app.py

import json
import gradio as gr
import pandas as pd
from pyspark.sql import SparkSession

# ── Initialize Spark ─────────────────────────────────────────────────────────
spark = (
    SparkSession
    .builder
    .master("local[*]")
    .appName("HF Spark Demo")
    .getOrCreate()
)

# ── Demo 1: Word Count ────────────────────────────────────────────────────────
def count_words(text: str) -> str:
    df = spark.createDataFrame([(text,)], ["sentence"])
    result = df.selectExpr("size(split(sentence, ' ')) as word_count").collect()[0]
    return f"Your input has {result['word_count']} words."

# ── Demo 2: JSON Data Explorer ───────────────────────────────────────────────
def load_example_json() -> pd.DataFrame:
    # Read the example JSON file into a Spark DataFrame
    df = spark.read.json("example_data.json")
    # Convert to pandas for Gradio display
    return df.toPandas()

# ── Build Gradio Interface ────────────────────────────────────────────────────
with gr.Blocks() as demo:
    gr.Markdown("## 🔥 Spark + Gradio Demo on 0.0.0.0")

    with gr.Tab("Word Count"):
        txt = gr.Textbox(lines=3, placeholder="Type something here...", label="Input Text")
        out = gr.Textbox(label="Word Count Result")
        txt.submit(count_words, txt, out)
        gr.Button("Count Words").click(count_words, txt, out)

    with gr.Tab("JSON Data Explorer"):
        df_table = gr.Dataframe(
            value=load_example_json(),
            label="Example Data",
            interactive=False
        )
        gr.Button("Reload Data").click(load_example_json, None, df_table)

# ── Launch ───────────────────────────────────────────────────────────────────
if __name__ == "__main__":
    demo.launch(
        server_name="0.0.0.0",
        server_port=7860,
        enable_queue=True
    )