cyclop/openai_generic_telegram_inference_bot.py

import json
import os
import importlib
import inspect
import logging
import asyncio
from telegram import error as TelegramErrors, Update, __version__ as telegram_version, InlineKeyboardButton, InlineKeyboardMarkup
from telegram.ext import Application, CommandHandler, MessageHandler, filters, ContextTypes, CallbackQueryHandler
from dotenv import load_dotenv
from tools.base_tool import BaseTool
from tools.metrics_tool import MetricsTool
from openai import OpenAI

# Load environment variables
load_dotenv()

openai_client = OpenAI(
    api_key=os.environ.get("GLHF_CHAT_API_KEY"), 
    base_url=os.environ.get("GLHF_CHAT_API_BASE", "https://api.openai.com/v1")
)

# Set up logging to console and file
logging.basicConfig(level=logging.WARNING, handlers=[
    logging.StreamHandler(),
    logging.FileHandler('logs/output.log', mode='a')
])

# Set up Telegram bot
TELEGRAM_BOT_TOKEN = os.getenv('TELEGRAM_BOT_TOKEN')

# Load system prompt
with open("prompts/developer_prompt.txt", "r") as file:
    system_prompt = file.read().strip()

# Dictionary to store conversation history for each user
conversation_history = {}

# Dictionary to store processing status for each user
processing_status = {}

# Load tools
tools = [MetricsTool()]  # Add MetricsTool instance
tools_dir = os.path.join(os.path.dirname(__file__), 'tools')
for filename in os.listdir(tools_dir):
    if filename.endswith('.py') and filename not in ['__init__.py', 'base_tool.py', 'metrics_tool.py']:
        module_name = f'tools.{filename[:-3]}'
        module = importlib.import_module(module_name)
        for name, obj in inspect.getmembers(module):
            if inspect.isclass(obj) and issubclass(obj, BaseTool) and obj != BaseTool:
                tools.append(obj())

# Collect all function definitions
functions = []
for tool in tools:
    functions.extend(tool.get_functions())

async def start(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
    logging.info("Bot started")
    await update.message.reply_text(
        "Hello! I'm your AI assistant. How can I help you today?"
    )

async def clear(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
    user_id = update.effective_user.id
    if user_id in conversation_history:
        del conversation_history[user_id]
    for tool in tools:
        tool.clear()

    logging.info(f"Cleared conversation history for user {user_id}")
    await update.message.reply_text("Conversation history cleared. Let's start fresh!")

async def update_status_message(context: ContextTypes.DEFAULT_TYPE, chat_id: int, message_id: int, status: str):
    keyboard = [
        [InlineKeyboardButton("Abort", callback_data='abort')]
    ]
    reply_markup = InlineKeyboardMarkup(keyboard)
    await context.bot.edit_message_text(
        chat_id=chat_id,
        message_id=message_id,
        text=f"Current status: {status}",
        reply_markup=reply_markup
    )

async def handle_message(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
    try:
        user_id = update.effective_user.id
        user_message = update.message.text

        logging.info(f"Message from user {user_id}: {user_message}")

        if user_id not in conversation_history:
            conversation_history[user_id] = []

        conversation_history[user_id].append({"role": "user", "content": user_message})

        # Send initial status message
        status_message = await update.message.reply_text("Processing your request...", reply_markup=InlineKeyboardMarkup([[InlineKeyboardButton("Abort", callback_data='abort')]]))
        processing_status[user_id] = {"processing": True, "message_id": status_message.message_id}

        messages = conversation_history[user_id]

        response = get_chat_response(messages)
        tool_calls = []
        for choice in response.choices:
            if choice.finish_reason == "tool_calls":
                tool_calls.append(choice)
            else:
                assistant_message = choice.message.content
                messages.append({"role": "assistant", "content": assistant_message})

        toolUseCount = 0
        previous_function_name = ""

        while tool_calls and toolUseCount < 50 and processing_status[user_id]["processing"]:
            tool_use_results = []
            for tool_call in tool_calls:
                function_name = tool_call['function']['name']

                if previous_function_name != function_name:
                    await update_status_message(context, update.effective_chat.id, status_message.message_id, f"Using tool: {function_name}")
                    previous_function_name = function_name

                tool_response = call_tool(tool_call['function'])
                tool_use_results.append({"tool_call_id": tool_call['id'], "role": "tool", "name": function_name, "content": json.dumps(tool_response)})

            messages.extend(tool_use_results)

            response = get_chat_response(messages)
            tool_calls = response.get('tool_calls', [])
            assistant_message = response.get('content', '')
            messages.append({"role": "assistant", "content": assistant_message})

            toolUseCount += 1

        if toolUseCount == 0:
            conversation_history[user_id].append({"role": "assistant", "content": assistant_message})
        
        if len(conversation_history[user_id]) > 20:
            conversation_history[user_id] = conversation_history[user_id][-20:]

        # Remove the status message
        await context.bot.delete_message(chat_id=update.effective_chat.id, message_id=status_message.message_id)
        del processing_status[user_id]
        try:
            await update.message.reply_text(assistant_message)
        except TelegramErrors.BadRequest as e:
            logging.error(f"An error occurred when trying to send a message in telegram: {str(e)}")

    except Exception as e:
        logging.error(f"An error occurred: {str(e)}")
        await update.message.reply_text("Sorry, an error occurred while processing your request.")

def call_tool(function_call):
    function_name = function_call['name']
    function_args = json.loads(function_call['arguments'])
    for tool in tools:
        if function_name in [f["name"] for f in tool.get_functions()]:
            return tool.execute(function_name, **function_args)

def get_chat_response(messages):
    try:
        response = openai_client.chat.completions.create(
            model="hf:meta-llama/Meta-Llama-3.1-405B-Instruct",  # You can change this to your desired model
            messages=[{"role": "system", "content": system_prompt}] + messages,
            max_tokens=4096,
            temperature=0.6,
            top_p=1,
            frequency_penalty=0,
            presence_penalty=0,
            tools=functions
            #tool_choice={  }
        )
        return response
    except Exception as e:
        logging.error(f"An error occurred: {str(e)}")
        return None

async def status(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
    await update.message.reply_text("Currently using gpt-3.5-turbo")

async def abort_processing(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
    query = update.callback_query
    await query.answer()
    
    user_id = query.from_user.id
    if user_id in processing_status:
        processing_status[user_id]["processing"] = False
        await context.bot.edit_message_text(
            chat_id=query.message.chat_id,
            message_id=query.message.message_id,
            text="Processing aborted."
        )
        await clear(update, context)
    else:
        await query.edit_message_text(text="No active processing to abort.")

def main() -> None:
    # Create the Application and pass it your bot's token
    application = Application.builder().token(TELEGRAM_BOT_TOKEN).build()

    # Add handlers
    application.add_handler(CommandHandler("start", start))
    application.add_handler(CommandHandler("clear", clear))
    application.add_handler(CommandHandler("status", status))
    application.add_handler(MessageHandler(filters.TEXT & ~filters.COMMAND, handle_message))
    application.add_handler(CallbackQueryHandler(abort_processing, pattern='^abort$'))

    # Start the Bot
    logging.info("Bot is running...")
    application.run_polling()

if __name__ == '__main__':
    main()