app.py

import os
import fitz  # PyMuPDF for parsing PDF
import streamlit as st
from transformers import pipeline
import re

# Function to extract text from a PDF file
def extract_text_from_pdf(pdf_path):
    text = ""
    with fitz.open(pdf_path) as pdf_document:
        for page_num in range(pdf_document.page_count):
            page = pdf_document.load_page(page_num)
            page_text = page.get_text()
            text += page_text
            yield page_num + 1, page_text  # Return the page number (1-based) and the extracted text

# Function to truncate text to the nearest word boundary
def truncate_to_word_boundary(text, max_words=100):
    words = re.findall(r'\w+', text)
    truncated_text = ' '.join(words[:max_words])
    return truncated_text

# Function to perform question-answering
def question_answering(question, pdf_text_with_pages):
    pdf_text = "\n".join([text for _, text in pdf_text_with_pages])

    # Perform question-answering using Hugging Face's Transformers
    question_answerer = pipeline("question-answering", model="distilbert-base-cased-distilled-squad", tokenizer="distilbert-base-cased-distilled-squad")
    answer = question_answerer(question=question, context=pdf_text)

    return answer

def main():
    st.title("Question Answering using a PDF Document")

    pdf_file = st.file_uploader("Upload a PDF file:", type=["pdf"])
    question = st.text_input("Ask your question:")

    if st.button("Answer"):
        if pdf_file:
            pdf_path = os.path.join(os.getcwd(), pdf_file.name)
            with open(pdf_path, "wb") as f:
                f.write(pdf_file.read())

            # Extract text from the PDF along with page numbers
            pdf_text_with_pages = list(extract_text_from_pdf(pdf_path))

            # Perform question-answering
            answer = question_answering(question, pdf_text_with_pages)

            os.remove(pdf_path)  # Delete the uploaded file after processing

            st.write(f"Question: '{question}'")
            st.write("Answer:", answer['answer'])
            st.write("Score:", answer['score'])
            st.write("Page Number:", answer['start'] + 1)  # Add 1 to convert 0-based index to 1-based page number

            # Display truncated context
            start_page = answer['start']
            context = pdf_text_with_pages[start_page][1]
            truncated_context = truncate_to_word_boundary(context)
            st.write("Context:", truncated_context)

if __name__ == "__main__":
    main()