Confirmed cases of COVID-19, country comparison (https://public.flourish.studio/visualisation/1619458/embed)

covid19_country_comparison.R

library(jsonlite)
library(xml2)
library(rvest)
library(dplyr)
library(tidyr)
library(ggplot2)
library(plotly)

d <- xml2::read_html('https://public.flourish.studio/visualisation/1619458/embed')
js <- rvest::xml_nodes(d, 'script') %>% html_text
js <- js[grep('_Flourish_data_column_names', js)] %>% strsplit('\n') %>% .[[1]]
labels <- fromJSON(gsub('var _Flourish_data_column_names = |,$', '', 
                        js[grep('_Flourish_data_column_names =', js)][1]))

values <- fromJSON(gsub('_Flourish_data = |;$', '', 
                        js[grep('_Flourish_data =', js)][1]))$data
  
dat <- do.call(rbind, setNames(values$value, values$label)) %>% 
  as.data.frame(stringsAsFactors=FALSE) %>% 
  setNames(labels$data$value) %>% 
  dplyr::as.tbl() %>% 
  dplyr::mutate_all(as.numeric) %>% 
  dplyr::mutate(day=dplyr::row_number()) %>% 
  dplyr::select(day, everything()) %>% 
  tidyr::pivot_longer(-day, names_to='country', values_to='cases')


# cumulative cases since first infection (raw)
(ggplot(dat, aes(x=day, y=cases, color=country)) +
    geom_line() +
    xlab('Days since first case') +
    ylab('Total confirmed cases')) %>% 
  ggplotly

# cumulative cases since first infection (log10)
(ggplot(dat, aes(x=day, y=log10(cases), color=country)) +
    geom_line() +
    xlab('Days since first case') +
    ylab('log10(Total confirmed cases)')) %>% 
  ggplotly