app.py

import os
import gradio as gr
import requests
import pandas as pd
import time
import threading
import random
import litellm
from litellm import RateLimitError

# Enable debug mode to see detailed error information
litellm.verbose = True
from smolagents import (
    CodeAgent,
    DuckDuckGoSearchTool,
    VisitWebpageTool,
    PythonInterpreterTool,
    WikipediaSearchTool,
    SpeechToTextTool,
    LiteLLMModel,
)
from libs.questionHelper.file_tools import fetch_task_files
from libs.chess.chess_tools import analyze_chess_image, analyze_chess_position
from libs.transcription.transcription_tools import transcribe_audio
from libs.youtube.youtube_tools import analyze_youtube_video, get_youtube_video_info


# (Keep Constants as is)
# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
FILES_AVAILABLE_PREFIX = "FILES_AVAILABLE: "
FILES_AVAILABLE_SUFFIX = "\n\n"


# --- Basic Agent Definition ---
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------


class TokenBucketRateLimiter:
    """Simple token-bucket rate limiter.

    capacity: max tokens in bucket (burst size)
    refill_rate: tokens added per second
    """

    def __init__(self, capacity: int, refill_rate: float):
        self.capacity = float(capacity)
        self._tokens = float(capacity)
        self.refill_rate = float(refill_rate)
        self._lock = threading.Lock()
        self._last = time.monotonic()

    def acquire(self, tokens: float = 1.0):
        with self._lock:
            now = time.monotonic()
            elapsed = now - self._last
            # Refill
            self._tokens = min(self.capacity, self._tokens + elapsed * self.refill_rate)
            self._last = now
            if self._tokens >= tokens:
                self._tokens -= tokens
                return 0.0
            # Need to wait for enough tokens
            required = tokens - self._tokens
            wait_time = required / self.refill_rate
            # consume what will be available after waiting
            self._tokens = 0.0
            return wait_time


class RateLimitedModel:
    """Wraps a model-like callable and enforces a TokenBucketRateLimiter before each call with retry logic."""

    def __init__(
        self,
        model_obj,
        rpm: int = 8,
        burst: int | None = None,
        max_retries: int = 10,
        base_delay: float = 30.0,
    ):
        self._model = model_obj
        self.max_retries = max_retries
        self.base_delay = base_delay
        # rpm -> tokens per minute
        capacity = burst if burst is not None else max(1, rpm)
        refill_rate = float(rpm) / 60.0
        self._limiter = TokenBucketRateLimiter(
            capacity=capacity, refill_rate=refill_rate
        )

    def _call_with_retry(self, func, *args, **kwargs):
        """Call a function with retry logic for rate limit errors."""
        last_exception = None
        for attempt in range(1, self.max_retries + 1):
            try:
                # Apply rate limiting before each attempt
                wait = self._limiter.acquire(1.0)
                if wait > 0:
                    jitter = random.uniform(0.0, 0.5)
                    total_wait = wait + jitter
                    print(
                        f"RateLimitedModel sleeping {total_wait:.2f}s to respect RPM limit"
                    )
                    time.sleep(total_wait)

                print(f"Model call attempt {attempt} of {self.max_retries}")
                result = func(*args, **kwargs)
                print(f"Model call attempt {attempt} succeeded")
                return result

            except Exception as e:
                last_exception = e
                error_str = str(e).lower()

                # Check if this is a rate limit error (various ways it might be reported)
                is_rate_limit = (
                    isinstance(e, RateLimitError)
                    or "rate limit" in error_str
                    or "quota" in error_str
                    or "429" in error_str
                    or "resource_exhausted" in error_str
                    or "too many requests" in error_str
                )

                # Check if this is a 503 server overload error
                is_server_overload = (
                    "503" in error_str
                    or "overloaded" in error_str
                    or "unavailable" in error_str
                    or "service unavailable" in error_str
                    or "internalservererror" in error_str
                )

                # Retry for both rate limit and server overload errors
                if is_rate_limit or is_server_overload:
                    error_type = (
                        "Rate limit" if is_rate_limit else "Server overload (503)"
                    )
                    print(f"{error_type} error on attempt {attempt}: {e}")
                    if attempt < self.max_retries:
                        # Use exponential backoff for 503 errors, longer delays
                        if is_server_overload:
                            delay = min(
                                120, self.base_delay * (2**attempt)
                            ) + random.uniform(0, 10)
                        else:
                            delay = self.base_delay + random.uniform(0, 5)
                        print(f"Waiting {delay:.1f}s before retry {attempt + 1}...")
                        time.sleep(delay)
                        continue
                else:
                    # Non-retryable error
                    print(f"Non-retryable error on attempt {attempt}: {e}")
                    raise e

        # All retries exhausted
        print(f"All {self.max_retries} attempts failed. Raising last exception.")
        raise last_exception

    def __call__(self, *args, **kwargs):
        return self._call_with_retry(self._model, *args, **kwargs)

    def __getattr__(self, name: str):
        """Proxy attribute access to the underlying model.

        For callable attributes (like `generate`) we wrap the call so the
        token-bucket rate limiter and retry logic are applied consistently.
        """
        # Avoid recursion
        if name.startswith("_"):
            raise AttributeError(name)

        attr = getattr(self._model, name)

        if callable(attr):

            def wrapped(*args, **kwargs):
                return self._call_with_retry(attr, *args, **kwargs)

            # Preserve original metadata where possible
            try:
                wrapped.__name__ = getattr(attr, "__name__", wrapped.__name__)
            except Exception:
                pass
            return wrapped

        return attr


# Wrap the model with a rate-limiter and retry logic. Default RPM is reduced to 8
# but can be configured via the MODEL_RPM environment variable.
_configured_rpm = int(os.getenv("MODEL_RPM", "8"))
_configured_burst = None
_configured_max_retries = int(os.getenv("MODEL_MAX_RETRIES", "10"))
_configured_base_delay = float(os.getenv("MODEL_BASE_DELAY", "30.0"))

# You can switch models here if Gemini continues to have issues
# Alternative options:
# - "gemini/gemini-1.5-flash" (older but more stable)
# - "gemini/gemini-1.5-pro" (more expensive but more capacity)
# - "gpt-4o-mini" or "gpt-3.5-turbo" (OpenAI alternatives)
_model_id = os.getenv("MODEL_ID", "gemini/gemini-2.5-flash")
print(f"Using model: {_model_id}")

model = RateLimitedModel(
    LiteLLMModel(model_id=_model_id, temperature=0.2),
    rpm=_configured_rpm,
    burst=_configured_burst,
    max_retries=_configured_max_retries,
    base_delay=_configured_base_delay,
)


class BasicAgent:
    def __init__(self, name: str = "GGSAgent"):
        self.name = name
        self.code_agent = CodeAgent(
            tools=[
                DuckDuckGoSearchTool(),
                VisitWebpageTool(),
                PythonInterpreterTool(),
                WikipediaSearchTool(),
                SpeechToTextTool(),
                transcribe_audio,
                analyze_youtube_video,
                get_youtube_video_info,
                analyze_chess_position,
                analyze_chess_image,
            ],
            model=model,
            max_steps=20,
            verbosity_level=1,
            additional_authorized_imports=[
                "json",
                "math",
                "pandas",
                "yt_dlp",
                "tempfile",
                "os",
                "torch",
                "whisper",
                "re",
                "litellm",
                "requests",
                "time",
                "threading",
                "random",
                "cv2",
                "numpy",
                "PIL",
                "base64",
                "io",
                "pathlib",
                "subprocess",
            ],
            add_base_tools=True,
        )
        print("BasicAgent initialized.")

    def __call__(self, question: str) -> str:
        print(f"Agent received question (first 50 chars): {question[:50]}...")
        print(f"Starting agent execution with model retry logic enabled...")

        start_time = time.time()
        try:
            # The retry logic is now handled at the model level within RateLimitedModel
            # so we can call the agent directly
            response = self.code_agent(question)

            duration = time.time() - start_time
            print(f"Agent completed successfully in {duration:.1f}s")
            print(f"Agent returning response: {response}")
            return response
        except Exception as e:
            duration = time.time() - start_time
            print(f"Error in code agent after {duration:.1f}s: {e}")
            return f"AGENT ERROR: {e}"


CACHE_DIR = "cache/gaia_validation"


def run_and_submit_all(profile: gr.OAuthProfile | None):
    """
    Fetches all questions, runs the BasicAgent on them, submits all answers,
    and displays the results.
    """
    # --- Determine HF Space Runtime URL and Repo URL ---
    space_id = os.getenv("SPACE_ID")  # Get the SPACE_ID for sending link to the code

    if profile:
        username = f"{profile.username}"
        print(f"User logged in: {username}")
    else:
        print("User not logged in.")
        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"

    # 1. Instantiate Agent ( modify this part to create your agent)
    try:
        agent_name = os.getenv("AGENT_NAME", "GGSAgent")
        agent = BasicAgent(name=agent_name)
    except Exception as e:
        print(f"Error instantiating agent: {e}")
        return f"Error initializing agent: {e}", None
    # In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
    print(agent_code)

    # 2. Fetch Questions
    print(f"Fetching questions from: {questions_url}")
    try:
        response = requests.get(questions_url, timeout=15)
        response.raise_for_status()
        questions_data = response.json()
        if not questions_data:
            print("Fetched questions list is empty.")
            return "Fetched questions list is empty or invalid format.", None
        print(f"Fetched {len(questions_data)} questions.")
    except requests.exceptions.RequestException as e:
        print(f"Error fetching questions: {e}")
        return f"Error fetching questions: {e}", None
    except requests.exceptions.JSONDecodeError as e:
        print(f"Error decoding JSON response from questions endpoint: {e}")
        print(f"Response text: {response.text[:500]}")
        return f"Error decoding server response for questions: {e}", None
    except Exception as e:
        print(f"An unexpected error occurred fetching questions: {e}")
        return f"An unexpected error occurred fetching questions: {e}", None

    # 3. Run your Agent
    results_log = []
    answers_payload = []
    print(f"Running agent on {len(questions_data)} questions...")
    for item in questions_data:
        task_id = item.get("task_id")
        question_text = item.get("question")
        if not task_id or question_text is None:
            print(f"Skipping item with missing task_id or question: {item}")
            continue
        # Fetch any associated files from GAIA validation (if present) and prepend a brief summary to the question
        try:
            try:
                file_results = fetch_task_files(
                    task_id, dest_dir=CACHE_DIR, transcribe_mp3=False
                )
            except Exception as e:
                print(f"Warning: failed to fetch files for {task_id}: {e}")
                file_results = {}

            # Build a compact file summary for the agent prompt
            file_summaries = []
            for ext, info in (file_results or {}).items():
                status = info.get("status")
                path = info.get("path")
                if status == "ok" and path:
                    file_summaries.append(f"{ext}=OK@{path}")
                else:
                    file_summaries.append(f"{ext}={status}")

            files_note = (
                ""
                if not file_summaries
                else (
                    FILES_AVAILABLE_PREFIX
                    + "; ".join(file_summaries)
                    + FILES_AVAILABLE_SUFFIX
                )
            )

            prompt_with_files = files_note + question_text

            submitted_answer = agent(prompt_with_files)
            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:
            print(f"Error running agent on task {task_id}: {e}")
            error_answer = f"AGENT ERROR: {e}"
            answers_payload.append(
                {"task_id": task_id, "submitted_answer": error_answer}
            )
            results_log.append(
                {
                    "Task ID": task_id,
                    "Question": question_text,
                    "Submitted Answer": error_answer,
                }
            )

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

    # 4. Prepare Submission
    submission_data = {
        "username": username.strip(),
        "agent_code": agent_code,
        "agent_name": getattr(agent, "name", "BasicAgent"),
        "answers": answers_payload,
    }
    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
    print(status_update)

    # 5. Submit
    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
    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.')}"
        )
        print("Submission successful.")
        results_df = pd.DataFrame(results_log)
        return final_status, results_df
    except requests.exceptions.HTTPError as e:
        error_detail = f"Server responded with status {e.response.status_code}."
        try:
            error_json = e.response.json()
            error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
        except requests.exceptions.JSONDecodeError:
            error_detail += f" Response: {e.response.text[:500]}"
        status_message = f"Submission Failed: {error_detail}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except requests.exceptions.Timeout:
        status_message = "Submission Failed: The request timed out."
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except requests.exceptions.RequestException as e:
        status_message = f"Submission Failed: Network error - {e}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df
    except Exception as e:
        status_message = f"An unexpected error occurred during submission: {e}"
        print(status_message)
        results_df = pd.DataFrame(results_log)
        return status_message, results_df


# --- Build Gradio Interface using Blocks ---
with gr.Blocks() as demo:
    gr.Markdown("# Basic Agent Evaluation Runner")
    gr.Markdown(
        """
        **Instructions:**

        1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
        2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.

        ---
        **Disclaimers:**
        Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
        This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
        """
    )

    gr.LoginButton()

    run_button = gr.Button("Run Evaluation & Submit All Answers")

    status_output = gr.Textbox(
        label="Run Status / Submission Result", lines=5, interactive=False
    )
    # Removed max_rows=10 from DataFrame constructor
    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__":
    print("\n" + "-" * 30 + " App Starting " + "-" * 30)
    # Check for SPACE_HOST and SPACE_ID at startup for information
    space_host_startup = os.getenv("SPACE_HOST")
    space_id_startup = os.getenv("SPACE_ID")  # Get SPACE_ID at startup

    if space_host_startup:
        print(f"✅ SPACE_HOST found: {space_host_startup}")
        print(f"   Runtime URL should be: https://{space_host_startup}.hf.space")
    else:
        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")

    if space_id_startup:  # Print repo URLs if SPACE_ID is found
        print(f"✅ SPACE_ID found: {space_id_startup}")
        print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
        print(
            f"   Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main"
        )
    else:
        print(
            "ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined."
        )

    print("-" * (60 + len(" App Starting ")) + "\n")

    print("Launching Gradio Interface for Basic Agent Evaluation...")
    demo.launch(debug=True, share=False)