Rake: add customized allmaps harvest task

This uses our geomg_id_s as the sort field for cursor mark

config/schedule.rb

@@ -57,7 +57,7 @@
 # Blacklight::Allmaps
 # Harvest Maps
 every :day, at: '3:30am', roles: [:app] do
-  rake 'blacklight_allmaps:sidecars:harvest:allmaps'
+  rake 'rake geoportal:allmaps:harvest'
 end
 
 # Populate the Georeferenced Facet

lib/tasks/geoportal.rake

@@ -92,4 +92,32 @@ namespace :geoportal do
       Blacklight.default_index.connection.commit
     end
   end
+
+  # @CUSTOMIZATION
+  # Require geomg_id_s as sort parameter
+  namespace :allmaps do
+    desc "Sidecars - Harvest: Crawl SolrDocuments to store Allmaps data locally"
+    task harvest: [:environment] do
+      cursor_mark = "*"
+      loop do
+        response = Blacklight.default_index.connection.get(
+          "select", params: {
+            q: "*:*", # all docs
+            fl: "id",  # just id field
+            cursorMark: cursor_mark, # use the cursor mark to handle paging
+            rows: 1000,
+            sort: "geomg_id_s asc" # must sort by id to use the cursor mark
+          }
+        )
+
+        response["response"]["docs"].each do |doc|
+          puts "Harvesting Allmaps data for #{doc["id"]}"
+          Blacklight::Allmaps::StoreSidecarAnnotation.perform_later(doc["id"])
+        end
+
+        break if response["nextCursorMark"] == cursor_mark # this means the result set is finished
+        cursor_mark = response["nextCursorMark"]
+      end
+    end
+  end
 end