app.py

import os
import argparse
import asyncio
import gradio as gr
from difflib import Differ
from string import Template
from utils import load_prompt, setup_gemini_client
from configs.responses import SummaryResponses
from google.genai import types

def main(args):
    # style_css = open(args.css_path, "r").read()

    # global client, prompt_tmpl, system_instruction
    # client = None#setup_gemini_client(args)
    # prompt_tmpl = load_prompt(args)
    
    ## Gradio Blocks
    with gr.Blocks() as demo:
        # State per session
        state = gr.State({
            "messages": [],
            "attached_files": [],
            "summary": "",
            "summary_history": [],
            "summary_diff_history": []
        })

        with gr.Column():
            gr.Markdown("# Adaptive Summarization")
            gr.Markdown("AdaptSum stands for Adaptive Summarization. This project focuses on developing an LLM-powered system for dynamic summarization. Instead of generating entirely new summaries with each update, the system intelligently identifies and modifies only the necessary parts of the existing summary. This approach aims to create a more efficient and fluid summarization process within a continuous chat interaction with an LLM.")

    return demo

if __name__ == "__main__":
    # args = parse_args()
    demo = main(None)
    demo.launch()