app.py

import gradio as gr
from docling.document_converter import DocumentConverter
import json
import os
import tempfile

# 1) Initialize DocumentConverter
converter = DocumentConverter()

# 2) Conversion function
def convert(file, out_format):
    if file is None:
        return "Please upload a file first."

    try:
        # Convert the file
        doc = converter.convert(file.name)

        # Export based on the selected format
        if out_format == "Markdown":
            # Save as Markdown file
            with tempfile.NamedTemporaryFile(delete=False, suffix=".md") as temp_file:
                temp_filename = temp_file.name
                with open(temp_filename, 'w') as f:
                    f.write(doc.document.export_to_markdown())
                return temp_filename  # Return file for download

        elif out_format == "HTML":
            # Save as HTML file
            with tempfile.NamedTemporaryFile(delete=False, suffix=".html") as temp_file:
                temp_filename = temp_file.name
                with open(temp_filename, 'w') as f:
                    f.write(doc.document.export_to_html())
                return temp_filename  # Return file for download

        elif out_format == "JSON":
            # Save as JSON file (lossless serialization)
            doctags = doc.document.export_to_dict()  # Correct method for JSON export
            with tempfile.NamedTemporaryFile(delete=False, suffix=".json") as temp_file:
                temp_filename = temp_file.name
                with open(temp_filename, 'w') as f:
                    json.dump(doctags, f, indent=4)
                return temp_filename  # Return file for download

        elif out_format == "Text":
            # Save as Text file (plain text)
            with tempfile.NamedTemporaryFile(delete=False, suffix=".txt") as temp_file:
                temp_filename = temp_file.name
                with open(temp_filename, 'w') as f:
                    f.write(doc.document.export_to_text())  # Correct method for plain text extraction
                return temp_filename  # Return file for download

        elif out_format == "Doctags":
            # Save as Doctags file
            with tempfile.NamedTemporaryFile(delete=False, suffix=".doctags") as temp_file:
                temp_filename = temp_file.name
                doc.document.save_as_doctags(temp_filename)
                return temp_filename  # Return file for download

        else:
            return "Unsupported output format"

    except Exception as e:
        return f"Error converting document: {str(e)}"

# 3) Build Gradio interface
with gr.Blocks() as demo:
    # Top: explanatory text about Docling and a link to the official repository
    gr.Markdown(
        """
        # 📝 Docling Demo
        **Docling** is an open-source toolkit for converting various document formats into structured, machine-readable formats. It supports a wide variety of formats including PDF, DOCX, images, and more. 

        You can choose to convert your document into:
        - Markdown
        - HTML
        - JSON (Lossless serialization of Docling Document)
        - Text (Plain text, i.e., without Markdown markers)
        - Doctags (A JSON-like structure preserving the document's original format)

        Docling is designed to simplify document processing, making it easier to extract content and structure from documents.

        👉 [Visit the official Docling GitHub repository here](https://github.com/DS4SD/docling)

        Upload any supported file, choose your output format, and see the result instantly.
        """
    )

    # Main row: uploader + options + convert button
    with gr.Row():
        inp = gr.File(
            label="📂 Upload your document",
            file_count="single",
            file_types=[".pdf", ".docx", ".pptx", ".xlsx", 
                       ".html", ".png", ".jpg", ".jpeg", ".tiff", 
                       ".wav", ".mp3"]
        )
        fmt = gr.Dropdown(
            choices=["Markdown", "HTML", "JSON", "Text", "Doctags"],
            label="🔄 Select output format",
            value="Markdown"
        )
        btn = gr.Button("Convert")

    # Output panel (file output for all formats)
    out = gr.File(label="📄 Download Output File")

    # Wire it up
    btn.click(fn=convert, inputs=[inp, fmt], outputs=out)

# 4) Launch
if __name__ == "__main__":
    demo.launch()