Update app.py

app.py

@@ -2,147 +2,72 @@ import os
 import gradio as gr
 import requests
 import pandas as pd
-import subprocess
-import sys # Import sys
+import subprocess # Needed for runtime pip install
+import sys        # Needed for runtime pip install
 
-# --- START: Force DDGS installation workaround ---
+# --- START: Force ddgs installation workaround ---
+# This block ensures 'ddgs' (which provides 'duckduckgo_search') is installed
+# early, before smolagents tries to use its DuckDuckGoSearchTool.
 try:
-    # Check if duckduckgo_search (provided by ddgs) can be imported
+    # Attempt to import duckduckgo_search to check if it's already available
     import duckduckgo_search
     print("duckduckgo_search (via ddgs) is already installed.")
 except ImportError:
     print("duckduckgo_search not found. Attempting to install ddgs...")
     try:
-        # Use ddgs as it's the updated package name
+        # Use 'ddgs' as it's the updated package name
         subprocess.check_call([sys.executable, "-m", "pip", "install", "ddgs>=4.0.0"])
         print("ddgs installed successfully.")
     except Exception as e:
         print(f"Failed to install ddgs: {e}")
-        # Critical error, re-raise to stop startup if essential dependency fails
+        # Critical error: if ddgs can't be installed, the app can't function.
         raise RuntimeError(f"CRITICAL: Failed to install ddgs: {e}")
-# --- END: Force DDGS installation workaround ---
+# --- END: Force ddgs installation workaround ---
 
-from agent import GaiaAgent # This line should now run after ddgs is confirmed installed
+# Now import the agent, as its dependencies (smolagents, duckduckgo_search) should be ready
+from agent import GaiaAgent
 
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-def run_and_submit_all(profile: gr.OAuthProfile | None):
-    space_id = os.getenv("SPACE_ID")
-
-    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"
-
-    try:
-        agent = GaiaAgent()
-    except Exception as e:
-        return f"Error initializing agent: {e}", None
-
-    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
-
-    try:
-        response = requests.get(questions_url, timeout=15)
-        response.raise_for_status()
-        questions_data = response.json()
-    except Exception as e:
-        return f"Error fetching questions: {e}", None
-
-    results_log = []
-    answers_payload = []
-
-    print("\n--- STARTING AGENT RUN ---")
-    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:
-            continue
-        try:
-            final_answer, trace = agent(question_text)
-
-            print("\n--- QUESTION ---")
-            print(f"Task ID: {task_id}")
-            print(f"Question: {question_text}")
-            print("\n--- REASONING TRACE ---")
-            print(trace)
-            print("\n--- FINAL ANSWER (SUBMITTED) ---")
-            print(final_answer)
-
-            answers_payload.append({
-                "task_id": task_id,
-                "submitted_answer": final_answer,
-                "reasoning_trace": trace
-            })
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": final_answer})
-        except Exception as e:
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"ERROR: {e}"})
-
-    if not answers_payload:
-        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
-
-    submission_data = {
-        "username": username.strip(),
-        "agent_code": agent_code,
-        "answers": answers_payload
-    }
-
+def run_agent_and_score(task_description: str) -> str:
+    # Initialize the agent within the function, so it's fresh for each run
+    # This also helps if the agent initialization is heavy or stateful
+    gaia_agent = GaiaAgent()
+    
+    # Process the task
+    agent_output = gaia_agent.process_task(task_description)
+    
+    # Send output to the scoring API
     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.')}"
+        response = requests.post(
+            f"{DEFAULT_API_URL}/score_agent",
+            json={"task_description": task_description, "agent_response": agent_output}
         )
-        results_df = pd.DataFrame(results_log)
-        return final_status, results_df
+        response.raise_for_status()  # Raise an HTTPError for bad responses (4xx or 5xx)
+        scoring_result = response.json()
+        score_info = f"Scoring Result:\nTotal Score: {scoring_result.get('total_score')}\nCorrectness Score: {scoring_result.get('correctness_score')}\nExplanation: {scoring_result.get('explanation', 'No explanation provided.')}"
+        return f"Agent Output:\n{agent_output}\n\n---\n\n{score_info}"
+    except requests.exceptions.RequestException as e:
+        return f"Agent Output:\n{agent_output}\n\n---\n\nError connecting to scoring API: {e}"
     except Exception as e:
-        return f"Submission Failed: {e}", pd.DataFrame(results_log)
-
+        return f"Agent Output:\n{agent_output}\n\n---\n\nAn unexpected error occurred during scoring: {e}"
 
+# Gradio Interface setup
 with gr.Blocks() as demo:
-    gr.Markdown("# GAIA Agent Submission Interface")
-    gr.Markdown("""
-        Logga in och kör agenten.\n
-        Du behöver INTE en OpenAI API-nyckel längre. Agenten kör en lokal modell.
-    """)
-    gr.LoginButton()
-
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
-    status_output = gr.Textbox(label="Submission Result")
-    results_table = gr.DataFrame(label="Answers")
-
-    run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
-
-if __name__ == "__main__":
-    print("\n" + "-"*30 + " App Starting " + "-"*30)
-    space_host_startup = os.getenv("SPACE_HOST")
-    space_id_startup = os.getenv("SPACE_ID")
+    gr.Markdown("# GAIA Basic Agent Evaluator (Freddolin)")
+    gr.Markdown("Enter a task description for the agent to process. The agent's output will be displayed, followed by its score from the GAIA scoring API.")
 
-    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?).")
+    task_input = gr.Textbox(label="Task Description", placeholder="e.g., 'What is the capital of France?'")
+    output_text = gr.Textbox(label="Agent Output & Score", interactive=False)
 
-    if space_id_startup:
-        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.")
+    run_button = gr.Button("Run Agent & Score")
 
-    print("-"*(60 + len(" App Starting ")) + "\n")
+    run_button.click(
+        fn=run_agent_and_score,
+        inputs=task_input,
+        outputs=output_text
+    )
 
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+# Launch the Gradio app
+demo.launch(debug=True) # debug=True can provide more info in logs during development