import gradio as gr
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader

# Load documents from "data" directory and build the index
documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents=documents)
query_engine = index.as_query_engine()

# Function to handle user queries
def query_document(query):
    response = query_engine.query(query)
    return str(response)

# Build Gradio app using Blocks for better layout and UX
with gr.Blocks(css=".gradio-container {font-family: 'Arial'; background-color: #fafafa;}") as demo:
    gr.Markdown("<h1 style='text-align: center;'>📄 RAG Application with LlamaIndex</h1>")
    gr.Markdown(
        "Ask questions about the documents stored in the local directory. "
        "This app uses Retrieval-Augmented Generation (RAG) powered by LlamaIndex."
    )

    with gr.Box():
        query_input = gr.Textbox(
            label="Enter your query",
            placeholder="e.g., What is the refund policy mentioned in the document?",
            lines=3
        )
        submit_btn = gr.Button("Submit", variant="primary")
        response_output = gr.Textbox(label="Response", lines=8)

    submit_btn.click(fn=query_document, inputs=query_input, outputs=response_output)

# Run the app
if __name__ == "__main__":
    demo.launch()