app.py

import gradio as gr
import pandas as pd
import tempfile
import os
from keyword_extractor import KeywordExtractor

extractor = KeywordExtractor()

def extract_keywords_batch(file, download_format):
    try:
        # Read uploaded Excel file
        df = pd.read_excel(file.name)
        
        # Check for 'abstract' column
        if "Abstract" not in df.columns:
            return "❌ Error: 'Abstract' column not found in uploaded Excel file.", None

        results = []
        for i, row in df.iterrows():
            Abstract = str(row["Abstract"])
            keywords = extractor.extract(Abstract)
            results.append({
                "Abstract": Abstract,
                "keywords": ", ".join(keywords)
            })

        result_df = pd.DataFrame(results)

        # Save result to a temporary file (CSV or XLSX)
        tmp_dir = tempfile.mkdtemp()
        if download_format == "CSV":
            file_path = os.path.join(tmp_dir, "extracted_keywords.csv")
            result_df.to_csv(file_path, index=False)
        else:
            file_path = os.path.join(tmp_dir, "extracted_keywords.xlsx")
            result_df.to_excel(file_path, index=False)

        return result_df, file_path

    except Exception as e:
        return f"❌ Exception occurred: {str(e)}", None


# Gradio Interface
iface = gr.Interface(
    fn=extract_keywords_batch,
    inputs=[
        gr.File(label="Upload .xlsx File with Abstracts"),
        gr.Radio(["CSV", "XLSX"], label="Choose Download Format", value="CSV")
    ],
    outputs=[
        gr.Dataframe(label="Extracted Keywords"),
        gr.File(label="Download File")
    ],
    title="Batch Abstract Keyword Extractor",
    description="""
Upload an Excel (.xlsx) file with an **'abstract'** column to extract keywords from each abstract.
You can choose to download the results in CSV or XLSX format.
"""
)

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