Update app.py

app.py

@@ -708,13 +708,16 @@ with gr.Blocks(
             if model is None:
                 return history + [[message, "❌ Failed to initialize Gemini model. Please check your GEMINI_API_KEY."]]
             
-            # Use RAG to get relevant chunks from markdown
+            # Use RAG to get relevant chunks from markdown (smaller chunks for quota)
             if document_chunks and len(document_chunks) > 0:
-                relevant_chunks = retrieve_relevant_chunks(message, document_chunks, document_embeddings)
+                relevant_chunks = retrieve_relevant_chunks(message, document_chunks, document_embeddings, top_k=2)
                 context = "\n\n".join(relevant_chunks)
+                # Limit context size to avoid quota issues
+                if len(context) > 1000:
+                    context = context[:1000] + "..."
             else:
                 # Fallback to truncated document if RAG fails
-                context = processed_markdown[:2000] + "..." if len(processed_markdown) > 2000 else processed_markdown
+                context = processed_markdown[:800] + "..." if len(processed_markdown) > 800 else processed_markdown
             
             # Create prompt for Gemini
             prompt = f"""You are a helpful assistant that answers questions about documents. Use the provided context to answer questions accurately and concisely.
@@ -726,12 +729,26 @@ Question: {message}
 
 Please provide a clear and helpful answer based on the context provided."""
             
-            # Generate response using Gemini API
-            response = model.generate_content(prompt)
+            # Generate response using Gemini API with retry logic
+            import time
+            max_retries = 2
             
-            response_text = response.text if hasattr(response, 'text') else str(response)
-            
-            return history + [[message, response_text]]
+            for attempt in range(max_retries):
+                try:
+                    response = model.generate_content(prompt)
+                    response_text = response.text if hasattr(response, 'text') else str(response)
+                    return history + [[message, response_text]]
+                except Exception as api_error:
+                    if "429" in str(api_error) and attempt < max_retries - 1:
+                        # Rate limit hit, wait and retry
+                        time.sleep(3)
+                        continue
+                    else:
+                        # Other error or final attempt failed
+                        if "429" in str(api_error):
+                            return history + [[message, "❌ API quota exceeded. Please wait a moment and try again, or check your Gemini API billing."]]
+                        else:
+                            raise api_error
             
         except Exception as e:
             error_msg = f"❌ Error generating response: {str(e)}"