app.py

import gradio as gr
from grobid_client.grobid_client import GrobidClient


def extract_text(file):
    client = GrobidClient(config_path="./config.json")
    information = client.process_pdf(
        "processFulltextDocument",
        file.name,
        generateIDs=False,
        consolidate_header=False,
        consolidate_citations=False,
        include_raw_citations=False,
        include_raw_affiliations=False,
        tei_coordinates=False,
        segment_sentences=False,
    )
    return information


# Ask Dr Ahmad about which LLM to use and if we have a token for it
with gr.Blocks() as demo:
    file_input = gr.File(
        label="Upload a research paper as a pdf file", file_types=[".pdf"]
    )
    text_output = gr.Textbox(label="Extracted Text")

    file_input.upload(fn=extract_text, inputs=file_input, outputs=text_output)

demo.launch()