Merge gamification with knowledge gap

toptal · Apr 12, 2024 · c569929 · c569929
1 parent b640338
commit c569929
Show file tree

Hide file tree

Showing 3 changed files with 16 additions and 18 deletions.
diff --git a/features/gamification.feature b/features/gamification.feature
diff --git a/features/knowledge_gaps.feature b/features/knowledge_gaps.feature
@@ -22,17 +22,27 @@
 # check that quiz_questions last row has a thread_id
 
 # Use white_check_mark emoji on TA message
-
 # check that user in user_scores table appears and has filled luminary_score eq 1
 
 Feature: Knowledge gaps
 
   Scenario: TopAssist initiates conversations for unanswered questions
-    Given there are no answers for question "What power superman has?"
+    Given databases contain documents with associated embeddings:
+      | title           | contents                                                  |
+      | Holiday Policy  | Full time employees are eligible for 30 days of per year. |
+      | Hardware Policy | Company hardware cannot be used for personal purposes.    |
+    When I submit phrase: "What power superman has?"
+    Then listener asks OpenAI for embeddings for the given phrase
+    And corresponding documents are not found in the database
+    And question "What power superman has?" is saved as unanswered
+    And I get rewarded 1 revealer point
     When I run CLI command "main.py knowledge-gap superman"
     Then listener asks OpenAI for embeddings for the given phrase
-    And corresponding unanswered questions are found in the database
+    And unanswered questions for given embeddings are found:
+      | question                 |
+      | What power superman has? |
     And thread "What power superman has?" is created in #topassist-dev-knowledge-gap channel
     When I reply in the thread with: "Flying"
-    And I react to that reply with :bookmark:
-    Then the conversation is being uploaded to dedicated confluence space
+    And I react to that reply with :white_check_mark:
+    Then conversation is being uploaded to dedicated confluence space
+    And I am awarded 1 luminary point
diff --git a/features/slack_conversation.feature b/features/slack_conversation.feature
@@ -10,14 +10,10 @@ Feature: Slack conversation
 # 3. Feedback. User posts a message to a thread started from the original question -> same as Q&A but instead of creating new conversation it posts to an exiting one.
 
   Scenario: User interacts with TopAssist in Slack
-    Given document database contains documents:
+    Given databases contain documents with associated embeddings:
       | title           | contents                                                  |
       | Holiday Policy  | Full time employees are eligible for 30 days of per year. |
       | Hardware Policy | Company hardware cannot be used for personal purposes.    |
-    And vector database contains entries:
-      | title           | embeddings     |
-      | Holiday Policy  | 32 56 f2 3b d7 |
-      | Hardware Policy | 4e d6 b1 78 9a |
     When I submit phrase: "Totally not a question"
     Then listener  does not react
     When I submit phrase: "How many days off am I eligible for?"