Update app.py

app.py

@@ -3,103 +3,130 @@ import gradio as gr
 import requests
 import pandas as pd
 
+# -------------------------------
+# Constants
+# -------------------------------
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# -------------------------
-# Basic GAIA Agent
-# -------------------------
+# -------------------------------
+# Basic Agent (可以換成你的 GAIA Agent)
+# -------------------------------
 class BasicAgent:
     def __init__(self):
         print("BasicAgent initialized.")
 
     def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
+        # TODO: 這裡改成你的 GAIA Agent 邏輯
         return "This is a default answer."
 
-# -------------------------
+# -------------------------------
 # Run & Submit Function
-# -------------------------
-def run_and_submit_all(profile):
-    if not profile:
+# -------------------------------
+def run_and_submit_all(profile=None):
+    """
+    Fetch questions, run the agent, submit answers, return results
+    profile: Either HF profile (from LoginButton) or mock_profile dict
+    """
+    # --- HF Username ---
+    if profile and hasattr(profile, "username"):
+        username = profile.username
+    elif isinstance(profile, dict) and "username" in profile:
+        username = profile["username"]
+    else:
         return "❌ Please login with your Hugging Face account.", None
 
-    username = profile.get("username", "unknown_user")
-    print(f"Logged in as: {username}")
-
-    space_id = os.getenv("SPACE_ID", "unknown-space")
-    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
-
-    # Instantiate Agent
+    # --- Agent & Code Link ---
     agent = BasicAgent()
+    space_id = os.getenv("SPACE_ID", "unknown")
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
 
-    # Fetch questions
+    # --- Fetch Questions ---
     try:
-        resp = requests.get(f"{DEFAULT_API_URL}/questions", timeout=15)
-        resp.raise_for_status()
-        questions = resp.json()
-        if not questions:
-            return "No questions fetched.", pd.DataFrame()
+        response = requests.get(f"{DEFAULT_API_URL}/questions", timeout=15)
+        response.raise_for_status()
+        questions_data = response.json()
     except Exception as e:
-        return f"Error fetching questions: {e}", pd.DataFrame()
+        return f"❌ Failed to fetch questions: {e}", None
 
-    # Run Agent
+    if not questions_data:
+        return "❌ No questions fetched.", None
+
+    # --- Run Agent ---
     results_log = []
     answers_payload = []
-    for q in questions:
-        task_id = q.get("task_id")
-        question_text = q.get("question")
-        if not task_id or question_text is None:
+    for item in questions_data:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        if not task_id or not question_text:
             continue
         try:
-            answer = agent(question_text)
+            ans = agent(question_text)
+            answers_payload.append({"task_id": task_id, "submitted_answer": ans})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text,
+                "Submitted Answer": ans
+            })
         except Exception as e:
-            answer = f"AGENT ERROR: {e}"
-        answers_payload.append({"task_id": task_id, "submitted_answer": answer})
-        results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": answer})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text,
+                "Submitted Answer": f"AGENT ERROR: {e}"
+            })
+
+    if not answers_payload:
+        return "❌ Agent did not produce any answers.", pd.DataFrame(results_log)
+
+    # --- Submit Answers ---
+    submission_data = {
+        "username": username.strip(),
+        "agent_code": agent_code,
+        "answers": answers_payload
+    }
 
-    # Submit answers
-    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
     try:
         resp = requests.post(f"{DEFAULT_API_URL}/submit", json=submission_data, timeout=60)
         resp.raise_for_status()
-        result = resp.json()
-        status = (
-            f"✅ Submission Complete!\n"
-            f"User: {result.get('username')}\n"
-            f"Score: {result.get('score', 'N/A')}% "
-            f"({result.get('correct_count', '?')}/{result.get('total_attempted', '?')} correct)\n"
-            f"Message: {result.get('message', 'No message')}"
+        result_data = resp.json()
+        status_text = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"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', '')}"
         )
+        return status_text, pd.DataFrame(results_log)
     except Exception as e:
-        status = f"❌ Submission failed: {e}"
-
-    return status, pd.DataFrame(results_log)
+        return f"❌ Submission failed: {e}", pd.DataFrame(results_log)
 
-# -------------------------
+# -------------------------------
 # Gradio Interface
-# -------------------------
+# -------------------------------
 with gr.Blocks() as demo:
-    gr.Markdown("# 🎯 GAIA Agent Evaluation")
+    gr.Markdown("# Basic GAIA Agent Evaluation Runner")
     gr.Markdown(
         """
-        **Instructions:**
-        1. Log in with your Hugging Face account.
-        2. Click 'Run Evaluation & Submit All Answers' to fetch questions and submit your agent's answers.
+        1. Log in with Hugging Face (Space only) or test locally with mock user.
+        2. Click 'Run Evaluation & Submit All Answers'.
         """
     )
 
-    # 使用 gr.State() 存使用者資訊
-    user_state = gr.State()
-
-    # Login button，沒有任何參數
+    # HF LoginButton only works in HF Space
     login_btn = gr.LoginButton()
-    login_btn.click(lambda profile: profile, inputs=None, outputs=user_state)
 
-    run_btn = gr.Button("🚀 Run Evaluation & Submit All Answers")
-    status_box = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
+    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使用user_state當profile
-    run_btn.click(run_and_submit_all, inputs=user_state, outputs=[status_box, results_table])
+    # For HF Space, login_btn gives profile; locally we can pass mock_profile
+    run_button.click(
+        fn=lambda profile: run_and_submit_all(profile or {"username": "local_user"}),
+        inputs=login_btn,
+        outputs=[status_output, results_table]
+    )
 
 if __name__ == "__main__":
+    print("Launching Gradio Interface...")
     demo.launch(debug=True, share=False)