Update app.py

app.py

@@ -233,10 +233,11 @@ def generate_chunked_response(model, prompt, max_tokens=1000, max_chunks=5):
             full_response += chunk
         except Exception as e:
             print(f"Error in generate_chunked_response: {e}")
+            print(f"Prompt: {prompt}")
+            print(f"Full response so far: {full_response}")
             if "Input validation error" in str(e):
-                # If we hit the token limit, return what we have so far
                 return full_response if full_response else "The input was too long to process. Please try a shorter query."
-            break
+            raise  # Re-raise the exception to be caught in ask_question
     return full_response.strip()
 
 def extract_text_from_webpage(html):
@@ -515,6 +516,11 @@ def extract_answer(full_response, instructions=None):
     full_response = '\n'.join(lines)
 
     return full_response.strip()
+        except Exception as e:
+        print(f"Error in extract_answer: {e}")
+        print(f"Full response: {full_response}")
+        print(f"Instructions: {instructions}")
+        raise  # Re-raise the exception to be caught in ask_question
 
 # Gradio interface
 with gr.Blocks() as demo: