app.py

# app.py
import os
import streamlit as st
from PyPDF2 import PdfReader
import faiss
import requests
from groq import Groq
import tempfile

def download_pdf_from_drive(link):
    file_id = link.split("/d/")[1].split("/")[0]
    download_url = f"https://drive.google.com/uc?export=download&id={file_id}"
    response = requests.get(download_url)
    if response.status_code == 200:
        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".pdf")
        with open(temp_file.name, "wb") as f:
            f.write(response.content)
        return temp_file.name
    else:
        st.error("Failed to download PDF file from Google Drive.")
        return None

# Groq API Setup
GROQ_API_KEY = os.environ.get("GROQ_API_KEY")
client = Groq(api_key=GROQ_API_KEY)

# PDF Data Extraction Function
def extract_text_from_pdf(pdf_path):
    reader = PdfReader(pdf_path)
    text = ""
    for page in reader.pages:
        text += page.extract_text() + "\\n"
    return text

# FAISS Vector Store Setup
def create_faiss_index(text):
    # Dummy example: tokenize and embed text
    # In production, use an actual embedding model
    import numpy as np
    tokenized = text.split(" ")
    vectors = [np.random.rand(128) for _ in tokenized]  # Replace with real embeddings

    index = faiss.IndexFlatL2(128)
    faiss_vectors = np.array(vectors, dtype='float32')
    index.add(faiss_vectors)
    return index, tokenized

# Query Function
def query_faiss_index(index, tokenized_text, query):
    import numpy as np
    query_vector = np.random.rand(128).astype('float32')  # Replace with real embedding
    distances, indices = index.search(np.array([query_vector]), k=5)
    results = [tokenized_text[i] for i in indices[0]]
    return results

# Streamlit Frontend
def main():
    st.title("RAG-Based Application")

    drive_link = st.text_input("Enter Google Drive PDF Link")
    query = st.text_input("Enter your query")

    if drive_link and query:
        st.write("Downloading PDF from Google Drive...")
        pdf_path = download_pdf_from_drive(drive_link)

        if pdf_path:
            st.write("Extracting data from PDF...")
            text = extract_text_from_pdf(pdf_path)
            st.write("Data extracted successfully!")

            st.write("Creating FAISS index...")
            index, tokenized_text = create_faiss_index(text)
            st.write("Index created successfully!")

            st.write("Querying the index...")
            results = query_faiss_index(index, tokenized_text, query)
            st.write("Results:")
            for result in results:
                st.write(result)

    if st.button("Ask Groq API"):
        messages = [
            {
                "role": "user",
                "content": query,
            }
        ]
        chat_completion = client.chat.completions.create(
            messages=messages, model="llama-3.3-70b-versatile"
        )
        st.write(chat_completion.choices[0].message.content)

if __name__ == "__main__":
    main()