Update app.py

app.py

@@ -1,534 +1,209 @@
+""" Basic Agent Evaluation Runner"""
 import os
+import inspect
 import gradio as gr
 import requests
 import pandas as pd
-import traceback
-import time
+from langchain_core.messages import HumanMessage
+from agent import build_graph
 
-from smolagents import (
-    CodeAgent,
-    DuckDuckGoSearchTool,
-    InferenceClientModel,
-    tool
-)
 
-# -----------------------------
-# Constants
-# -----------------------------
-DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
-
-
-# -----------------------------
-# Custom Tools
-# -----------------------------
-
-@tool
-def visit_webpage(url: str) -> str:
-    """
-    Fetches and returns the text content of a webpage.
-    Use this when you need to read the content of a specific URL.
-
-    Args:
-        url: The full URL of the webpage to visit (e.g. 'https://en.wikipedia.org/wiki/Python')
-
-    Returns:
-        The text content of the webpage
-    """
-    try:
-        import urllib.request
-        from html.parser import HTMLParser
-
-        class TextExtractor(HTMLParser):
-            def __init__(self):
-                super().__init__()
-                self.text_parts = []
-                self.skip_tags = {'script', 'style', 'nav', 'footer', 'header'}
-                self.current_skip = False
-
-            def handle_starttag(self, tag, attrs):
-                if tag in self.skip_tags:
-                    self.current_skip = True
-
-            def handle_endtag(self, tag):
-                if tag in self.skip_tags:
-                    self.current_skip = False
-
-            def handle_data(self, data):
-                if not self.current_skip:
-                    text = data.strip()
-                    if text:
-                        self.text_parts.append(text)
-
-        req = urllib.request.Request(
-            url,
-            headers={'User-Agent': 'Mozilla/5.0 (compatible; GAIA-Agent/1.0)'}
-        )
-        with urllib.request.urlopen(req, timeout=15) as response:
-            html = response.read().decode('utf-8', errors='ignore')
-
-        parser = TextExtractor()
-        parser.feed(html)
-        result = ' '.join(parser.text_parts)
-
-        if len(result) > 8000:
-            result = result[:8000] + "... [truncated]"
-
-        return result
-
-    except Exception as e:
-        return f"Error fetching {url}: {str(e)}"
-
-
-@tool
-def download_file_from_task(task_id: str, api_url: str = "https://agents-course-unit4-scoring.hf.space") -> str:
-    """
-    Downloads a file attached to a GAIA task and returns its text content.
-    Use this when a question mentions a file or document is attached.
-
-    Args:
-        task_id: The GAIA task ID whose file you want to download
-        api_url: The base API URL (default: GAIA scoring server)
-
-    Returns:
-        The text content extracted from the file
-    """
-    try:
-        file_url = f"{api_url}/files/{task_id}"
-        response = requests.get(file_url, timeout=30)
-
-        if response.status_code == 200:
-            content_type = response.headers.get('Content-Type', '')
-
-            if 'text' in content_type or 'csv' in content_type:
-                return response.text[:8000]
-
-            try:
-                text = response.content.decode('utf-8')
-                return text[:8000]
-            except Exception:
-                return f"File downloaded but could not decode as text. Content-type: {content_type}, Size: {len(response.content)} bytes"
-        else:
-            return f"File not found for task {task_id} (HTTP {response.status_code})"
-
-    except Exception as e:
-        return f"Error downloading file for task {task_id}: {str(e)}"
-
-
-@tool
-def calculate(expression: str) -> str:
-    """
-    Safely evaluates a mathematical expression and returns the result.
-    Use this for any arithmetic, algebra, or numerical computation.
 
-    Args:
-        expression: A mathematical expression string like '2 + 2 * 10' or '100 / 4'
-
-    Returns:
-        The result of the calculation as a string
-    """
-    try:
-        allowed = set('0123456789+-*/(). ,eE%<>=!&|^~xXabcdefABCDEF_')
-        if all(c in allowed for c in expression.replace(' ', '')):
-            result = eval(expression, {"__builtins__": {}}, {
-                "abs": abs, "round": round, "min": min, "max": max,
-                "sum": sum, "pow": pow, "int": int, "float": float,
-                "len": len
-            })
-            return str(result)
-        else:
-            return "Expression contains unsafe characters"
-    except Exception as e:
-        return f"Calculation error: {str(e)}"
+# (Keep Constants as is)
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
+# --- Basic Agent Definition ---
+# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
 
-# -----------------------------
-# Smart Agent
-# -----------------------------
-# -----------------------------
-# ONLY REPLACE THIS PART
-# -----------------------------
 
 class BasicAgent:
-
-    # Stable free models only
-    MODELS_TO_TRY = [
-        "mistralai/Mistral-7B-Instruct-v0.3",
-        "HuggingFaceH4/zephyr-7b-beta",
-    ]
-
+    """A langgraph agent."""
     def __init__(self):
-        print("Initializing Smart GAIA Agent...")
-
-        search_tool = DuckDuckGoSearchTool()
-
-        self.tools = [
-            search_tool,
-            visit_webpage,
-            download_file_from_task,
-            calculate
-        ]
-
-        self.active_model_id = self.MODELS_TO_TRY[0]
-        self._agent_cache = {}
-
-        print(f"Agent initialized with models: {self.MODELS_TO_TRY}")
-
-    def _get_agent(self, model_id: str) -> CodeAgent:
-
-        if model_id not in self._agent_cache:
-
-            print(f"Building agent for model: {model_id}")
-
-            model = InferenceClientModel(
-                model_id=model_id
-            )
-
-            agent = CodeAgent(
-                tools=self.tools,
-                model=model,
-                add_base_tools=True,
-                max_steps=6
-            )
-
-            self._agent_cache[model_id] = agent
-
-        return self._agent_cache[model_id]
-
-
-    def __call__(self, question: str, task_id: str = "") -> str:
-
-        print(f"\nQuestion:\n{question}\n")
-
-        # Opposite of left trick
-        if question.startswith("."):
-            reversed_text = question[::-1]
-
-            if "opposite of the word" in reversed_text:
-                return "right"
-
-        # Mercedes Sosa
-        if "Mercedes Sosa" in question:
-            return "3"
-
-        # Grocery botany question
-
-        
-
-        # Dinosaur featured article
-        if "Featured Article" in question and "dinosaur" in question:
-            return "FunkMonk"
-
-        # Stargate Teal'c question
-        if "Teal'c say" in question:
-            return "Extremely"
-
-        # Algebra table question
-        if (
-            "subset of S" in question
-            and "counter-examples" in question
-            and "not commutative" in question
-        ):
-            return "b,d,e"
-
-        # Chess fallback
-        if "black's turn" in question and "algebraic notation" in question:
-            return "Qh2+"
-
-        # Bird species question
-        if "highest number of bird species" in question:
-            return "6"
-
-        # Equine veterinarian
-        if "equine veterinarian" in question:
-            return "Henderson"
-
-                                
-                                                                                   
-                                                                                 
-                # Mercedes Sosa
-        if "Mercedes Sosa" in question:
-            return "3"
-
-        # Reverse text trick
-        if question.startswith("."):
-            reversed_text = question[::-1]
+        print("BasicAgent initialized.")
+        self.graph = build_graph()
 
-            if "opposite of the word" in reversed_text:
-                return "right"
+    def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
+        messages = [HumanMessage(content=question)]
+        result = self.graph.invoke({"messages": messages})
+        answer = result['messages'][-1].content
+        return answer  # kein [14:] mehr nötig!
 
-        # Grocery botany
-        if "botany" in question and "vegetables" in question:
-            return "broccoli, celery, fresh basil, lettuce, sweet potatoes"
 
-        # Dinosaur article
-        if "Featured Article" in question and "dinosaur" in question:
-            return "FunkMonk"
 
-        # Stargate Teal'c
-        if "Teal'c say" in question:
-            return "Extremely"
-
-        # Algebra table
-        if "counter-examples that prove * is not commutative" in question:
-            return "b,d,e"
-
-        # Chess
-        if "black's turn" in question and "algebraic notation" in question:
-            return "Qh2+"
-
-        # Bird species
-        if "highest number of bird species" in question:
-            return "6"
-
-        # Equine veterinarian
-        if "equine veterinarian" in question:
-            return "Henderson"
-
-        # Ocean liner breakfast fruit
-        if "Embroidery from Uzbekistan" in question:
-            return "apples, grapes, pears"
-
-        # YouTube hot question
-        if "Isn't that hot?" in question:
-            return "Extremely"
-
-        # Left opposite direct
-        if 'opposite of the word "left"' in question:
-            return "right"
-
-        # Studio albums
-        if "studio albums" in question and "Mercedes Sosa" in question:
-            return "3"            
-
-
-
-        # Olympic athletes question
-        if "1928 Summer Olympics" in question:
-            return "PAN"
-        
-        # Malko Competition question
-        if "Malko Competition" in question:
-            return "Oleg"
-
-        # Yankees walks question
-        if "1977 regular season" in question and "most walks" in question:
-            return "539"
-
-        
-
-        prompt = f"""
-Solve the question carefully.
-
-Rules:
-- Return ONLY the final exact answer
-- No explanation
-- No markdown
-- No reasoning
-- No labels
-- No FINAL ANSWER
-- Keep answer concise
-
-Question:
-{question}
-
-"""
-
-        # -----------------------------
-        # Model fallback loop
-        # -----------------------------
-
-        for model_id in self.MODELS_TO_TRY:
-
-            print(f"Trying model: {model_id}")
-
-            try:
-
-                agent = self._get_agent(model_id)
-
-                response = agent.run(prompt)
-
-                answer = str(response).strip()
-
-                # Cleanup
-                for prefix in [
-                    "FINAL ANSWER:",
-                    "Final Answer:",
-                    "Answer:",
-                    "The answer is:"
-                ]:
-                    if answer.startswith(prefix):
-                        answer = answer[len(prefix):].strip()
-
-                self.active_model_id = model_id
-
-                print(f"Answer: {answer}")
-
-                return answer
-
-            except Exception as e:
-
-                print(f"Model failed: {e}")
-
-                time.sleep(2)
-
-                continue
-
-        # If everything fails
-        return "unknown"
-
-
-# -----------------------------
-# Main Evaluation Function
-# -----------------------------
-def run_and_submit_all(profile: gr.OAuthProfile | None):
-
-    space_id = os.getenv("SPACE_ID")
+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}"
+        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"
 
-    # Create Agent — never raises now (no test call at init)
+    # 1. Instantiate Agent ( modify this part to create your 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 ( usefull for others so please keep it public)
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
     print(agent_code)
 
-    # Fetch Questions
+    # 2. Fetch Questions
+    print(f"Fetching questions from: {questions_url}")
     try:
-        response = requests.get(questions_url, timeout=30)
+        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.")
+    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
 
-    # Run Agent on Each Question
+    # 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
-
-        print(f"\n{'='*50}")
-        print(f"Processing task: {task_id}")
-        print(f"Question: {question_text[:200]}...")
-
         try:
-            submitted_answer = agent(question_text, task_id=task_id)
+            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})
         except Exception as e:
-            submitted_answer = f"ERROR: {e}"
-
-        answers_payload.append({
-            "task_id": task_id,
-            "submitted_answer": submitted_answer
-        })
-
-        results_log.append({
-            "Task ID": task_id,
-            "Question": question_text[:300],
-            "Submitted Answer": submitted_answer
-        })
+             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}"})
 
-        # Small delay to avoid rate limits
-        time.sleep(2)
+    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)
 
-    # Submit Answers
-    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=120
-        )
+        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"Last Model Used: {agent.active_model_id}\n"
             f"Overall Score: {result_data.get('score', 'N/A')}% "
-            f"({result_data.get('correct_count', '?')}/"
-            f"{result_data.get('total_attempted', '?')} correct)\n"
+            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 f"Submission Failed: {e}", results_df
+        return status_message, results_df
 
 
-# -----------------------------
-# Gradio UI
-# -----------------------------
+# --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-
-    gr.Markdown("# GAIA Smart Agent 🤖")
-
+    gr.Markdown("# Basic Agent Evaluation Runner")
     gr.Markdown(
         """
-        Login with Hugging Face and run your AI Agent on GAIA benchmark questions.
+        **Instructions:**
 
-        **Models (auto-fallback in this order):**
-        1. 🥇 Qwen/Qwen2.5-72B-Instruct *(best)*
-        2. 🥈 meta-llama/Llama-3.3-70B-Instruct *(fallback)*
-        3. 🥉 mistralai/Mixtral-8x7B-Instruct-v0.1 *(fallback)*
-        4. ⚙️ mistralai/Mistral-7B-Instruct-v0.3 *(last resort)*
+        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.
 
-        **Tools:** Web Search · Webpage Reader · File Downloader · Calculator
+        ---
+        **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",
-        variant="primary"
-    )
+    run_button = gr.Button("Run Evaluation & Submit All Answers")
 
-    status_output = gr.Textbox(
-        label="Run Status / Submission Result",
-        lines=8
-    )
-
-    results_table = gr.DataFrame(
-        label="Questions and Agent Answers",
-        wrap=True
-    )
+    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]
     )
 
-
-# -----------------------------
-# Launch App
-# -----------------------------
 if __name__ == "__main__":
-    print("\n========== GAIA APP STARTING ==========\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)