app.py

import os
import gradio as gr
import requests
import pandas as pd
from huggingface_hub import InferenceClient
from duckduckgo_search import DDGS
import wikipediaapi

# ==== CONFIG ====
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
HF_TOKEN = os.getenv("HF_TOKEN")

# Supported conversational/text-gen models in order of preference
CONVERSATIONAL_MODELS = [
    "deepseek-ai/DeepSeek-LLM",
    "HuggingFaceH4/zephyr-7b-beta",
    "mistralai/Mistral-7B-Instruct-v0.2"
]

wiki_api = wikipediaapi.Wikipedia(
    language="en", 
    user_agent="SmartAgent/1.0 ([email protected])"
)

# ==== SEARCH TOOLS ====
def duckduckgo_search(query):
    with DDGS() as ddgs:
        results = [r for r in ddgs.text(query, max_results=3)]
        return "\n".join([r.get("body", "") for r in results if r.get("body")]) or "No DuckDuckGo results found."

def wikipedia_search(query):
    page = wiki_api.page(query)
    return page.summary if page.exists() and page.summary else "No Wikipedia page found."

# ==== HUGGING FACE CHAT/TEXT-GEN TOOL ====
def hf_chat_model(question):
    last_error = ""
    for model_id in CONVERSATIONAL_MODELS:
        try:
            hf_client = InferenceClient(model_id, token=HF_TOKEN)
            # Try conversational endpoint first, if it exists
            if hasattr(hf_client, "conversational"):
                try:
                    result = hf_client.conversational(
                        messages=[{"role": "user", "content": question}],
                        max_new_tokens=384,
                    )
                    if isinstance(result, dict) and "generated_text" in result:
                        return f"[{model_id}] " + result["generated_text"]
                    elif hasattr(result, "generated_text"):
                        return f"[{model_id}] " + result.generated_text
                    elif isinstance(result, str):
                        return f"[{model_id}] " + result
                except Exception as e:
                    last_error += f"({model_id}: conversational) {e}\n"
            # Fall back to text_generation for all other models
            result = hf_client.text_generation(question, max_new_tokens=384)
            if isinstance(result, dict) and "generated_text" in result:
                return f"[{model_id}] " + result["generated_text"]
            elif hasattr(result, "generated_text"):
                return f"[{model_id}] " + result.generated_text
            elif isinstance(result, str):
                return f"[{model_id}] " + result
            else:
                return f"[{model_id}] " + str(result)
        except Exception as e:
            last_error += f"({model_id}: text_generation) {e}\n"
    return f"HF LLM error: {last_error or 'No models produced output.'}"

# ==== SMART AGENT ====
class SmartAgent:
    def __init__(self):
        pass

    def __call__(self, question: str) -> str:
        q_lower = question.lower()
        # DuckDuckGo for current/event/internet questions
        if any(term in q_lower for term in [
            "current", "latest", "2024", "2025", "who is the president", "recent", "live", "now", "today"
        ]):
            duck_result = duckduckgo_search(question)
            if duck_result and "No DuckDuckGo" not in duck_result:
                return duck_result
        # Wikipedia for encyclopedic knowledge
        wiki_result = wikipedia_search(question)
        if wiki_result and "No Wikipedia page found" not in wiki_result:
            return wiki_result
        # Fallback to LLMs
        return hf_chat_model(question)

# ==== SUBMISSION LOGIC ====
def run_and_submit_all(profile: gr.OAuthProfile | None):
    space_id = os.getenv("SPACE_ID")
    if profile:
        username = profile.username
    else:
        return "Please Login to Hugging Face with the button.", None

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

    agent = SmartAgent()
    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"

    try:
        response = requests.get(questions_url, timeout=15)
        response.raise_for_status()
        questions_data = response.json()
    except Exception as e:
        return f"Error fetching questions: {e}", None

    results_log = []
    answers_payload = []

    for item in questions_data:
        task_id = item.get("task_id")
        question_text = item.get("question")
        if not task_id or not question_text:
            continue
        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})

    if not answers_payload:
        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)

    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}

    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', 'No message received.')}"
        )
        results_df = pd.DataFrame(results_log)
        return final_status, results_df
    except Exception as e:
        return f"Submission Failed: {e}", pd.DataFrame(results_log)

# ==== GRADIO UI ====
with gr.Blocks() as demo:
    gr.Markdown("# Smart Agent Evaluation Runner")
    gr.Markdown("""
        **Instructions:**
        1. Clone this space, define your agent logic, tools, packages, etc.
        2. Log in to Hugging Face.
        3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
    """)
    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])

if __name__ == "__main__":
    demo.launch(debug=True, share=False)