app.py

import tempfile
import os
import streamlit as st
from PyPDF2 import PdfReader
from sentence_transformers import SentenceTransformer
import faiss
import numpy as np

# Function to process the uploaded PDF and save it temporarily
def process_pdf(file):
    with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmpfile:
        tmpfile.write(file.read())  # Write the uploaded file's content to the temp file
        tmpfile_path = tmpfile.name  # Get the temporary file path
    return tmpfile_path

# Function to extract text from the PDF
def extract_text_from_pdf(pdf_path):
    reader = PdfReader(pdf_path)
    text = ""
    for page in reader.pages:
        text += page.extract_text()
    return text

# Main function to run the Streamlit app
def main():
    st.title("PDF Embedding and Query System")

    # File uploader for the user to upload a PDF
    uploaded_file = st.file_uploader("Upload a PDF file", type=["pdf"])

    if uploaded_file is not None:
        # Process the uploaded PDF and get its file path
        tmp_file_path = process_pdf(uploaded_file)

        # Extract text from the uploaded PDF
        pdf_text = extract_text_from_pdf(tmp_file_path)

        # Initialize Sentence-Transformer model for embeddings
        model = SentenceTransformer('all-MiniLM-L6-v2')

        # Generate embeddings for the text (split into chunks)
        text_chunks = pdf_text.split("\n")  # Split text into lines or paragraphs
        embeddings = model.encode(text_chunks, convert_to_numpy=True)

        # Build FAISS index
        dimension = embeddings.shape[1]
        index = faiss.IndexFlatL2(dimension)
        index.add(embeddings)

        # Query input field for users to enter their search queries
        query = st.text_input("Enter a query to search:")

        if query:
            # Generate embedding for the query
            query_embedding = model.encode([query], convert_to_numpy=True)

            # Perform similarity search using FAISS
            D, I = index.search(query_embedding, k=5)

            # Display the results
            for i in range(len(I[0])):
                st.write(f"Match {i + 1}: {text_chunks[I[0][i]]} (Distance: {D[0][i]:.4f})")

# Run the app if this script is executed directly
if __name__ == "__main__":
    main()