# app.py
import os
from opik import track
import gradio as gr
import requests
from smolagents import DuckDuckGoSearchTool, CodeAgent, WikipediaSearchTool, LiteLLMModel, CodeAgent, tool , OpenAIServerModel , PythonInterpreterTool
from pathlib import Path
import pathlib
import tempfile
import PyPDF2   
from opik.integrations.openai import track_openai
import pytesseract             
from PIL import Image       
from smolagents.tools import PipelineTool, Tool
from typing import Union, Optional
import pandas as pd
from tabulate import tabulate  # pragma: no cover – fallback path
import re
import opik
import time
from tenacity import retry, stop_after_attempt, wait_exponential, retry_if_exception_type, before_sleep_log
import logging
import random
import sys

# Set up logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)

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

GROQ_API_KEY = os.getenv("Grok_api")       # set as Secret in your Space
OPIK_API_KEY = os.getenv("OPIK_API_KEY")   # set as Secret in your Space
OPIK_WORKSPACE = os.getenv("OPIK_WORKSPACE")  # set as Variable in your Space

# ── 2) Configure litellm & OpikLogger ─────────────────────────────────────────
os.environ["GROQ_API_KEY"] = GROQ_API_KEY
os.environ["OPIK_API_KEY"] = OPIK_API_KEY
os.environ["OPIK_WORKSPACE"] = OPIK_WORKSPACE

class ExcelToTextTool(Tool):
    """Render an Excel worksheet as Markdown text."""

    # ------------------------------------------------------------------
    # Required smol‑agents metadata
    # ------------------------------------------------------------------
    name = "excel_to_text"
    description = (
        "Read an Excel file and return a Markdown table of the requested sheet. "
        "Accepts either the sheet name or the zero-based index."
    )

    inputs = {
        "excel_path": {
            "type": "string",
            "description": "Path to the Excel file (.xlsx / .xls).",
        },
        "sheet_name": {
            "type": "string",
            "description": (
                "Worksheet name or zero‑based index *as a string* (optional; default first sheet)."
            ),
            "nullable": True,
        },
    }

    output_type = "string"
    

    # ------------------------------------------------------------------
    # Core logic
    # ------------------------------------------------------------------
    def forward(
            self,
            excel_path: str,
            sheet_name: Optional[str] = None,
    ) -> str:
        """Load *excel_path* and return the sheet as a Markdown table."""

        path = pathlib.Path(excel_path).expanduser().resolve()
        if not path.exists():
            return f"Error: Excel file not found at {path}"

        try:
            # Interpret sheet identifier -----------------------------------
            sheet: Union[str, int]
            if sheet_name is None or sheet_name == "":
                sheet = 0  # first sheet
            else:
                # If the user passed a numeric string (e.g. "1"), cast to int
                sheet = int(sheet_name) if sheet_name.isdigit() else sheet_name

            # Load worksheet ----------------------------------------------
            df = pd.read_excel(path, sheet_name=sheet)

            # Render to Markdown; fall back to tabulate if needed ---------
            if hasattr(pd.DataFrame, "to_markdown"):
                return df.to_markdown(index=False)
            from tabulate import tabulate  # pragma: no cover – fallback path

            return tabulate(df, headers="keys", tablefmt="github", showindex=False)

        except Exception as exc:  # broad catch keeps the agent chat‑friendly
            return f"Error reading Excel file: {exc}"


def download_file_if_any(base_api_url: str, task_id: str) -> str | None:
    """
    Try GET /files/{task_id}.
    • On HTTP 200 → save to a temp dir and return local path.
    • On 404 → return None.
    • On other errors → raise so caller can log / handle.
    """
    url = f"{base_api_url}/files/{task_id}"
    try:
        resp = requests.get(url, timeout=30)
        if resp.status_code == 404:
            return None          # no file
        resp.raise_for_status()   # raise on 4xx/5xx ≠ 404
    except requests.exceptions.HTTPError as e:
        # propagate non-404 errors (403, 500, …)
        raise e

    # ▸ Save bytes to a named file inside the system temp dir
    #    Try to keep original extension from Content-Disposition if present.
    cdisp = resp.headers.get("content-disposition", "")
    filename = task_id                                 # default base name
    if "filename=" in cdisp:
        m = re.search(r'filename="([^"]+)"', cdisp)
        if m:
            filename = m.group(1)                      # keep provided name

    tmp_dir = Path(tempfile.gettempdir()) / "gaia_files"
    tmp_dir.mkdir(exist_ok=True)
    file_path = tmp_dir / filename
    with open(file_path, "wb") as f:
        f.write(resp.content)
    return str(file_path)

@tool
def pdf_to_text_tool(pdf_path: str) -> str:
    """
    Extract all text from a PDF file.
    
    Args:
    pdf_path : Path to pdf's
    
    Returns:
    Analysis result or error message
    """
    path = Path(pdf_path).expanduser().resolve()
    if not path.exists():
        return f"Error: PDF file not found at {path}"
    try:
        reader = PyPDF2.PdfReader(str(path))
        text = "\n".join(page.extract_text() or "" for page in reader.pages)
        return text
    except Exception as e:
        return f"Error reading PDF file: {e}"

@tool
def analyze_image_tool(image_path: str) -> str:
    """
    Analyze an image: return dimensions and OCR-extracted text.
    Args:
    image_path : Image path
    
    Returns:
    Analysis result or error message
  
    """
    path = Path(image_path).expanduser().resolve()
    if not path.exists():
        return f"Error: Image not found at {path}"
    try:
        img = Image.open(path)
        w, h = img.size
        ocr_text = pytesseract.image_to_string(img)
        return f"Dimensions: {w}×{h}\n\nOCR Text:\n{ocr_text}"
    except Exception as e:
        return f"Error analyzing image: {e}"


# --- Basic Agent Definition ---

class BasicAgent:
    def __init__(self):
        self.agent = CodeAgent(
            model=OpenAIServerModel(model_id="gpt-4o"),
            tools=[DuckDuckGoSearchTool(), WikipediaSearchTool(), ExcelToTextTool(), pdf_to_text_tool, analyze_image_tool,PythonInterpreterTool()],
            add_base_tools=True,
            additional_authorized_imports=['pandas', 'numpy', 'csv', 'subprocess']
        )
        # Response cache to avoid duplicate API calls
        self.response_cache = {}
        logger.info("BasicAgent initialized.")

    def __call__(self, question: str) -> str:
        logger.info(f"Agent received question (first 50 chars): {question[:50]}...")
        
        # Check cache first
        if question in self.response_cache:
            logger.info("Using cached response")
            return self.response_cache[question]
        
        try:
            # Use the retry wrapper to handle rate limits
            fixed_answer = call_llm_with_retry(self.agent, question)
            logger.info(f"Agent returning answer (first 50 chars): {fixed_answer[:50] if fixed_answer else 'None'}...")
            
            # Cache the response
            self.response_cache[question] = fixed_answer
            return fixed_answer
        except Exception as e:
            error_msg = f"Failed after multiple retries: {e}"
            logger.error(error_msg)
            return f"The model experienced an issue that couldn't be resolved with retries: {str(e)}"

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 = "l3xv/Final_Assignment_Template"

    if profile:
        username = f"{profile.username}"
        logger.info(f"User logged in: {username}")
    else:
        logger.warning("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
    try:
        agent = BasicAgent()
    except Exception as e:
        logger.error(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
    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
    logger.info(f"Agent code URL: {agent_code}")

    # 2. Fetch Questions
    logger.info(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:
            logger.warning("Fetched questions list is empty.")
            return "Fetched questions list is empty or invalid format.", None
        logger.info(f"Fetched {len(questions_data)} questions.")
    except requests.exceptions.RequestException as e:
        logger.error(f"Error fetching questions: {e}")
        return f"Error fetching questions: {e}", None
    except requests.exceptions.JSONDecodeError as e:
        logger.error(f"Error decoding JSON response from questions endpoint: {e}")
        logger.error(f"Response text: {response.text[:500]}")
        return f"Error decoding server response for questions: {e}", None
    except Exception as e:
        logger.error(f"An unexpected error occurred fetching questions: {e}")
        return f"An unexpected error occurred fetching questions: {e}", None

    # 3. Run your Agent with rate limit handling and batching
    results_log = []
    answers_payload = []
    logger.info(f"Running agent on {len(questions_data)} questions...")
    
    # Process questions with a small delay between them to avoid rate limits
    batch_size = 1  # Process one at a time for rate limiting
    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:
            logger.warning(f"Skipping item with missing task_id or question: {item}")
            continue
            
        # ----------fetch any attached file ----------
        try:
            file_path = download_file_if_any(api_url, task_id)
        except Exception as e:
            file_path = None
            logger.error(f"[file fetch error] {task_id}: {e}")

        # ---------- Build the prompt sent to the agent ----------
        if file_path:
            q_for_agent = (
                f"{question_text}\n\n"
                f"---\n"
                f"A file was downloaded for this task and saved locally at:\n"
                f"{file_path}\n"
                f"---\n\n"
            )
        else:
            q_for_agent = question_text
            
        try:
            logger.info(f"Processing question {i+1}/{len(questions_data)}: {task_id}")
            submitted_answer = agent(q_for_agent)
            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})
            
            # Add a delay between questions to manage rate limits
            if i < len(questions_data) - 1:  # Don't delay after the last question
                delay = random.uniform(5, 10)  # Random delay between 5-10 seconds
                logger.info(f"Processed question {i+1}/{len(questions_data)}. Waiting {delay:.2f}s before next question...")
                time.sleep(delay)
                
        except Exception as e:
            logger.error(f"Error running agent on task {task_id}: {e}")
            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})

    if not answers_payload:
        logger.warning("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, "answers": answers_payload}
    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
    logger.info(status_update)

    # 5. Submit
    logger.info(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.')}"
        )
        logger.info("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}"
        logger.error(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."
        logger.error(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}"
        logger.error(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}"
        logger.error(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)
    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__":
    banner = "\n" + "-"*30 + " App Starting " + "-"*30
    logger.info(banner)
    
    # Check for SPACE_HOST and SPACE_ID at startup for information
    space_host_startup = os.getenv("SPACE_HOST")
    space_id_startup = "l3xv/Final_Assignment_Template"

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

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

    logger.info("-"*(60 + len(" App Starting ")) + "\n")

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