update codes

app.py

@@ -6,69 +6,103 @@ import pandas as pd
 import time
 import re
 
+# --- Course Architecture Modules ---
 from smolagents import CodeAgent, InferenceClientModel, Tool
 
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Custom Tool Definition ---
-class DuckDuckGoSearchTool(Tool):
+# --- Deep Research Tool 1: 웹 검색 및 스니펫 수집 툴 ---
+class DeepSearchTool(Tool):
     name = "web_search"
-    description = "Searches the web for a given query and returns snippet results."
-    inputs = {"query": {"type": "string", "description": "The search query"}}
+    description = "Searches the web for a given query and returns snippet results with URLs. Useful for finding initial links."
+    inputs = {"query": {"type": "string", "description": "The search query to look up"}}
     output_type = "string"
 
     def forward(self, query: str) -> str:
         try:
             url = f"https://html.duckduckgo.com/html/?q={requests.utils.quote(query)}"
-            headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"}
+            headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}
             response = requests.get(url, headers=headers, timeout=10)
             if response.status_code != 200:
-                return f"Search failed with status code {response.status_code}"
+                return f"Search temporary unavailable (Status: {response.status_code})."
             
             html = response.text
+            links = re.findall(r'<a class="result__url"[^>]* href="(.*?)"', html)
             snippets = re.findall(r'<a class="result__snippet"[^>]*>(.*?)</a>', html, re.DOTALL)
-            cleaned_snippets = []
-            for snip in snippets[:5]:
-                clean = re.sub(r'<[^>]+>', '', snip)
-                clean = clean.replace('\n', ' ').strip()
-                cleaned_snippets.append(clean)
             
-            if cleaned_snippets:
-                return "\n".join(f"- {s}" for s in cleaned_snippets)
-            return "No results found."
+            cleaned_results = []
+            for i in range(min(len(links), 4)):
+                lnk = links[i]
+                snip = re.sub(r'<[^>]+>', '', snippets[i]).replace('\n', ' ').strip() if i < len(snippets) else ""
+                cleaned_results.append(f"[{i+1}] URL: {lnk}\nSnippet: {snip}")
+                
+            return "\n\n".join(cleaned_results) if cleaned_results else "No results found."
         except Exception as e:
-            return f"Search error: {str(e)}"
+            return f"Search execution failed: {str(e)}"
+
+# --- Deep Research Tool 2: 특정 페이지 심층 텍스트 스크래핑 툴 ---
+class VisitWebpageTool(Tool):
+    name = "visit_webpage"
+    description = "Downloads and extracts all plain text content from a specific URL. Use this to read full articles, CSV contents, or text documents provided in GAIA tasks."
+    inputs = {"url": {"type": "string", "description": "The exact web URL to visit and read"}}
+    output_type = "string"
+
+    def forward(self, url: str) -> str:
+        try:
+            headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}
+            response = requests.get(url, headers=headers, timeout=15)
+            if response.status_code != 200:
+                return f"Failed to retrieve webpage. Status code: {response.status_code}"
+            
+            # HTML 태그 제거 및 순수 텍스트 추출 가공
+            text = response.text
+            text = re.sub(r'<script[^>]*>.*?</script>', '', text, flags=re.DOTALL)
+            text = re.sub(r'<style[^>]*>.*?</style>', '', text, flags=re.DOTALL)
+            text = re.sub(r'<[^>]+>', ' ', text)
+            text = re.sub(r'\s+', ' ', text).strip()
+            
+            # 컨텍스트 윈도우 오염 방지를 위해 최대 4000자 슬라이싱 제한
+            return text[:4000] if len(text) > 4000 else text
+        except Exception as e:
+            return f"Failed to visit webpage: {str(e)}"
 
 # --- Basic Agent Definition ---
-# ----- THIS IS WHERE YOU CAN BUILD WHAT YOU WANT ------
+# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
 class BasicAgent:
     def __init__(self):
+        # Open Deep Research 명세에 따른 최적의 무료 오픈소스 추론 백엔드 클라이언트 바인딩
         self.model = InferenceClientModel(
             model_id="Qwen/Qwen2.5-Coder-32B-Instruct",
             token=os.getenv("HF_TOKEN")
         )
-        self.search_tool = DuckDuckGoSearchTool()
+        self.search_tool = DeepSearchTool()
+        self.visit_tool = VisitWebpageTool()
+        
+        # 자율 컴파일 및 오류 자가 수정(Self-correction) 루프 가동을 위한 CodeAgent 패키징
         self.agent = CodeAgent(
-            tools=[self.search_tool],
+            tools=[self.search_tool, self.visit_tool],
             model=self.model,
-            max_steps=5,
-            additional_authorized_imports=["pandas", "numpy", "json", "math", "re", "datetime"]
+            max_steps=10,  # 복잡한 웹 서핑 및 데이터 교차 대조를 위해 실행 마일스톤을 10단계로 확장
+            additional_authorized_imports=["requests", "pandas", "numpy", "json", "math", "re", "datetime"]
         )
+        print("BasicAgent initialized.")
 
     def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
         try:
             refined_prompt = (
-                f"{question}\n\n"
-                f"Solve this task step by step using your tools. "
-                f"Provide the final short answer clearly at the very end."
+                f"Task: {question}\n\n"
+                f"You have deep research tools. Search information and visit specific URLs if necessary. "
+                f"Provide the final short answer clearly at the very end of your response."
             )
             result = self.agent.run(refined_prompt)
             if result is None:
                 return "unknown"
             return str(result).strip()
         except Exception as e:
-            return f"Error: {type(e).__name__}"
+            print(f"Error running agent: {e}")
+            return "unknown"
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
@@ -134,8 +168,6 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
             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})
-            
-            time.sleep(2)
         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}"})
@@ -224,4 +256,25 @@ with gr.Blocks() as demo:
     )
 
 if __name__ == "__main__":
-    demo.launch()
+    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)