
    rna_full[[d]]$region
    rna_full[[d]]$Date <- ymd(d)
    rna_full[[d]] <- left_join(rna_full[[d]], cas_NA_ARS_dpt, by=c("region"="dep_name", "Date"))
    rna_full[[d]]$Nombre[is.na(rna_full[[d]]$Nombre)] <- 0
  }
  rna_full2plot <- do.call(rbind.data.frame, rna_full)
  rna_full2plot <- rna_full2plot %>% 
    arrange(Date) %>% 
    group_by(order) %>% 
    mutate(CumNb = cumsum(Nombre)) %>% 
    select(-subregion, -Dpt, -region_name)
  
  #rna_full2plot %>% filter(region == "Gironde")
  #View(rna_full2plot %>% filter(order == 9107))
  
  
  #temp <- full_join(rna, cas_NA_ARS_dpt, by=c("region"="dep_name"))
  p <- rna_full2plot %>% 
    #filter(Date == ymd("2020-03-12")) %>% 
    ggplot(aes(x = long, y = lat, group = group)) +
    geom_polygon(aes(fill=CumNb), colour = "black") +
    scale_fill_gradientn("Incidence cumulée\nd'infections à COVID19\nconfirmées",
                         colours=c("ivory", "lightskyblue1", "deepskyblue", "dodgerblue2", 
                                   "blue", "midnightblue")
                         , values=scales::rescale(c(0,2,10,20,30,40), to=c(0,1))
                         #, minor_breaks=c(5,20)
                         , breaks=c(0,10,20,30,40)
                         #, trans = "log1p"
                         , limits=c(0,40) 
    ) +
    coord_map() +
    theme_void() +
    labs(caption = "Source : ARS Nouvelle Aquitaine") +
    #ggtitle("Nouvelle Aquitaine") +
    #facet_wrap(~Date) +
    NULL
  p + ggtitle("Nouvelle Aquitaine") +
    facet_wrap(~Date) +
    NULL
  
  
  anim <- p + 
    ggtitle("Nouvelle Aquitaine", subtitle = '{closest_state}') +
    transition_states(Date, transition_length = 0, state_length = 1)
  animate(anim, fps=10, end_pause=15, nframes = 170)
  if(export_gif){
    anim_save(filename = paste0(xlsx_file, ".gif"))
  }
}