app.py

import asyncio
import streamlit as st
from pydantic_ai import Agent
from pydantic_ai.models.groq import GroqModel
import nest_asyncio
import pdfplumber
import os

api_key = os.getenv("API_KEY")
data = []

#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

model = GroqModel('llama-3.1-70b-versatile', api_key = api_key)

def split_long_string(long_string, chunk_size=6000):
    return [long_string[i:i+chunk_size] for i in range(0, len(long_string), chunk_size)]


async def ppt_content(data):
    agent = Agent(model,system_prompt=(
        "You are an expert in making power-point perssentation",
        "Title Slide: Include the document's title, subtitle, author, and date.",
        "Methodology Slide: Summarize the methodology in detail in bullet points",
        "Results Slide: Present key findings in detail in simple text and bullet points.",
        "Discussion Slide: Summarize the implications and limitations.",
        "Conclusion Slide: State the overall conclusion.",
        "Reference Slide: Include all citations used."
    ))
    listOfString = split_long_string("".join(data))
    print(len(listOfString))

    
    result_1 = agent.run_sync(user_prompt = listOfString[0])
    result_2 = agent.run_sync(user_prompt = listOfString[1],message_history=result_1.new_messages())
    print(result_2.data)

def ai_ppt(data):
    asyncio.run(ppt_content(data=data))


def extract_data(feed):
    
    with pdfplumber.open(feed) as pdf:
        pages = pdf.pages
        for p in pages:
            data.append(p.extract_text())
    return None 



# if data is not None:
#     st.caption(data)
#     ai_ppt(data=data)

def main():
    uploaded_file = st.file_uploader('Choose your .pdf file', type="pdf")
    if uploaded_file is not None:
        extract_data(uploaded_file)
    if st.button("Search"):
        ai_ppt(data)

if __name__ == '__main__':
    import asyncio
    nest_asyncio.apply()
    main()