Opengraph

blogs/cararthompson.com.json

@@ -2,7 +2,7 @@
   "title": "Building Stories with Data", 
   "subtitle": "Pars pas les mains vides", 
   "type": "blog", 
-  "url": "https://cararthompson.com/blog", 
+  "url": "https://cararthompson.com/posts", 
   "rss_feed": "https://www.cararthompson.com/posts.xml",
   "photo_url": "https://www.cararthompson.com/images/home/profile-pic.png", 
   "description": "My website and portfolio, which includes talks and a blog about the things I've picked up along the way in the weird and wonderful worlds of #rstats, datascience and dataviz.",

scripts/get_og_data.R

@@ -0,0 +1,48 @@
+library(httr2)
+library(rvest)
+library(jsonlite)
+
+extract_opengraph_info <- function(url) {
+  # Fetch webpage content
+  response <- request(url) |>
+    req_perform()
+
+  # Parse the webpage content
+  content <- response |> resp_body_string()
+  html <- read_html(content)
+
+  # Extract Open Graph tags
+  og_tags <- html |>
+    html_nodes(xpath = '//meta[starts-with(@property, "og:")]')
+
+  # Extract the properties and their content
+  og_info <- lapply(og_tags, function(tag) {
+    html_attr(tag, "content")
+  })
+  names(og_info) <- sapply(og_tags, function(tag){
+    gsub("og:", "", html_attr(tag, "property"))
+  })
+  og_info
+}
+
+add_og_info <- function(path){
+  jsoncontent <- jsonlite::read_json(path)
+
+  if(jsoncontent$type != "youtube"){
+    jsoncontent$opengraph <- extract_opengraph_info(jsoncontent$url)
+    jsonlite::write_json(
+      jsoncontent, 
+      path, 
+      auto_unbox = TRUE,
+      pretty = TRUE
+    )
+  }
+}
+
+fileslist <- list.files(
+  here::here("blogs/"), 
+  pattern = "json$",
+  full.names = TRUE
+)
+
+lapply(fileslist, add_og_info)