FEATURE: allow limiting results in related topics section (#30)

Also:

- Normalizes behavior between logged in and anon,
 we only show related topics in the related topic section

- Renames "suggested" to "related" given this only exists in related section
- Adds a spec section to ensure anon does not regress
- Adds `ai_embeddings_semantic_related_topics` to limit related topics

Renamed settings:

ai_embeddings_semantic_suggested_model -> ai_embeddings_semantic_related_model
ai_embeddings_semantic_suggested_topics_enabled -> ai_embeddings_semantic_related_topics_enabled

Plugins is still in an experimental phase and not much is overidden hence
avoiding adding site setting migrations.


Co-authored-by: Krzysztof Kotlarek <kotlarek.krzysztof@gmail.com>

config/locales/server.en.yml

@@ -42,9 +42,10 @@ en:
     ai_embeddings_discourse_service_api_endpoint: "URL where the API is running for the embeddings module"
     ai_embeddings_discourse_service_api_key: "API key for the embeddings API"
     ai_embeddings_models: "Discourse will generate embeddings for each of the models enabled here"
-    ai_embeddings_semantic_suggested_model: "Model to use for suggested topics."
+    ai_embeddings_semantic_related_model: "Model to use for related topics."
     ai_embeddings_generate_for_pms: "Generate embeddings for personal messages."
-    ai_embeddings_semantic_suggested_topics_enabled: "Use Semantic Search for related topics."
+    ai_embeddings_semantic_related_topics_enabled: "Use Semantic Search for related topics."
+    ai_embeddings_semantic_related_topics: "Maximum number of topics to show in related topic section."
     ai_embeddings_pg_connection_string: "PostgreSQL connection string for the embeddings module. Needs pgvector extension enabled and a series of tables created. See docs for more info."
 
   reviewables:

config/settings.yml

@@ -124,7 +124,7 @@ plugins:
      - msmarco-distilbert-base-v4
      - msmarco-distilbert-base-tas-b
      - text-embedding-ada-002
-  ai_embeddings_semantic_suggested_model:
+  ai_embeddings_semantic_related_model:
     type: enum
     default: all-mpnet-base-v2
     choices:
@@ -134,5 +134,6 @@ plugins:
       - multi-qa-mpnet-base-dot-v1
       - paraphrase-multilingual-mpnet-base-v2
   ai_embeddings_generate_for_pms: false
-  ai_embeddings_semantic_suggested_topics_enabled: false
+  ai_embeddings_semantic_related_topics_enabled: false
+  ai_embeddings_semantic_related_topics: 5
   ai_embeddings_pg_connection_string: ""

lib/modules/embeddings/entry_point.rb

@@ -7,13 +7,12 @@ module DiscourseAi
         require_relative "models"
         require_relative "topic"
         require_relative "jobs/regular/generate_embeddings"
-        require_relative "semantic_suggested"
+        require_relative "semantic_related"
       end
 
       def inject_into(plugin)
         plugin.add_to_class(:topic_view, :related_topics) do
-          if !@guardian&.user || topic.private_message? ||
-               !SiteSetting.ai_embeddings_semantic_suggested_topics_enabled
+          if topic.private_message? || !SiteSetting.ai_embeddings_semantic_related_topics_enabled
             return nil
           end
 
@@ -21,7 +20,7 @@ module DiscourseAi
             TopicList.new(
               :suggested,
               nil,
-              DiscourseAi::Embeddings::SemanticSuggested.candidates_for(topic),
+              DiscourseAi::Embeddings::SemanticRelated.candidates_for(topic),
             ).topics
         end
 
@@ -35,7 +34,7 @@ module DiscourseAi
 
         %i[topic_view TopicViewPosts].each do |serializer|
           plugin.add_to_serializer(serializer, :related_topics) do
-            if object.next_page.nil? && !object.topic.private_message? && scope.authenticated?
+            if object.next_page.nil? && !object.topic.private_message?
               object.related_topics.map do |t|
                 SuggestedTopicSerializer.new(t, scope: scope, root: false)
               end
@@ -44,7 +43,7 @@ module DiscourseAi
 
           # custom include method so we also check on semantic search
           plugin.add_to_serializer(serializer, :include_related_topics?) do
-            plugin.enabled? && SiteSetting.ai_embeddings_semantic_suggested_topics_enabled
+            plugin.enabled? && SiteSetting.ai_embeddings_semantic_related_topics_enabled
           end
         end
 
@@ -57,11 +56,6 @@ module DiscourseAi
 
         plugin.on(:topic_created, &callback)
         plugin.on(:topic_edited, &callback)
-
-        DiscoursePluginRegistry.register_list_suggested_for_provider(
-          SemanticSuggested.method(:build_suggested_topics),
-          plugin,
-        )
       end
     end
   end

lib/modules/embeddings/semantic_related.rb

@@ -2,16 +2,10 @@
 
 module DiscourseAi
   module Embeddings
-    class SemanticSuggested
-      def self.build_suggested_topics(topic, pm_params, topic_query)
-        return unless SiteSetting.ai_embeddings_semantic_suggested_topics_enabled
-        return if topic_query.user
-        return if topic.private_message?
-
-        { result: candidates_for(topic), params: {} }
-      end
-
+    class SemanticRelated
       def self.candidates_for(topic)
+        return ::Topic.none if SiteSetting.ai_embeddings_semantic_related_topics < 1
+
         cache_for =
           case topic.created_at
           when 6.hour.ago..Time.now
@@ -30,7 +24,7 @@ module DiscourseAi
                 search_suggestions(topic)
               end
         rescue StandardError => e
-          Rails.logger.error("SemanticSuggested: #{e}")
+          Rails.logger.error("SemanticRelated: #{e}")
           Jobs.enqueue(:generate_embeddings, topic_id: topic.id)
           return ::Topic.none
         end
@@ -42,10 +36,11 @@ module DiscourseAi
           .secured
           .where(id: candidate_ids)
           .order("array_position(ARRAY#{candidate_ids}, id)")
+          .limit(SiteSetting.ai_embeddings_semantic_related_topics)
       end
 
       def self.search_suggestions(topic)
-        model_name = SiteSetting.ai_embeddings_semantic_suggested_model
+        model_name = SiteSetting.ai_embeddings_semantic_related_model
         model = DiscourseAi::Embeddings::Models.list.find { |m| m.name == model_name }
         function =
           DiscourseAi::Embeddings::Models::SEARCH_FUNCTION_TO_PG_FUNCTION[model.functions.first]

spec/lib/modules/embeddings/semantic_related_spec.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+require "rails_helper"
+
+describe DiscourseAi::Embeddings::SemanticRelated do
+  fab!(:target) { Fabricate(:topic) }
+  fab!(:normal_topic_1) { Fabricate(:topic) }
+  fab!(:normal_topic_2) { Fabricate(:topic) }
+  fab!(:normal_topic_3) { Fabricate(:topic) }
+  fab!(:unlisted_topic) { Fabricate(:topic, visible: false) }
+  fab!(:private_topic) { Fabricate(:private_message_topic) }
+  fab!(:secured_category) { Fabricate(:category, read_restricted: true) }
+  fab!(:secured_category_topic) { Fabricate(:topic, category: secured_category) }
+
+  before { SiteSetting.ai_embeddings_semantic_related_topics_enabled = true }
+
+  describe "#candidates_for" do
+    before do
+      Discourse.cache.clear
+      described_class.stubs(:search_suggestions).returns(
+        Topic.unscoped.order(id: :desc).limit(10).pluck(:id),
+      )
+    end
+
+    after { Discourse.cache.clear }
+
+    it "returns the related topics without non public topics" do
+      results = described_class.candidates_for(target).to_a
+      expect(results).to include(normal_topic_1)
+      expect(results).to include(normal_topic_2)
+      expect(results).to include(normal_topic_3)
+      expect(results).to_not include(unlisted_topic)
+      expect(results).to_not include(private_topic)
+      expect(results).to_not include(secured_category_topic)
+    end
+  end
+end

spec/lib/modules/embeddings/semantic_suggested_spec.rb

@@ -1,38 +0,0 @@
-# frozen_string_literal: true
-
-require "rails_helper"
-
-describe DiscourseAi::Embeddings::SemanticSuggested do
-  fab!(:target) { Fabricate(:topic) }
-  fab!(:normal_topic_1) { Fabricate(:topic) }
-  fab!(:normal_topic_2) { Fabricate(:topic) }
-  fab!(:normal_topic_3) { Fabricate(:topic) }
-  fab!(:unlisted_topic) { Fabricate(:topic, visible: false) }
-  fab!(:private_topic) { Fabricate(:private_message_topic) }
-  fab!(:secured_category) { Fabricate(:category, read_restricted: true) }
-  fab!(:secured_category_topic) { Fabricate(:topic, category: secured_category) }
-
-  before { SiteSetting.ai_embeddings_semantic_suggested_topics_enabled = true }
-
-  describe "#build_suggested_topics" do
-    before do
-      Discourse.cache.clear
-      described_class.stubs(:search_suggestions).returns(
-        Topic.unscoped.order(id: :desc).limit(10).pluck(:id),
-      )
-    end
-
-    after { Discourse.cache.clear }
-
-    it "returns the suggested topics without non public topics" do
-      suggested = described_class.build_suggested_topics(target, {}, TopicQuery.new(nil))
-      suggested_results = suggested[:result]
-      expect(suggested_results).to include(normal_topic_1)
-      expect(suggested_results).to include(normal_topic_2)
-      expect(suggested_results).to include(normal_topic_3)
-      expect(suggested_results).to_not include(unlisted_topic)
-      expect(suggested_results).to_not include(private_topic)
-      expect(suggested_results).to_not include(secured_category_topic)
-    end
-  end
-end

spec/requests/topic_spec.rb

@@ -6,25 +6,36 @@ describe ::TopicsController do
   fab!(:topic) { Fabricate(:topic) }
   fab!(:topic1) { Fabricate(:topic) }
   fab!(:topic2) { Fabricate(:topic) }
+  fab!(:topic3) { Fabricate(:topic) }
   fab!(:user) { Fabricate(:admin) }
 
   before do
     Discourse.cache.clear
-    SiteSetting.ai_embeddings_semantic_suggested_topics_enabled = true
+    SiteSetting.ai_embeddings_semantic_related_topics_enabled = true
+    SiteSetting.ai_embeddings_semantic_related_topics = 2
   end
 
   after { Discourse.cache.clear }
 
   context "when a user is logged on" do
     it "includes related topics in payload when configured" do
-      DiscourseAi::Embeddings::SemanticSuggested.stubs(:search_suggestions).returns([topic2.id])
+      DiscourseAi::Embeddings::SemanticRelated.stubs(:search_suggestions).returns(
+        [topic1.id, topic2.id, topic3.id],
+      )
+
+      get("#{topic.relative_url}.json")
+      json = response.parsed_body
+
+      expect(json["suggested_topics"].length).to eq(0)
+      expect(json["related_topics"].length).to eq(2)
+
       sign_in(user)
 
       get("#{topic.relative_url}.json")
       json = response.parsed_body
 
       expect(json["suggested_topics"].length).to eq(0)
-      expect(json["related_topics"].length).to be > 0
+      expect(json["related_topics"].length).to eq(2)
     end
   end
 end