backfill job handles url_names correctly

app/lib/fa_backfill_favs.rb

@@ -15,9 +15,10 @@ class FaBackfillFavs
   def initialize(start_at: nil, limit: nil, batch_size: nil)
     @start_at = start_at
     @limit = limit
-    @batch_size = T.let(batch_size || 100, Integer)
+    @batch_size = T.let(batch_size || 32, Integer)
     @unique_users_seen = T.let(Set.new, T::Set[String])
     @total_favs_seen = T.let(0, Integer)
+    @total_log_entries_processed = T.let(0, Integer)
   end
 
   sig { void }
@@ -30,16 +31,15 @@ class FaBackfillFavs
   sig { void }
   def collect_favs_from_logs
     scope =
-      HttpLogEntry.where(uri_host: "www.furaffinity.net").where(
-        "uri_path LIKE ?",
-        "/favorites/%",
-      )
-
-    total_processed = 0
+      HttpLogEntry
+        .includes(:response)
+        .where(uri_host: "www.furaffinity.net")
+        .where("uri_path LIKE ?", "/favorites/%")
 
     scope.find_in_batches(
       batch_size: @batch_size,
       start: @start_at,
+      order: :desc,
     ) do |entries|
       user_favs =
         T.let(
@@ -54,13 +54,25 @@ class FaBackfillFavs
         contents = T.let(response.contents, T.nilable(String))
         next unless contents
 
-        page = Domain::Fa::Parser::Page.new(contents)
+        page =
+          begin
+            Domain::Fa::Parser::Page.new(contents, require_logged_in: false)
+          rescue Domain::Fa::Parser::NotLoggedInError
+            next
+          end
         uri_path = T.let(entry.uri_path, T.nilable(String))
         next unless uri_path
 
-        url_name = uri_path.split("/").last
+        # url is like /favorites/iiszed/1183854116/next
+        # extract url_name from the path with regex
+        url_name = uri_path.match(%r{/favorites/([^\/]+)})&.[](1)
         next unless url_name
 
+        @total_log_entries_processed += 1
+        logger.info(
+          "[entry #{entry.id}] [processed #{@total_log_entries_processed}] favs for user #{url_name} - #{entry.uri_path}",
+        )
+
         favs = T.must(user_favs[url_name])
 
         page.submissions_parsed.each do |submission|
@@ -68,16 +80,18 @@ class FaBackfillFavs
 
           favs.add(submission.id)
         end
+
+        break if @limit && @total_log_entries_processed >= @limit
       end
 
       process_users_favs(user_favs)
 
-      total_processed += entries.length
-      logger.info("Processed #{total_processed} HttpLogEntries")
+      @total_log_entries_processed += entries.length
+      logger.info("Processed #{@total_log_entries_processed} HttpLogEntries")
       logger.info(
         "Unique users seen: #{@unique_users_seen.length}, Total favs seen: #{@total_favs_seen}",
       )
-      if @limit && total_processed >= @limit
+      if @limit && @total_log_entries_processed >= @limit
         logger.info("Reached limit of #{@limit} HttpLogEntries, stopping")
         break
       end
@@ -117,12 +131,12 @@ class FaBackfillFavs
     user_favs.each do |url_name, post_fa_ids|
       user_id =
         T.must_because(url_name_to_user_id[url_name]) do
-          "url_name #{url_name} not found in url_name_to_user_id"
+          "url_name '#{url_name.inspect}' not found in url_name_to_user_id"
         end
       post_fa_ids.each do |post_fa_id|
         post_id =
           T.must_because(fa_id_to_post_id[post_fa_id]) do
-            "post fa_id #{post_fa_id} not found in fa_id_to_post_id"
+            "post fa_id '#{post_fa_id.inspect}' not found in fa_id_to_post_id"
           end
         user_id_post_id_pairs << [user_id, post_id]
       end

spec/lib/fa_backfill_favs_spec.rb

@@ -42,7 +42,7 @@ describe FaBackfillFavs do
     # Create HttpLogEntry records for each user's favorites page
     HttpLogEntry.create!(
       uri_host: "www.furaffinity.net",
-      uri_path: "/favorites/iiszed",
+      uri_path: "/favorites/iiszed/1183854116/next",
       uri_scheme: "https",
       verb: :get,
       performed_by: "direct",