Update main.py

main.py

@@ -21,7 +21,6 @@ from langchain.chains import (
 
 import chainlit as cl
 from chainlit.input_widget import TextInput, Select, Switch, Slider
-from chainlit.types import ThreadDict
 
 from deep_translator import GoogleTranslator
 
@@ -144,42 +143,6 @@ async def Search(input, categorie):
             
     results = [sources_text, verbatim_text, sources_offres]
     return results
-    
-@cl.step(type="llm")
-async def setup_conversationalChain():
-    model = await LLModel()
-    retriever = await Retriever(cl.user_session.get("selectRequest"))
-    ########## Chain with streaming ##########
-    message_history = ChatMessageHistory()
-    memory = ConversationBufferMemory(memory_key="chat_history",output_key="answer",chat_memory=message_history,return_messages=True)
-    
-    qa = ConversationalRetrievalChain.from_llm(
-        model,
-        memory=cl.user_session.get("memory"),
-        chain_type="stuff",
-        return_source_documents=True,
-        verbose=False,
-        retriever=retriever
-    )
-    cl.user_session.set("runnable", qa)
-    cl.user_session.set("memory", memory)
-
-@cl.step(type="tool")
-async def switch(value):
-    if value == "Pédagogie durable":
-        return "bibliographie-OPP-DGDIN"
-    elif value == "Lieux d'apprentissage":
-        return "bibliographie-OPP-DGDIN"
-    elif value == "Journée de La Pédagogie":
-        return "year"
-    elif value == "Compétences du CFA Descartes":
-        return "skills"
-    elif value == "Formations Gustave Eiffel":
-        return "OF"
-    elif value == "Vidéos paroles de confiné.es":
-        return "videosTC"
-    elif value == "Offres d'emploi France Travail":
-        return "offreST"
         
 @cl.on_chat_start
 async def on_chat_start():
@@ -222,27 +185,23 @@ async def on_chat_start():
         await cl.Message(f"Vous pouvez requêter sur la thématique : {res.get('value')}").send()
         cl.user_session.set("selectRequest", res.get("name"))
         
-    await setup_conversationalChain()
+    model = await LLModel()
+    retriever = await Retriever(cl.user_session.get("selectRequest"))
+    ########## Chain with streaming ##########
+    message_history = ChatMessageHistory()
+    memory = ConversationBufferMemory(memory_key="chat_history",output_key="answer",chat_memory=message_history,return_messages=True)
     
-@cl.on_chat_resume
-async def on_chat_resume(thread: ThreadDict):
-    memory = ConversationBufferMemory(return_messages=True)
-    root_messages = [m for m in thread["steps"] if m["parentId"] == None]
-    for message in root_messages:
-        if message["type"] == "assistant_message" and message["output"][28:37] == "thématique":
-            resName = await switch(message["output"][42:])
-            cl.user_session.set("selectRequest", resName)
-            print(message["output"])
-        
-        if message["type"] == "user_message":
-            memory.chat_memory.add_user_message(message["output"])
-        else:
-            memory.chat_memory.add_ai_message(message["output"])
-
+    qa = ConversationalRetrievalChain.from_llm(
+        model,
+        memory=memory,
+        chain_type="stuff",
+        return_source_documents=True,
+        verbose=False,
+        retriever=retriever
+    )
+    cl.user_session.set("runnable", qa)
     cl.user_session.set("memory", memory)
-
-    await setup_conversationalChain()
-
+    
 @cl.on_message
 async def on_message(message: cl.Message):
     memory = cl.user_session.get("memory")