Scrape bird transect lines urls

scrape_urls.R

library("dplyr")
library("magrittr")
library("rvest")

urls <- c("http://koivu.luomus.fi/seurannat/linjalaskenta/vakiolinjat.php")

site <- html(urls)
linjat_table <- site %>% 
                  html_node("table") %>%
                  html_nodes("tr")

extract_urls <- function(x) {
  # Create an empty list to hold the results
  transect_data <- list()
  # Get all table data nodes from a single row
  td_nodes <- html_nodes(x, "td")
  # Extract the transect ID
  transect_data[["id"]] <- html_text(td_nodes[[1]])
  # Extract kansalaisen karttapaikka link
  transect_data[["map_url"]] <- html_attr(html_node(td_nodes[[13]], "a"), 
                                          "href")
  # Extract kartta pdf link
  transect_data[["map_pdf_url"]] <- html_attr(html_node(td_nodes[[15]], "a"), 
                                              "href")
  # Extract form pdf link
  transect_data[["form_pdf_url"]] <- html_attr(html_node(td_nodes[[16]], "a"), 
                                               "href")
  # Coerce the list to data frame
  return(dplyr::as_data_frame(transect_data))
}

# Skip the first row (it's the header)
transect_data <- lapply(linjat_table[2:length(linjat_table)], 
                        function(x) {return(extract_urls(x))})
# Create a data frame
transect_data <- dplyr::bind_rows(transect_data)