app.py

# app.py (Minimal "Dummy" Version for Step 1)

import os
import gradio as gr
import requests
import pandas as pd
import logging

# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"

# -----------------------------------------------------------------
# --- AGENT LOGIC IS COMPLETELY REMOVED OR COMMENTED OUT ---
# -----------------------------------------------------------------


# We are replacing the entire GaiaAgent with this simple class.
# It has no external dependencies.
class DummyAgent:
    def __init__(self):
        logging.info("DummyAgent initialized. No models or tools loaded.")

    def __call__(self, question: str) -> str:
        logging.info(f"DummyAgent received question: {question[:50]}...")
        fixed_answer = "dummy_answer"
        logging.info(f"DummyAgent returning fixed answer: {fixed_answer}")
        return fixed_answer


# All other imports and tool definitions are removed for this test.
# from langchain_community.llms import HuggingFaceHub
# from langchain_community.tools import DuckDuckGoSearchRun
# ... and so on ...

# All tool definitions are removed.
# @tool
# def web_search...

# The AgentState class is removed.
# class AgentState...

# The GaiaAgent class is replaced by DummyAgent above.

# -----------------------------------------------------------------
# --- GRADIO APP AND SUBMISSION LOGIC (largely unchanged) ---
# -----------------------------------------------------------------


def run_and_submit_all(profile: gr.OAuthProfile | None):
    """
    Fetches all questions, runs the DUMMY agent, and submits the answers.
    """
    if not profile:
        logging.warning("User not logged in.")
        return "Please Login to Hugging Face with the button.", None

    username = profile.username
    logging.info(f"User logged in: {username}")

    space_id = os.getenv("SPACE_ID")
    if not space_id:
        logging.error("SPACE_ID environment variable is not set.")
        return "CRITICAL ERROR: SPACE_ID environment variable is not set.", None

    api_url = DEFAULT_API_URL
    questions_url = f"{api_url}/questions"
    submit_url = f"{api_url}/submit"

    # 1. Instantiate Agent (Using the DummyAgent for this test)
    try:
        # We instantiate our simple, harmless agent.
        agent = DummyAgent()
    except Exception as e:
        logging.critical(
            f"Fatal error instantiating even the DummyAgent: {e}", exc_info=True
        )
        return f"Fatal error initializing agent: {e}", None

    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
    logging.info(f"Agent code URL: {agent_code}")

    # 2. Fetch Questions
    logging.info(f"Fetching questions from: {questions_url}")
    try:
        response = requests.get(questions_url, timeout=20)
        response.raise_for_status()
        questions_data = response.json()
        if not questions_data:
            logging.warning("Fetched questions list is empty.")
            return "Fetched questions list is empty.", None
        logging.info(f"Fetched {len(questions_data)} questions.")
    except Exception as e:
        logging.error(f"Error fetching questions: {e}")
        return f"Error fetching questions: {e}", None

    # 3. Run your Agent
    results_log = []
    answers_payload = []
    logging.info(f"Running dummy agent on {len(questions_data)} questions...")
    for i, item in enumerate(questions_data):
        task_id = item.get("task_id")
        question_text = item.get("question")
        if not task_id or question_text is None:
            continue
        try:
            # The agent call is now super fast and simple.
            submitted_answer = agent(question_text)
            answers_payload.append(
                {"task_id": task_id, "submitted_answer": submitted_answer}
            )
            results_log.append(
                {
                    "Task ID": task_id,
                    "Question": question_text,
                    "Submitted Answer": submitted_answer,
                }
            )
        except Exception as e:
            logging.error(
                f"Error running dummy agent on task {task_id}: {e}", exc_info=True
            )
            results_log.append(
                {
                    "Task ID": task_id,
                    "Question": question_text,
                    "Submitted Answer": f"DUMMY AGENT ERROR: {e}",
                }
            )

    if not answers_payload:
        logging.warning("Dummy agent did not produce any answers.")
        return "Dummy agent did not produce any answers.", pd.DataFrame(results_log)

    # 4. Prepare and Submit
    submission_data = {
        "username": username.strip(),
        "agent_code": agent_code,
        "answers": answers_payload,
    }
    logging.info(f"Submitting {len(answers_payload)} answers for user '{username}'...")
    try:
        response = requests.post(submit_url, json=submission_data, timeout=60)
        response.raise_for_status()
        result_data = response.json()
        final_status = (
            f"Submission Successful!\n"
            f"User: {result_data.get('username')}\n"
            f"Overall Score: {result_data.get('score', 'N/A')}% "
            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
            f"Message: {result_data.get('message', 'This was a test with a dummy agent.')}"
        )
        logging.info("Submission successful.")
        return final_status, pd.DataFrame(results_log)
    except Exception as e:
        logging.critical(
            f"An unexpected error occurred during submission: {e}", exc_info=True
        )
        return f"An unexpected error occurred during submission: {e}", pd.DataFrame(
            results_log
        )


# --- Build Gradio Interface (Unchanged) ---
with gr.Blocks() as demo:
    gr.Markdown("# GAIA Agent Evaluation Runner (Minimal Test)")
    gr.Markdown("This is a minimal version to test the basic app stability.")
    gr.LoginButton()
    run_button = gr.Button("Run Evaluation & Submit All Answers")
    status_output = gr.Textbox(
        label="Run Status / Submission Result", lines=5, interactive=False
    )
    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
    run_button.click(
        fn=run_and_submit_all,
        outputs=[status_output, results_table],
        api_name="run_evaluation",
    )

if __name__ == "__main__":
    logging.basicConfig(
        level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
    )
    logging.info("App Starting (Minimal Version)...")
    demo.launch(debug=True, share=False)