import subprocess
import camelot
import pandas as pd
import gradio as gr
try:
    camelot.__version__
except ImportError:
    subprocess.run(["apt-get", "update"])
    subprocess.run(["apt-get", "install", "-y", "ghostscript"])

def extract_tables(pdf_file):
    tables = camelot.read_pdf(pdf_file.name, pages="all")
    df = pd.concat([table.df for table in tables], ignore_index=True)
    df.to_excel("output.xlsx", index=False)
    return "output.xlsx"

interface = gr.Interface(
    fn=extract_tables,
    inputs=gr.File(label="Upload PDF"),
    outputs=gr.File(label="Download Excel"),
    title="PDF Table Extractor",
    description="Extract tables from PDF and output as Excel file.",
)

if __name__ == "__main__":
    interface.launch()