CV2-5087 move Articles side effecting saves to to it via presto

app/models/concerns/alegre_v2.rb

@@ -1,7 +1,7 @@
 require 'active_support/concern'
 class AlegreTimeoutError < StandardError; end
 class TemporaryProjectMedia
-  attr_accessor :team_id, :id, :url, :type
+  attr_accessor :team_id, :id, :url, :text, :type
   def media
     media_type_map = {
       "claim" => "Claim",
@@ -55,11 +55,18 @@ def sync_path_for_type(type)
     end
 
     def async_path(project_media)
-      "/similarity/async/#{get_type(project_media)}"
+      self.async_path_for_type(get_type(project_media))
+    end
+
+    def async_path_for_type(type)
+      "/similarity/async/#{type}"
     end
 
     def delete_path(project_media)
-      type = get_type(project_media)
+      self.delete_path_for_type(get_type(project_media))
+    end
+
+    def delete_path_for_type(type)
       "/#{type}/similarity/"
     end
 
@@ -122,6 +129,10 @@ def request(method, path, params, retries=3)
       end
     end
 
+    def request_delete_from_raw(params, type)
+      request("delete", delete_path_for_type(type), params)
+    end
+
     def request_delete(data, project_media)
       request("delete", delete_path(project_media), data)
     end
@@ -148,18 +159,22 @@ def get_type(project_media)
       type
     end
 
+    def content_hash_for_value(value)
+      Digest::MD5.hexdigest(value)
+    end
+
     def content_hash(project_media, field)
       if Bot::Alegre::ALL_TEXT_SIMILARITY_FIELDS.include?(field)
-        Digest::MD5.hexdigest(project_media.send(field))
+        content_hash_for_value(project_media.send(field))
       else
         if project_media.is_link?
-          return Digest::MD5.hexdigest(project_media.media.url)
+          return content_hash_for_value(project_media.media.url)
         elsif project_media.is_a?(TemporaryProjectMedia)
           return Rails.cache.read("url_sha:#{project_media.url}")
         elsif !project_media.is_text?
           return project_media.media.file.filename.split(".").first
         else
-          return Digest::MD5.hexdigest(project_media.send(field).to_s)
+          return content_hash_for_value(project_media.send(field).to_s)
         end
       end
     end
@@ -267,6 +282,14 @@ def store_package_text(project_media, field, params)
       generic_package_text(project_media, field, params)
     end
 
+    def get_sync_raw_params(params, type)
+      request("post", sync_path_for_type(type), params)
+    end
+
+    def get_async_raw_params(params, type)
+      request("post", async_path_for_type(type), params)
+    end
+
     def get_sync(project_media, field=nil, params={})
       request_sync(
         store_package(project_media, field, params),

app/models/explainer.rb

@@ -63,24 +63,26 @@ def self.update_paragraphs_in_alegre(id, previous_paragraphs_count, timestamp)
 
     # Index title
     params = {
+      content_hash: Bot::Alegre.content_hash_for_value(explainer.title),
       doc_id: Digest::MD5.hexdigest(['explainer', explainer.id, 'title'].join(':')),
+      context: base_context.merge({ field: 'title' }),
       text: explainer.title,
       models: ALEGRE_MODELS_AND_THRESHOLDS.keys,
-      context: base_context.merge({ field: 'title' })
     }
-    Bot::Alegre.request('post', '/text/similarity/', params)
+    Bot::Alegre.get_async_raw_params(params, "text")
 
     # Index paragraphs
     count = 0
     explainer.description.to_s.gsub(/\r\n?/, "\n").split(/\n+/).reject{ |paragraph| paragraph.strip.blank? }.each do |paragraph|
       count += 1
       params = {
+        content_hash: Bot::Alegre.content_hash_for_value(paragraph.strip),
         doc_id: Digest::MD5.hexdigest(['explainer', explainer.id, 'paragraph', count].join(':')),
+        context: base_context.merge({ paragraph: count }),
         text: paragraph.strip,
         models: ALEGRE_MODELS_AND_THRESHOLDS.keys,
-        context: base_context.merge({ paragraph: count })
       }
-      Bot::Alegre.request('post', '/text/similarity/', params)
+      Bot::Alegre.get_async_raw_params(params, "text")
     end
 
     # Remove paragraphs that don't exist anymore (we delete after updating in order to avoid race conditions)
@@ -91,7 +93,7 @@ def self.update_paragraphs_in_alegre(id, previous_paragraphs_count, timestamp)
         quiet: true,
         context: base_context.merge({ paragraph: count })
       }
-      Bot::Alegre.request('delete', '/text/similarity/', params)
+      Bot::Alegre.request_delete_from_raw(params, type)
     end
   end
 
@@ -106,7 +108,7 @@ def self.search_by_similarity(text, language, team_id)
         language: language
       }
     }
-    response = Bot::Alegre.request('post', '/text/similarity/search/', params)
+    Bot::Alegre.get_async_raw_params(params, "text")
     results = response['result'].to_a.sort_by{ |result| result['_score'] }
     explainer_ids = results.collect{ |result| result.dig('_source', 'context', 'explainer_id').to_i }.uniq.first(3)
     explainer_ids.empty? ? Explainer.none : Explainer.where(team_id: team_id, id: explainer_ids)