Update app.py

app.py

@@ -2,393 +2,177 @@ import os
 import gradio as gr
 import requests
 import pandas as pd
-import re
-from typing import Optional
-import json
+from typing import Dict, List
 
+# custom imports
+from agents import Agent
+from tool import get_tools
+from model import get_model
+
+# (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+MODEL_ID = "gemini/gemini-2.5-flash-preview-04-17"
 
-# --- Enhanced GAIA Agent with Web Search ---
-class BasicAgent:
-    """
-    Enhanced agent for GAIA benchmark questions.
-    Includes web search, Wikipedia lookup, and improved reasoning.
-    """
-    
-    def __init__(self):
-        print("BasicAgent initialized with enhanced capabilities.")
-        self.knowledge_base = self._build_knowledge_base()
-        self.session = requests.Session()
-        self.session.headers.update({
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36'
-        })
-    
-    def _build_knowledge_base(self):
-        """Build knowledge base with known answers"""
+# --- Async Question Processing ---
+async def process_question(agent, question: str, task_id: str) -> Dict:
+    """Process a single question and return both answer AND full log entry"""
+    try:
+        answer = agent(question)
         return {
-            # Specific factual answers from GAIA
-            "equine_vet_agnew": {
-                "keywords": ["equine veterinarian", "chemistry materials", "marisa alviar-agnew"],
-                "answer": "Agnew"
-            },
-            "malta_olympics": {
-                "keywords": ["1928", "summer olympics", "least number", "athletes"],
-                "answer": "Malta"
-            },
-            "tsai_video": {
-                "keywords": ["1htkbjuuwec", "teal'c", "isn't that hot"],
-                "answer": "Indeed"
-            },
-            # Add more as discovered
+            "submission": {"task_id": task_id, "submitted_answer": answer},
+            "log": {"Task ID": task_id, "Question": question, "Submitted Answer": answer}
         }
+    except Exception as e:
+        error_msg = f"ERROR: {str(e)}"
+        return {
+            "submission": {"task_id": task_id, "submitted_answer": error_msg},
+            "log": {"Task ID": task_id, "Question": question, "Submitted Answer": error_msg}
+        }
+
+async def run_questions_async(agent, questions_data: List[Dict]) -> tuple:
+    """Process questions sequentially instead of in batch"""
+    submissions = []
+    logs = []
     
-    def __call__(self, question: str) -> str:
-        """
-        Main entry point for answering questions.
-        """
-        print(f"Processing: {question[:100]}...")
-        
-        # Strategy order matters - try most specific first
-        answer = (
-            self._check_knowledge_base(question) or
-            self._handle_file_questions(question) or
-            self._handle_video_questions(question) or
-            self._handle_web_search_questions(question) or
-            self._handle_wikipedia_questions(question) or
-            self._extract_numbers(question) or
-            self._handle_math(question) or
-            "Unknown"
-        )
-        
-        print(f"Answer: {answer}")
-        return answer
-    
-    def _check_knowledge_base(self, question: str) -> Optional[str]:
-        """Check knowledge base for exact matches"""
-        q_lower = question.lower()
-        
-        for key, data in self.knowledge_base.items():
-            if all(keyword in q_lower for keyword in data["keywords"]):
-                print(f"✓ Matched: {key}")
-                return data["answer"]
-        
-        return None
-    
-    def _handle_file_questions(self, question: str) -> Optional[str]:
-        """Handle questions about files (images, code, Excel, etc.)"""
-        q_lower = question.lower()
-        
-        # Questions explicitly mentioning attachments or images
-        if any(phrase in q_lower for phrase in [
-            "review the chess position",
-            "provided in the image",
-            "attached python code",
-            "attached excel file",
-            "attached file"
-        ]):
-            print("File-based question detected")
-            return "File not found"
-        
-        # Code execution questions
-        if "python code" in q_lower and "output" in q_lower:
-            return "File not found"
-        
-        # Excel/spreadsheet questions
-        if "excel file" in q_lower or "spreadsheet" in q_lower:
-            return "File not found"
-        
-        return None
-    
-    def _handle_video_questions(self, question: str) -> Optional[str]:
-        """Handle YouTube video questions"""
-        q_lower = question.lower()
-        
-        # Extract YouTube video ID
-        youtube_pattern = r'youtube\.com/watch\?v=([a-zA-Z0-9_-]+)'
-        match = re.search(youtube_pattern, question)
-        
-        if match:
-            video_id = match.group(1)
-            print(f"YouTube video detected: {video_id}")
-            
-            # Specific known answers
-            if "1htkbjuuwec" in q_lower.replace(" ", ""):
-                if "teal'c" in q_lower or "isn't that hot" in q_lower:
-                    return "Indeed"
-            
-            # Try to get video title/description (limited without API key)
-            try:
-                # Basic approach - check if question contains answer hints
-                if "say in response" in q_lower:
-                    # Common Stargate SG-1 Teal'c responses
-                    return "Indeed"
-            except Exception as e:
-                print(f"Video processing error: {e}")
-        
-        return None
-    
-    def _handle_web_search_questions(self, question: str) -> Optional[str]:
-        """Handle questions requiring web search"""
-        q_lower = question.lower()
-        
-        # Article/publication questions
-        if "article" in q_lower and "published" in q_lower:
-            # Extract date and publication
-            date_match = re.search(r'(january|february|march|april|may|june|july|august|september|october|november|december)\s+\d{1,2},?\s+\d{4}', q_lower)
-            if date_match:
-                print(f"Article question: {date_match.group(0)}")
-                # Would need actual web search here
-                return "Unknown"
-        
-        # Sports statistics questions
-        if any(word in q_lower for word in ["yankee", "pitcher", "at bats", "walks"]):
-            print("Sports statistics question")
-            # Known baseball stats - Yankees 1977
-            if "1977" in question and "walks" in q_lower:
-                # Reggie Jackson had most walks on 1977 Yankees
-                return "Unknown"  # Would need actual lookup
-        
-        return None
-    
-    def _handle_wikipedia_questions(self, question: str) -> Optional[str]:
-        """Handle Wikipedia-specific questions"""
-        q_lower = question.lower()
-        
-        if "wikipedia" in q_lower:
-            print("Wikipedia question detected")
-            # Would implement Wikipedia API search here
-            return "Unknown"
-        
-        # Questions about specific people/places/things that are likely on Wikipedia
-        if any(phrase in q_lower for phrase in [
-            "who did the actor",
-            "what country had",
-            "where were the specimens",
-            "who are the pitchers"
-        ]):
-            print("Likely Wikipedia question")
-            return "Unknown"
-        
-        return None
-    
-    def _extract_numbers(self, question: str) -> Optional[str]:
-        """Extract numerical answers"""
-        q_lower = question.lower()
-        
-        # "How many" questions
-        if "how many" in q_lower:
-            # Look for explicit numbers mentioned
-            numbers = re.findall(r'\b\d+\b', question)
-            if numbers:
-                for num in numbers:
-                    n = int(num)
-                    if 1 <= n <= 1000:  # Reasonable range
-                        return num
-        
-        return None
+    for q in questions_data:
+        result = await process_question(agent, q["question"], q["task_id"])
+        submissions.append(result["submission"])
+        logs.append(result["log"])
     
-    def _handle_math(self, question: str) -> Optional[str]:
-        """Handle mathematical calculations"""
-        try:
-            # Simple arithmetic
-            pattern = r'(\d+\.?\d*)\s*([\+\-\*\/])\s*(\d+\.?\d*)'
-            match = re.search(pattern, question)
-            
-            if match:
-                num1 = float(match.group(1))
-                op = match.group(2)
-                num2 = float(match.group(3))
-                
-                operations = {
-                    '+': lambda a, b: a + b,
-                    '-': lambda a, b: a - b,
-                    '*': lambda a, b: a * b,
-                    '/': lambda a, b: a / b if b != 0 else None
-                }
-                
-                result = operations.get(op, lambda a, b: None)(num1, num2)
-                if result is not None:
-                    return str(int(result)) if result == int(result) else str(round(result, 2))
-            
-            # Factorial
-            if "factorial" in question.lower():
-                numbers = re.findall(r'\b\d+\b', question)
-                if numbers:
-                    n = int(numbers[0])
-                    if n <= 20:
-                        result = 1
-                        for i in range(2, n + 1):
-                            result *= i
-                        return str(result)
-        
-        except Exception as e:
-            print(f"Math error: {e}")
-        
-        return None
+    return submissions, logs
 
 
-def run_and_submit_all(profile: gr.OAuthProfile | None):
+async def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
-    Fetches questions, runs agent, and submits answers.
+    Fetches all questions, runs the BasicAgent on them, submits all answers,
+    and displays the results.
     """
-    space_id = os.getenv("SPACE_ID")
+    # --- 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: {username}")
+        username= f"{profile.username}"
+        print(f"User logged in: {username}")
     else:
-        return "Please Login to Hugging Face", None
+        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. Initialize Agent
+    # 1. Instantiate Agent
     try:
-        agent = BasicAgent()
+        agent = Agent(
+            model=get_model("LiteLLMModel", MODEL_ID),
+            tools=get_tools()
+        )
     except Exception as e:
-        return f"Agent initialization error: {e}", None
-    
+        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 from: {questions_url}")
+    print(f"Fetching questions from: {questions_url}")
     try:
         response = requests.get(questions_url, timeout=15)
         response.raise_for_status()
         questions_data = response.json()
-        print(f"Fetched {len(questions_data)} questions")
-    except Exception as e:
+        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.")
+        questions_data = questions_data[:2]
+    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. Process Questions
-    results_log = []
-    answers_payload = []
-    total = len(questions_data)
-    
-    print(f"\n{'='*60}")
-    print(f"Processing {total} questions...")
-    print(f"{'='*60}\n")
-    
-    for idx, 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:
-            # Run agent
-            answer = agent(question_text)
-            answers_payload.append({
-                "task_id": task_id,
-                "submitted_answer": answer
-            })
-            results_log.append({
-                "Task ID": task_id,
-                "Question": question_text[:120] + "..." if len(question_text) > 120 else question_text,
-                "Submitted Answer": answer
-            })
-            
-            # Progress
-            if (idx + 1) % 3 == 0 or idx == total - 1:
-                print(f"Progress: {idx + 1}/{total} ({100*(idx+1)/total:.0f}%)")
-                
-        except Exception as e:
-            print(f"Error on task {task_id}: {e}")
-            results_log.append({
-                "Task ID": task_id,
-                "Question": question_text[:120] + "...",
-                "Submitted Answer": f"ERROR: {e}"
-            })
+    # 3. Run your Agent
+    print(f"Running agent on {len(questions_data)} questions...")
+    answers_payload, results_log = await run_questions_async(agent, questions_data)
 
     if not answers_payload:
-        return "No answers generated", pd.DataFrame(results_log)
+        print("Agent did not produce any answers to submit.")
+        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
 
-    # 4. Submit
-    submission_data = {
-        "username": username.strip(),
-        "agent_code": agent_code,
-        "answers": answers_payload
-    }
-    
-    print(f"\nSubmitting {len(answers_payload)} answers...")
-    
+    # 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 = response.json()
-        
-        score = result.get('score', 'N/A')
-        correct = result.get('correct_count', '?')
-        total_attempted = result.get('total_attempted', '?')
-        
-        status = (
-            f"✅ SUBMISSION SUCCESSFUL!\n\n"
-            f"User: {result.get('username')}\n"
-            f"Score: {score}% ({correct}/{total_attempted} correct)\n\n"
-            f"{result.get('message', '')}\n\n"
-            f"Leaderboard: {api_url}/leaderboard"
+        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(f"\n{'='*60}")
-        print(f"Score: {score}% ({correct}/{total_attempted})")
-        print(f"{'='*60}\n")
-        
-        return status, pd.DataFrame(results_log)
-        
+        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 = e.response.json()
-            detail = error.get('detail', e.response.text)
-        except:
-            detail = e.response.text[:500]
-        return f"❌ Submission failed: {detail}", pd.DataFrame(results_log)
-        
+            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:
-        return f"❌ Error: {e}", pd.DataFrame(results_log)
+        status_message = f"An unexpected error occurred during submission: {e}"
+        print(status_message)
+        results_df = pd.DataFrame(results_log)
+        return status_message, results_df
 
 
-# --- Gradio Interface ---
-with gr.Blocks(title="GAIA Agent", theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# 🤖 GAIA Benchmark Agent")
+# --- Build Gradio Interface using Blocks ---
+with gr.Blocks() as demo:
+    gr.Markdown("# Basic Agent Evaluation Runner")
     gr.Markdown(
         """
-        ### Enhanced Agent Features:
-        - ✓ Knowledge base for known factual questions
-        - ✓ File-based question detection
-        - ✓ YouTube video question handling
-        - ✓ Mathematical expression evaluation
-        - ✓ Web search detection (extensible)
-        - ✓ Wikipedia question detection
-        
-        ### Current Capabilities:
-        - Correctly answers: Agnew (veterinarian), Malta (Olympics), and more
-        - Handles file/image questions appropriately
-        - Processes video questions (with known answer database)
-        
-        ### To Improve Further:
-        Add API keys for: Wikipedia API, YouTube Data API, Web Search API
+        **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.
         """
     )
 
     gr.LoginButton()
-    
-    with gr.Row():
-        run_button = gr.Button("🚀 Run Evaluation", variant="primary", scale=2)
-    
-    status_output = gr.Textbox(
-        label="📊 Results",
-        lines=10,
-        interactive=False
-    )
-    
-    results_table = gr.DataFrame(
-        label="📋 Detailed Answers",
-        wrap=True,
-        max_height=400
-    )
+
+    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,
@@ -396,17 +180,25 @@ with gr.Blocks(title="GAIA Agent", theme=gr.themes.Soft()) as demo:
     )
 
 if __name__ == "__main__":
-    print("\n" + "="*70)
-    print("🤖 GAIA Agent Starting")
-    print("="*70)
-    
-    space_host = os.getenv("SPACE_HOST")
-    space_id = os.getenv("SPACE_ID")
-    
-    if space_host:
-        print(f"✅ Runtime: https://{space_host}.hf.space")
-    if space_id:
-        print(f"✅ Repo: https://huggingface.co/spaces/{space_id}")
-    
-    print("="*70 + "\n")
+    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)