app.py

import asyncio
import os
import gradio as gr
from core.central_ai_hub import CentralAIHub
from loguru import logger

# Initialize the Central AI Hub
hub = None

async def initialize_hub():
    global hub
    if hub is None:
        logger.info("Initializing Central AI Hub...")
        # Check if a local model path is provided as an environment variable
        model_path = os.getenv("LOCAL_MODEL_PATH")
        hub = CentralAIHub(model_path=model_path)
        await hub.start()
        logger.info("Central AI Hub initialized.")

async def process_task(task_type, task_content, task_requirements):
    await initialize_hub()
    task = {
        'type': task_type,
        'content': task_content,
        'requirements': task_requirements.split(',') if task_requirements else []
    }
    task_id = await hub.delegate_task(task)
    status = await hub.get_task_status(task_id)
    return f"Task ID: {task_id}, Status: {status['status']}, Result: {status.get('result', 'N/A')}"

if __name__ == "__main__":
    with gr.Blocks() as demo:
        gr.Markdown("# Central AI Hub")
        with gr.Row():
            task_type = gr.Dropdown(choices=['code_analysis', 'code_generation', 'error_fixing'], label="Task Type")
            task_content = gr.Textbox(label="Task Content")
            task_requirements = gr.Textbox(label="Task Requirements (comma separated)")
        submit_button = gr.Button("Submit Task")
        output_text = gr.Textbox(label="Task Status")

        submit_button.click(
            process_task,
            inputs=[task_type, task_content, task_requirements],
            outputs=[output_text]
        )

    demo.launch(server_name="0.0.0.0", server_port=7860)