try and except for solving 'while a run run is active'

app.py

@@ -1,6 +1,7 @@
 import os
 import re
 import streamlit as st
+import openai
 from dotenv import load_dotenv
 from langchain.agents.openai_assistant import OpenAIAssistantRunnable
 
@@ -20,33 +21,57 @@ def remove_citation(text: str) -> str:
     pattern = r"【\d+†\w+】"
     return re.sub(pattern, "📚", text)
 
-# Initialize session state for messages, thread_id and processing state
+# Initialize session state for messages and thread_id
 if "messages" not in st.session_state:
     st.session_state["messages"] = []
 if "thread_id" not in st.session_state:
     st.session_state["thread_id"] = None
-if "is_processing" not in st.session_state:
-    st.session_state["is_processing"] = False
 
 st.title("Solution Specifier A")
 
 def predict(user_input: str) -> str:
     """
     This function calls our OpenAIAssistantRunnable to get a response.
-    If we don't have a thread_id yet, we create a new thread on the first call.
-    Otherwise, we continue the existing thread.
+    We either create a new thread (no thread_id yet) or continue the existing thread.
+    If the server complains that a run is still active, we reset the thread_id and retry once.
     """
-    if st.session_state["thread_id"] is None:
-        response = extractor_llm.invoke({"content": user_input})
-        st.session_state["thread_id"] = response.thread_id
-    else:
-        response = extractor_llm.invoke(
-            {"content": user_input, "thread_id": st.session_state["thread_id"]}
-        )
-    output = response.return_values["output"]
-    return remove_citation(output)
+    try:
+        if st.session_state["thread_id"] is None:
+            # Start a new thread
+            response = extractor_llm.invoke({"content": user_input})
+            st.session_state["thread_id"] = response.thread_id
+        else:
+            # Continue the existing thread
+            response = extractor_llm.invoke(
+                {"content": user_input, "thread_id": st.session_state["thread_id"]}
+            )
+        output = response.return_values["output"]
+        return remove_citation(output)
+
+    except openai.error.BadRequestError as e:
+        # If the error says a run is still active, reset to a new thread and re-invoke once
+        err_msg = str(e)
+        if "while a run" in err_msg:
+            st.session_state["thread_id"] = None
+            # Re-invoke once to get a fresh thread
+            try:
+                response = extractor_llm.invoke({"content": user_input})
+                st.session_state["thread_id"] = response.thread_id
+                output = response.return_values["output"]
+                return remove_citation(output)
+            except Exception as e2:
+                st.error(f"Error after resetting thread: {e2}")
+                return ""
+        else:
+            # Some other bad request
+            st.error(err_msg)
+            return ""
+    except Exception as e:
+        # Catch-all for any other error
+        st.error(str(e))
+        return ""
 
-# Display any existing messages (from a previous run or refresh)
+# Display any existing messages
 for msg in st.session_state["messages"]:
     if msg["role"] == "user":
         with st.chat_message("user"):
@@ -56,35 +81,23 @@ for msg in st.session_state["messages"]:
             st.write(msg["content"])
 
 # Create the chat input widget at the bottom of the page
-# Disable it while processing a message
-user_input = st.chat_input(
-    "Type your message here...",
-    disabled=st.session_state["is_processing"]
-)
+user_input = st.chat_input("Type your message here...")
 
 # When the user hits ENTER on st.chat_input
-if user_input and not st.session_state["is_processing"]:
-    try:
-        # Set processing state to True
-        st.session_state["is_processing"] = True
-        
-        # Add the user message to session state
-        st.session_state["messages"].append({"role": "user", "content": user_input})
-        
-        # Display the user's message
-        with st.chat_message("user"):
-            st.write(user_input)
-            
-        # Get the assistant's response
-        response_text = predict(user_input)
-        
-        # Add the assistant response to session state
-        st.session_state["messages"].append({"role": "assistant", "content": response_text})
-        
-        # Display the assistant's reply
-        with st.chat_message("assistant"):
-            st.write(response_text)
-            
-    finally:
-        # Reset processing state when done
-        st.session_state["is_processing"] = False
+if user_input:
+    # Add the user message to session state
+    st.session_state["messages"].append({"role": "user", "content": user_input})
+
+    # Display the user's message
+    with st.chat_message("user"):
+        st.write(user_input)
+
+    # Get the assistant's response
+    response_text = predict(user_input)
+
+    # Add the assistant response to session state
+    st.session_state["messages"].append({"role": "assistant", "content": response_text})
+
+    # Display the assistant's reply
+    with st.chat_message("assistant"):
+        st.write(response_text)