import gradio as gr
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

def execute_sql(user_query):
    model_name = "microsoft/tapex-large-sql-execution"  # Tapex large SQL execution model
    tokenizer = AutoTokenizer.from_pretrained(model_name)
    model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

    inputs = tokenizer(user_query, return_tensors="pt", padding=True)
    outputs = model.generate(inputs['input_ids'], attention_mask=inputs['attention_mask'], max_length=1024)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    return response

'''
def chatbot_response(user_message):
    # Your chatbot code goes here (using GPT-2 or any other text generation model)
    # For example, you can use the GPT-2 code from the previous responses

    return chatbot_generated_response

# Define the chatbot and SQL execution interface using Gradio
chatbot_interface = gr.Interface(
    fn=chatbot_response,
    inputs=gr.Textbox(prompt="You:"),
    outputs=gr.Textbox(),
    live=True,
    capture_session=True,
    title="Chatbot",
    description="Type your message in the box above, and the chatbot will respond.",
)
'''

sql_execution_interface = gr.Interface(
    fn=execute_sql,
    inputs=gr.Textbox(prompt="Enter your SQL query:"),
    outputs=gr.Textbox(),
    live=True,
    capture_session=True,
    title="SQL Execution",
    description="Type your SQL query in the box above, and the chatbot will execute it.",
)

# Combine the chatbot and SQL execution interfaces
#combined_interface = gr.Interface([chatbot_interface, sql_execution_interface], layout="horizontal")

# Launch the combined Gradio interface
if __name__ == "__main__":
    sql_execution_interface.launch()