app.py

import gradio as gr
import subprocess
import os
import uuid
import shutil

def convert_pdf_to_html(pdf_file):
    job_id = str(uuid.uuid4())
    work_dir = f"/tmp/{job_id}"
    os.makedirs(work_dir, exist_ok=True)

    input_path = os.path.join(work_dir, "input.pdf")
    output_path = os.path.join(work_dir, "input.html")

    with open(input_path, "wb") as f:
        f.write(pdf_file.read())

    try:
        result = subprocess.run(
            ["pdf2htmlEX", "--dest-dir", work_dir, "--embed", "cfijo", input_path],
            capture_output=True,
            text=True
        )

        if result.returncode != 0:
            return f"Error during conversion:\n{result.stderr}"

        if not os.path.exists(output_path):
            return "Conversion failed: output HTML file not found."

        return output_path

    finally:
        shutil.rmtree(work_dir, ignore_errors=True)

gr.Interface(
    fn=convert_pdf_to_html,
    inputs=gr.File(label="Upload PDF", file_types=[".pdf"]),
    outputs=gr.File(label="Download HTML"),
    title="PDF to HTML Converter",
    description="Uses pdf2htmlEX to convert PDFs to standalone HTML."
).launch()