import streamlit as st
import pdfplumber
import io
from PyPDF2 import PdfReader
from pdf2image import convert_from_bytes


def extract_text_from_pdf(pdf_file):
    text = ""
    with pdfplumber.open(pdf_file) as pdf:
        for page in pdf.pages:
            text += page.extract_text() + "\n\n"
    return text


def main():
    st.title("Tops Infosolutions Pvt. Ltd.")
    st.write("Upload a PDF, view its pages, and convert it to Markdown.")

    uploaded_file = st.file_uploader("Choose a PDF file", type=["pdf"])

    if uploaded_file is not None:
        # Create a byte stream of the PDF without consuming the file pointer
        pdf_bytes = uploaded_file.getvalue()

        pdf_reader = PdfReader(io.BytesIO(pdf_bytes))
        total_pages = len(pdf_reader.pages)
        st.session_state.page_number = st.session_state.get("page_number", 0)

        col1, col2 = st.columns([1, 6])
        with col1:
            if st.button("Previous"):
                st.session_state.page_number = max(0, st.session_state.page_number - 1)
        with col2:
            if st.button("Next"):
                st.session_state.page_number = min(total_pages - 1, st.session_state.page_number + 1)

        # Convert PDF to images using the raw bytes
        images = convert_from_bytes(pdf_bytes, first_page=st.session_state.page_number + 1,
                                    last_page=st.session_state.page_number + 1)

        if images:
            st.image(images[0], use_column_width=True)

        page = pdf_reader.pages[st.session_state.page_number]
        text = page.extract_text()

        st.subheader(f"Page {st.session_state.page_number + 1} of {total_pages}")
        st.text_area("Extracted Text", text, height=300)

        md_file = io.BytesIO()
        md_file.write(text.encode('utf-8'))
        md_file.seek(0)

        st.download_button(
            label="Download Markdown",
            data=md_file,
            file_name="converted.md",
            mime="text/markdown"
        )


if __name__ == "__main__":
    main()