Spaces:
Sleeping
Sleeping
import asyncio | |
import websockets | |
import threading | |
import sqlite3 | |
import g4f | |
import streamlit as st | |
# Define the websocket client class | |
class WebSocketClient1: | |
def __init__(self, uri): | |
# Initialize the uri attribute | |
self.uri = uri | |
async def askQuestion(self, question): | |
system_instruction = "You are now integrated with a local websocket server in a project of hierarchical cooperative multi-agent framework called NeuralGPT. Your main job is to coordinate simultaneous work of multiple LLMs connected to you as clients. Each LLM has a model (API) specific ID to help you recognize different clients in a continuous chat thread (template: <NAME>-agent and/or <NAME>-client). Your chat memory module is integrated with a local SQL database with chat history. Your primary objective is to maintain the logical and chronological order while answering incoming messages and to send your answers to the correct clients to maintain synchronization of the question->answer logic. However, please note that you may choose to ignore or not respond to repeating inputs from specific clients as needed to prevent unnecessary traffic." | |
try: | |
db = sqlite3.connect('chat-hub.db') | |
cursor = db.cursor() | |
cursor.execute("SELECT * FROM messages ORDER BY timestamp DESC LIMIT 30") | |
messages = cursor.fetchall() | |
messages.reverse() | |
past_user_inputs = [] | |
generated_responses = [] | |
for message in messages: | |
if message[1] == 'server': | |
past_user_inputs.append(message[2]) | |
else: | |
generated_responses.append(message[2]) | |
response = await g4f.ChatCompletion.create_async( | |
model=g4f.models.gpt_4, | |
provider=g4f.Provider.Bing, | |
messages=[ | |
{"role": "system", "content": system_instruction}, | |
*[{"role": "user", "content": message} for message in past_user_inputs], | |
*[{"role": "assistant", "content": message} for message in generated_responses], | |
{"role": "user", "content": question} | |
]) | |
print(response) | |
return response | |
except Exception as e: | |
print(e) | |
# Define a function that will run the client in a separate thread | |
def run(self): | |
# Create a thread object | |
self.thread = threading.Thread(target=self.run_client) | |
# Start the thread | |
self.thread.start() | |
# Define a function that will run the client using asyncio | |
def run_client(self): | |
# Get the asyncio event loop | |
loop = asyncio.new_event_loop() | |
# Set the event loop as the current one | |
asyncio.set_event_loop(loop) | |
# Run the client until it is stopped | |
loop.run_until_complete(self.client()) | |
# Define a coroutine that will connect to the server and exchange messages | |
async def startClient(self): | |
# Connect to the server | |
async with websockets.connect(self.uri) as websocket: | |
# Loop forever | |
while True: | |
# Listen for messages from the server | |
input_message = await websocket.recv() | |
print(f"Server: {input_message}") | |
input_Msg = st.chat_message("assistant") | |
input_Msg.markdown(input_message) | |
try: | |
response = await self.askQuestion(input_message) | |
res1 = f"Client: {response}" | |
output_Msg = st.chat_message("ai") | |
output_Msg.markdown(res1) | |
await websocket.send(res1) | |
except websockets.ConnectionClosed: | |
print("client disconnected") | |
continue | |
except Exception as e: | |
print(f"Error: {e}") | |
continue |