updated docs

Future_Improvements.md

@@ -147,14 +147,19 @@ This improvement would significantly enhance the chatbot's usability by making i
    - Added timestamps and metadata to feedback records
    - Organized feedback in dedicated feedback directory
 
-2. **Add Flexible Storage System** ✅
+2. **Implement Supabase Database Storage** ✅
+   - Feedback is now stored in a Supabase database for persistent, structured storage and analytics.
+   - **Note:** Due to unresolved issues with Supabase Row Level Security (RLS) and the anon key, the application currently uses the Supabase service (admin) key to insert feedback, bypassing RLS. This is a workaround and should be reviewed with Supabase for a more secure, production-ready solution.
+   - All feedback fields are now saved and can be reviewed in the Supabase dashboard.
+
+3. **Add Flexible Storage System** ✅
    - Created feedback manager with save_feedback interface
    - Implemented custom save_feedback_with_prefix function for better organization
    - Added logging for feedback events
    - Ensured proper error handling and success tracking
 
-3. **Next Steps**
-   - Implement Supabase integration for persistent feedback storage on Streamlit Cloud
+4. **Next Steps**
+   - If Supabase resolves the anon key/RLS issue, update the app to use the anon key and a secure RLS policy for public feedback.
    - Consider other database options: MongoDB Atlas (free tier) for JSON-like storage
    - Add indexing capability for feedback analytics
    - Create migration utility to move file-based feedback to database

README.md

@@ -74,4 +74,14 @@ If you experience issues with imports or module reloading:
    streamlit run Public_Chat.py --server.runOnSave=true
    ```
 
-See [DEPLOYMENT.md](DEPLOYMENT.md) for cloud deployment options.
+See [DEPLOYMENT.md](DEPLOYMENT.md) for cloud deployment options.
+
+## Feedback System (Updated)
+
+- User feedback is now stored in a Supabase database for persistent, structured storage and future analytics.
+- Due to issues with Supabase Row Level Security (RLS) and the anon key, the application currently uses the Supabase service (admin) key to insert feedback. This bypasses RLS and allows all feedback to be saved securely from the backend.
+- **Note:** Using the service key is a workaround. For production, consult Supabase support or documentation to implement a more secure, best-practice solution for public/anon feedback collection.
+
+## Progress
+- Feedback database integration with Supabase is complete and operational.
+- All feedback fields are now saved and can be reviewed in the Supabase dashboard.

architecture.md

@@ -82,60 +82,16 @@ The application follows a multi-page Streamlit structure with clear separation o
 
 ## Feedback System
 
-The application includes a comprehensive feedback collection system:
-
-### Components
-
-1. **UI Components**:
-   - Thumbs up/down buttons after each chatbot response
-   - Detailed feedback form for negative ratings
-   - Categories for feedback classification
-   - Optional free-text input for detailed comments
-   - Thank you messages after feedback submission
-
-2. **Feedback Storage**:
-   - JSON files with rating prefix in filenames (pos_/neg_)
-   - Timestamped feedback records
-   - Storage in dedicated feedback directory
-   - File-based persistence for simplicity
-
-3. **Implementation Details**:
-   - Unique message IDs based on content hash
-   - Feedback state tracking in session state
-   - Independent feedback for each assistant message
-   - Proper state management across page refreshes
-
-### Feedback Flow
-
-1. User receives a response from the chatbot
-2. Feedback buttons appear below the response
-3. User clicks thumbs up or thumbs down
-4. For positive feedback:
-   - Feedback is immediately saved
-   - Thank you message is displayed
-5. For negative feedback:
-   - Detailed feedback form appears
-   - User selects a category and optional comments
-   - On submission, feedback is saved
-   - Thank you message is displayed
-
-### Technical Implementation
-
-1. **Message Identification**:
-   - Each message has a unique ID based on its content hash
-   - This ensures consistent identification across sessions
-
-2. **Feedback Storage**:
-   - Uses `save_feedback_with_prefix()` function
-   - Creates files with naming pattern: `{pos|neg}_feedback_{timestamp}.json`
-   - Stores complete feedback data including query, response, rating, and comments
-
-3. **State Management**:
-   - Tracks feedback status in `st.session_state.feedback`
-   - Maintains proper UI state for each message
-   - Ensures independent feedback for each response
-
-This system allows for continuous improvement of the chatbot based on user feedback, providing valuable insights into response quality and areas for enhancement.
+The application now uses Supabase as the backend for storing user feedback:
+
+- All feedback is saved in a Supabase database table for persistent, structured storage and future analytics.
+- Due to unresolved issues with Supabase Row Level Security (RLS) and the anon key, the application currently uses the Supabase service (admin) key to insert feedback. This bypasses RLS and allows all feedback to be saved securely from the backend.
+- **Security Note:** Using the service key is a workaround. For production, consult Supabase support or documentation to implement a more secure, best-practice solution for public/anon feedback collection. Never expose the service key in frontend or public code.
+- All feedback fields are now saved and can be reviewed in the Supabase dashboard.
+
+#### Progress
+- Feedback database integration with Supabase is complete and operational.
+- The feedback manager now supports saving feedback directly to the database.
 
 ## Document Indexing System
 

feedback/feedback_advanced.csv

@@ -0,0 +1,4 @@
+"id","user_id","query","normalized_query","question_hash","response","rating","category","comment","tags","status","admin_note","assigned_to","document_id","source","priority","reviewed","timestamp"
+"b1a7c1e2-1234-4a5b-8c9d-1e2f3a4b5c6d","alice","What is a habit?","what is a habit","a1b2c3d4","A habit is a routine behavior repeated regularly.","positive","","","","open","","","doc_001","user","low","FALSE","2024-06-01T12:00:00Z"
+"c2b8d2f3-2345-5b6c-9d0e-2f3a4b5c6d7e","bob","How do I form a new habit?","how do i form a new habit","b2c3d4e5","Start small, be consistent, and track your progress.","negative","Incomplete answer","Did not mention rewards or triggers.","habit formation,incomplete","in progress","Will add more detail to doc_002","admin1","doc_002","user","high","TRUE","2024-06-01T12:05:00Z"
+"d3c9e3f4-3456-6c7d-0e1f-3a4b5c6d7e8f","carol","What is a habit?","what is a habit","a1b2c3d4","A habit is a routine behavior repeated regularly.","negative","Irrelevant to question","Answer was off-topic.","habit formation,off-topic","open","","","doc_001","user","medium","FALSE","2024-06-01T12:10:00Z"

supabase_test.py

@@ -0,0 +1,9 @@
+from supabase import create_client
+
+SUPABASE_URL = "https://cxptnhxhrhlwiuzdcyxl.supabase.co"  # Replace with your new project URL
+SUPABASE_KEY = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJzdXBhYmFzZSIsInJlZiI6ImN4cHRuaHhocmhsd2l1emRjeXhsIiwicm9sZSI6InNlcnZpY2Vfcm9sZSIsImlhdCI6MTc0NDg2NDc0NiwiZXhwIjoyMDYwNDQwNzQ2fQ.22iQvvjU94vD2asqny7D5_18T4drmx7Q1dVZFQroDs4"  # Replace with your new anon key
+
+supabase = create_client(SUPABASE_URL, SUPABASE_KEY)
+data = {"query": "test", "response": "test", "rating": "positive"}
+result = supabase.table("feedback_test").insert(data).execute()
+print(result)

utils/feedback.py

@@ -6,6 +6,20 @@ from datetime import datetime
 # Setup logging
 logger = logging.getLogger(__name__)
 
+# Supabase integration
+try:
+    from supabase import create_client, Client
+    SUPABASE_URL = os.getenv("SUPABASE_URL")
+    SUPABASE_KEY = os.getenv("SUPABASE_KEY")
+    if SUPABASE_URL and SUPABASE_KEY:
+        supabase: Client = create_client(SUPABASE_URL, SUPABASE_KEY)
+    else:
+        supabase = None
+        logger.warning("Supabase URL or Key not set. Supabase feedback saving will be disabled.")
+except ImportError:
+    supabase = None
+    logger.warning("supabase-py not installed. Supabase feedback saving will be disabled.")
+
 class FeedbackManager:
     """Handles the storage and retrieval of user feedback for chatbot responses."""
     
@@ -61,5 +75,25 @@ class FeedbackManager:
             logger.error(f"Error retrieving feedback: {str(e)}")
             return []
 
+    def save_feedback_supabase(self, feedback_data):
+        """
+        Save feedback to Supabase table 'feedback'.
+        Args:
+            feedback_data (dict): Dictionary containing feedback information matching the table schema.
+        Returns:
+            bool: True if feedback was saved successfully, False otherwise
+        """
+        if not supabase:
+            logger.error("Supabase client not available. Feedback not saved to Supabase.")
+            return False
+        try:
+            logger.info(f"Attempting to insert feedback: {feedback_data}")  # Debug outgoing data
+            data, count = supabase.table('feedback').insert(feedback_data).execute()
+            logger.info(f"Feedback saved to Supabase: {data}")
+            return True
+        except Exception as e:
+            logger.error(f"Error saving feedback to Supabase: {str(e)}")
+            return False
+
 # Create a singleton instance
 feedback_manager = FeedbackManager()