better telegraaf.nl parser

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: paperboy
 Title: Comprehensive Collection of News Media Scrapers
 Version: 0.0.5.9000
-Date: 2023-11-08
+Date: 2023-11-10
 Authors@R:
     person(given = "Johannes B.",
            family = "Gruber",

R/deliver_telegraaf_nl.R

@@ -4,44 +4,53 @@ pb_deliver_paper.telegraaf_nl <- function(x, verbose = NULL, pb, ...) {
   # raw html is stored in column content_raw
   html <- rvest::read_html(x$content_raw)
 
+  data <- html %>%
+    rvest::html_elements("[data-name=\"PageTracking\"]") %>%
+    rvest::html_text2() %>%
+    jsonlite::fromJSON()
+
+  type <- purrr::pluck(data, "article", "type")
+  paywall <- purrr::pluck(data, "article", "premium")
+
   # datetime
-  datetime <- html %>%
-    rvest::html_element("[property=\"article:published_time\"]") %>%
-    rvest::html_attr("content") %>%
+  datetime <- purrr::pluck(data, "article", "publishDate") %>%
     lubridate::as_datetime()
 
   # headline
-  headline <- html %>%
-    rvest::html_element("[name=\"title\"]") %>%
-    rvest::html_attr("content")
+  headline <- purrr::pluck(data, "article", "title")
 
   # author
-  author <- html %>%
-    rvest::html_element(".DetailBylineBlock__author")  %>%
-    rvest::html_text2() %>%
-    toString()
+  author <- purrr::pluck(data, "article", "author", .default = NA_character_)
 
   # text
-  text <- html %>%
-    rvest::html_elements(".Article__intro,.DetailBodyBlocks p") %>%
-    rvest::html_text2() %>%
-    paste(collapse = "\n")
+  if (type == "normal") {
+    text <- html %>%
+      rvest::html_elements(".Article__intro,.DetailBodyBlocks p") %>%
+      rvest::html_text2() %>%
+      paste(collapse = "\n")
+  } else {
+    text <- paste0("[", type, "]")
+  }
 
   cover_image_html <- html %>%
     rvest::html_element(".DetailArticleImage img") %>%
     as.character()
 
   cover_image_url <- html %>%
     rvest::html_element(".DetailArticleImage img") %>%
-    rvest::html_attr("src") %>%
-    paste0("https://www.telegraaf.nl", .)
+    rvest::html_attr("src")
+
+  if (!is.na(cover_image_url))
+    cover_image_url <- paste0("https://www.telegraaf.nl", cover_image_url)
 
   # the helper function safely creates a named list from objects
   s_n_list(
     datetime,
     author,
     headline,
     text,
+    type,
+    paywall,
     cover_image_url,
     cover_image_html
   )