app.py

import os
import gradio as gr
import requests
import pandas as pd
from smolagents import CodeAgent, DuckDuckGoSearchTool, OpenAIServerModel, Tool, PythonInterpreterTool
import os
import re
import requests
import pandas as pd
from youtube_transcript_api import YouTubeTranscriptApi
import whisper
from SPARQLWrapper import SPARQLWrapper, JSON
import chess
import chess.engine
import shutil
import traceback
# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
class YouTubeTranscriptTool(Tool):
    name = "youtube_transcript"
    description = (
        "Fetches the transcript of a YouTube video given its URL or ID.\n"
        "Returns plain text (no timestamps) or raw with timestamps."
    )
    inputs = {
        "video_url": {"type": "string", "description": "YouTube URL or video ID."},
        "raw": {"type": "boolean", "description": "Include timestamps?", "nullable": True}
    }
    output_type = "string"

    def forward(self, video_url: str, raw: bool = False) -> str:
        # Extract video ID
        video_id = video_url.strip().split("?v=")[-1]
        transcript = YouTubeTranscriptApi.get_transcript(video_id)
        if raw:
            return "\n".join(f"{int(e['start'])}s: {e['text']}" for e in transcript)
        return " ".join(e['text'] for e in transcript)


class SpeechToTextTool(Tool):
    name = "speech_to_text"
    description = (
        "Converts an audio file to text using OpenAI Whisper."
    )
    inputs = {
        "audio_path": {"type": "string", "description": "Path to audio file (.mp3, .wav)"},
    }
    output_type = "string"

    def __init__(self):
        super().__init__()
        self.model = whisper.load_model("base")

    def forward(self, audio_path: str) -> str:
        if not os.path.exists(audio_path):
            return f"Error: File not found at {audio_path}"
        result = self.model.transcribe(audio_path)
        return result.get("text", "")

class WikidataQueryTool(Tool):
    name = "wikidata_query"
    description = (
        "Runs SPARQL queries on Wikidata and returns JSON results."
    )
    inputs = {
        "query": {"type": "string", "description": "The SPARQL query to execute."}
    }
    output_type = "dict"

    def forward(self, query: str) -> dict:
        sparql = SPARQLWrapper("https://query.wikidata.org/sparql")
        sparql.setQuery(query)
        sparql.setReturnFormat(JSON)
        results = sparql.query().convert()
        return results


class WikipediaAPIExtractor(Tool):
    name = "wikipedia_api"
    description = (
        "Fetches and parses a Wikipedia page's sections as structured JSON."  
    )
    inputs = {
        "page_title": {"type": "string", "description": "Title of the Wikipedia page."}
    }
    output_type = "dict"

    def forward(self, page_title: str) -> dict:
        api_url = (
            f"https://en.wikipedia.org/api/rest_v1/page/mobile-sections/{page_title}"
        )
        resp = requests.get(api_url)
        resp.raise_for_status()
        data = resp.json()
        # merge lead with sections without requiring Python 3.9's dict union
        lead = data.get("lead", {})
        sections = data.get("remaining", {}).get("sections", [])
        return {**lead, "sections": sections}
    
class TableParseTool(Tool):
    name = "table_parse"
    description = (
        "Parses an ASCII or markdown table (or image) into a pandas DataFrame."
    )
    inputs = {
        "table_text": {"type": "string", "description": "The raw table string."}
    }
    output_type = "pandas.DataFrame"

    def forward(self, table_text: str) -> pd.DataFrame:
        # Leveraging pandas read_csv on StringIO with markdown separators
        from io import StringIO
        # Clean pipes and extra spaces
        clean = re.sub(r"^\||\|$", "", table_text.strip(), flags=re.MULTILINE)
        return pd.read_csv(StringIO(clean), sep=r"\s*\|\s*", engine="python")

class ChessEngineTool(Tool):
    name = "chess_engine"
    description = "Analyzes a chess position (FEN) with Stockfish and returns the best move."
    inputs = {
        "fen": {"type": "string", "description": "FEN string of the position."},
        "time_limit": {"type": "number", "description": "Time in seconds for engine analysis.", "nullable": True}
    }
    output_type = "string"

    def forward(self, fen: str, time_limit: float = 0.1) -> str:
        # figure out where the binary actually is
        sf_bin = shutil.which("stockfish") or "/usr/games/stockfish"
        if not sf_bin:
            raise RuntimeError(
                f"Cannot find stockfish on PATH or at /usr/games/stockfish. "
                "Did you install it in apt.txt or via apt-get?"
            )

        board = chess.Board(fen)
        engine = chess.engine.SimpleEngine.popen_uci(sf_bin)
        result = engine.play(board, chess.engine.Limit(time=time_limit))
        engine.quit()
        return board.san(result.move)
    
class RegexTool(Tool):
    name = "regex"
    description = (
        "Performs regex search and replace on an input string."
    )
    inputs = {
        "text": {"type": "string", "description": "Input text."},
        "pattern": {"type": "string", "description": "Regex pattern."},
        "replacement": {"type": "string", "description": "Replacement string."}
    }
    output_type = "string"

    def forward(self, text: str, pattern: str, replacement: str) -> str:
        return re.sub(pattern, replacement, text)


class MathSolverTool(Tool):
    name = "math_solver"
    description = (
        "Solves arithmetic or symbolic expressions via sympy or numpy."
    )
    inputs = {
        "expression": {"type": "string", "description": "Math expression to solve."}
    }
    output_type = "string"

    def forward(self, expression: str) -> str:
        try:
            import sympy as sp
            expr = sp.sympify(expression)
            solution = sp.solve(expr)
            return str(solution)
        except Exception:
            try:
                result = eval(expression, {"__builtins__": None}, {})
                return str(result)
            except Exception as e:
                return f"Error evaluating expression: {e}"

# Custom file reading tool
class FileReadTool(Tool):
    name = "file_reader"
    description = """
    This tool reads the content of text files.
    It's useful for processing plain text files (.txt, .csv, .json, etc).
    """
    inputs = {
        "file_path": {
            "type": "string",
            "description": "The path to the file to read",
        }
    }
    output_type = "string"
    
    def forward(self, file_path: str) -> str:
        """
        Reads the content of the given file.
        """
        try:
            # Check if the file exists
            if not os.path.exists(file_path):
                return f"Error: File not found at {file_path}"
                
            # Read the file
            with open(file_path, 'r', encoding='utf-8') as file:
                content = file.read()
                
            # If the content is too long, truncate it
            if len(content) > 10000:
                content = content[:10000] + "...\n[Text truncated due to length]"
                
            return content or "File is empty."
            
        except Exception as e:
            return f"Error reading file: {str(e)}"

class PDFReaderTool(Tool):
    name = "pdf_reader"
    description = """
    This tool extracts text content from PDF files.
    It's useful for reading research papers, reports, or other document types.
    """
    inputs = {
        "pdf_path": {
            "type": "string",
            "description": "The path to the PDF file to read",
        }
    }
    output_type = "string"
    
    def forward(self, pdf_path: str) -> str:
        """
        Extracts text from the given PDF file.
        """
        try:
            # Check if the file exists
            if not os.path.exists(pdf_path):
                return f"Error: PDF file not found at {pdf_path}"
                
            import PyPDF2
            
            # Open the PDF file
            with open(pdf_path, 'rb') as file:
                # Create a PDF reader object
                pdf_reader = PyPDF2.PdfReader(file)
                
                # Get the number of pages
                num_pages = len(pdf_reader.pages)
                
                # Extract text from all pages
                text = ""
                for page_num in range(num_pages):
                    page = pdf_reader.pages[page_num]
                    text += page.extract_text() + "\n\n"
                
                # If the text is too long, truncate it
                if len(text) > 10000:
                    text = text[:10000] + "...\n[Text truncated due to length]"
                
                return text or "No text could be extracted from the PDF."
                
        except Exception as e:
            return f"Error reading PDF: {str(e)}"

class ExcelReaderTool(Tool):
    name = "excel_reader"
    description = """
    This tool reads and processes Excel files (.xlsx, .xls).
    It can extract data, calculate statistics, and perform data analysis on spreadsheets.
    """
    inputs = {
        "excel_path": {
            "type": "string",
            "description": "The path to the Excel file to read",
        },
        "sheet_name": {
            "type": "string",
            "description": "The name of the sheet to read (optional, defaults to first sheet)",
            "nullable": True
        }
    }
    output_type = "string"
    
    def forward(self, excel_path: str, sheet_name: str = None) -> str:
        """
        Reads and processes the given Excel file.
        """
        try:
            # Check if the file exists
            if not os.path.exists(excel_path):
                return f"Error: Excel file not found at {excel_path}"
                
            import pandas as pd
            
            # Read the Excel file
            if sheet_name:
                df = pd.read_excel(excel_path, sheet_name=sheet_name)
            else:
                df = pd.read_excel(excel_path)
                
            # Get basic info about the data
            info = {
                "shape": df.shape,
                "columns": list(df.columns),
                "dtypes": df.dtypes.to_dict(),
                "head": df.head(5).to_dict()
            }
            
            # Return formatted info
            result = f"Excel file: {excel_path}\n"
            result += f"Shape: {info['shape'][0]} rows × {info['shape'][1]} columns\n\n"
            result += "Columns:\n"
            for col in info['columns']:
                result += f"- {col} ({info['dtypes'].get(col)})\n"
            
            result += "\nPreview (first 5 rows):\n"
            result += df.head(5).to_string()
            
            return result
            
        except Exception as e:
            return f"Error reading Excel file: {str(e)}"

class ImageAnalysisTool(Tool):
    name = "image_analysis"
    description = """
    This tool analyzes an image and extracts relevant information from it.
    It can describe image content, extract text from images, identify objects, etc.
    """
    inputs = {
        "image_path": {
            "type": "string",
            "description": "The path to the image file to analyze",
        }
    }
    output_type = "string"
    
    def forward(self, image_path: str) -> str:
        """
        Analyzes the given image and returns relevant information using OpenAI's ChatGPT API.
        """
        try:
            # Check if the file exists
            if not os.path.exists(image_path):
                return f"Error: Image file not found at {image_path}"
                
            import requests
            import base64
            import json
            from PIL import Image
            
            # Load the image
            with open(image_path, "rb") as image_file:
                image_bytes = image_file.read()
                
            # Convert to base64 for OpenAI API
            encoded_image = base64.b64encode(image_bytes).decode('utf-8')
            
            # Get API key from environment
            api_key = os.getenv('OPENAI_API_KEY', '')
            if not api_key:
                return "OpenAI API key not configured. Please add the OPENAI_API_KEY to your environment variables."
            
            # Prepare the API request for ChatGPT with vision capabilities
            api_url = "https://api.openai.com/v1/chat/completions"
            headers = {
                "Content-Type": "application/json",
                "Authorization": f"Bearer {api_key}"
            }
            
            payload = {
                "model": "gpt-4o-mini-2024-07-18",  # Vision-capable model
                "messages": [
                    {
                        "role": "user",
                        "content": [
                            {
                                "type": "text",
                                "text": "Analyze this image in detail. Describe what you see, including main subjects, activities, background elements, colors, and any text visible in the image. If there's text in the image, please extract it."
                            },
                            {
                                "type": "image_url",
                                "image_url": {
                                    "url": f"data:image/jpeg;base64,{encoded_image}"
                                }
                            }
                        ]
                    }
                ],
                "max_tokens": 500
            }
            
            response = requests.post(
                api_url,
                headers=headers,
                json=payload
            )
            
            if response.status_code != 200:
                return f"Error: OpenAI API returned status code {response.status_code}. Details: {response.text}"
                
            result = response.json()
            
            # Extract the response content
            if "choices" in result and len(result["choices"]) > 0:
                analysis = result["choices"][0]["message"]["content"]
                return f"Image analysis result: {analysis}"
            else:
                return f"Error: Unexpected response format from OpenAI API: {result}"
                
        except Exception as e:
            return f"Error analyzing image: {str(e)}"

# --- Basic Agent Definition ---
class BasicAgent:
    def __init__(self):
        print("BasicAgent initialized.")
        # Initialize the model
        model = OpenAIServerModel(model_id="gpt-4o-mini-2024-07-18")
        
        # Initialize tools
        self.tools = [
            YouTubeTranscriptTool(),
            SpeechToTextTool(),
            ChessEngineTool(),
            
            DuckDuckGoSearchTool(),  # Built-in web search tool
            FileReadTool(),          # Custom file reader
            PDFReaderTool(),         # PDF reader
            ExcelReaderTool(),       # Excel reader
            ImageAnalysisTool(),     # Image analysis
                   # Code execution
        ]
        
        # Initialize Agent
        self.agent = CodeAgent(
            model=model,
            tools=self.tools,
            add_base_tools=True, # Add basic tools like math
        )
        
    def __call__(self, question: str) -> str:
        print(f"Agent received question (first 50 chars): {question[:50]}...")
        try:
            answer = self.agent.run(question)
            print(f"Agent returned answer (first 50 chars): {answer[:50]}...")
            return answer
        except Exception as e:
            print(traceback.format_exc())
            error_msg = f"Error running agent: {str(e)}"
            print(error_msg)
            return f"I encountered an issue while processing your question: {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 = 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
    try:
        agent = BasicAgent()
    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
    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
        try:
            print(f"Processing task {task_id}: {question_text[:50]}...")
            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})
            print(f"Completed task {task_id}")
        except Exception as e:
             print(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:
        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, "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("# Advanced Agent Evaluation Runner")
    gr.Markdown(
        """
        **Instructions:**

        1. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
        2. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.

        ---
        **Note:**
        Once you click on the "submit" button, it may take quite some time as the agent processes all the questions.
        The agent is using SmolaAgents with multiple tools including web search, file processing, and code execution.
        """
    )

    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__":
    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 Advanced Agent Evaluation...")
    demo.launch(debug=True, share=False)