Code for converting the tables in the ESR measles outbreak weekly report pdfs into excel files

ESR_measles_report_pdfs_to_excel

library(rvest)
library(janitor)
library(pdftools)
library(dplyr)
library(tabulizer) #needs Jav, see https://github.com/ropensci/tabulizer
library(writexl)
library(purrr)

## store pdfs locally in  folder ESR_measles
if(!dir.exists("ESR_measles")){dir.create("ESR_measles")}

start_URL <- "https://surv.esr.cri.nz/surveillance/WeeklyMeaslesRpt.php"
start_page <- read_html(start_URL)
all_start_URLS <- start_page %>% html_nodes("a") %>% html_attr("href")
all_start_text <- start_page %>% html_nodes("a") %>% html_text()

stage2_urls <- paste0("https://surv.esr.cri.nz",
                      all_start_URLS[grep("Measles Report 2019",
                                          all_start_text)])

get_second_links <- function(x){
  every_urls <- read_html(x) %>% html_nodes("a") %>% html_attr("href")
  base_url <- unique(grep("/PDF_surveillance/MeaslesRpt/2019/", every_urls, value=TRUE))
  pdf_url <- paste0("https://surv.esr.cri.nz", base_url)
  return(pdf_url)
}
second_link <- sapply(stage2_urls, get_second_links)
save_path <- paste0("ESR_measles/",
                     basename(second_link))

get_reports <- function(x,y){
  if(!file.exists(y)){
    download.file(url = x, destfile = y, method = "libcurl", mode = "wb")
    return("new")
  }
  return("current")
}
  
downloads <- mapply(get_reports, x=second_link, y=save_path)
###
data_from_pdf <- function(x){
  PDFextracted <-  extract_tables(x) #tabulizer
  # 1 character matrix per table, want dataframes for writexl
  PDFtables <- lapply(PDFextracted,as.data.frame, stringsAsFactors=FALSE)
  target_no_suffix <- gsub("\\.pdf$","", x, ignore.case = TRUE)
  write_xlsx(PDFtables, paste0(target_no_suffix, ".xlsx"), col_names=FALSE) #writexl
}

PDFtext <- walk(save_path, data_from_pdf)