Download mpox data for Africa from WHO Shiny App

mpox_shiny_africa.R

library(httr)
library(rvest)
library(base64enc)
library(readr)

get_shiny_data <- function() {
  
  url <- "https://worldhealthorg.shinyapps.io/mpx_global/#26_Case_definitions" # URL of the webpage
  
 
  response <- GET(url)  # Fetch the page content
  
  
  if (status_code(response) == 200) { # Check if the request was successful
    
    page <- content(response, "text") # Parse the HTML content with rvest
    soup <- read_html(page)
    
    
    button <- html_node(soup, xpath = "//button[contains(@class, 'btn-primary')]") # Find the button tag with the specific class might break if WHO changes arrangement of their page
    
   
    if (!is.null(button)) {  # Extract the value of the 'onclick' attribute
      onclick_value <- html_attr(button, "onclick")
      
      
      if (grepl("data:text/csv;base64,", onclick_value)) { # If the attribute contains 'data:text/csv;base64,', extract and decode it
        base64_data <- strsplit(onclick_value, "data:text/csv;base64,")[[1]][2]
        base64_data <- gsub("[')]", "", base64_data)
        
        
        decoded_csv <- rawToChar(base64decode(base64_data)) # Decode base64 and clean it
        
        
        cleaned_csv <- iconv(decoded_csv, "latin1", "ASCII", sub = "") # Clean any problematic characters, such as null or non-printable characters
        
        
        csv_data <- read_csv(I(cleaned_csv), skip_empty_rows = TRUE) # Read cleaned CSV into a dataframe
        
        return(csv_data)
      }
    }
  } else {
    print(paste("Failed to retrieve the webpage. Status code:", status_code(response)))
  }
  
  return(NULL)
}


df <- get_shiny_data() # Run the function to get the data
print(df)

df %>% 
  filter(week_end_date > as.Date('2023-12-31')) %>% 
ggplot(., aes(x = week_end_date, y = total_suspected_cases, group = country))+
  geom_point()+
  facet_wrap(~country, scales = 'free')