Update app.py

app.py

@@ -1,10 +1,9 @@
-from smolagents import CodeAgent,DuckDuckGoSearchTool, HfApiModel,load_tool,tool
+from smolagents import CodeAgent,DuckDuckGoSearchTool,HfApiModel,load_tool,tool
 import datetime
 import requests
 import pytz
 import yaml
 from tools.final_answer import FinalAnswerTool
-from bs4 import BeautifulSoup
 
 from Gradio_UI import GradioUI
 
@@ -21,65 +20,42 @@ def my_cutom_tool(arg1:str, arg2:int)-> str: #it's import to specify the return
 
 # Tool to fetch latest news headlines from different news outlets
 @tool
-def get_latest_news(source: str) -> str:
-    """Fetches latest headlines from specified news source.
+def search_news_headlines(source: str) -> str:
+    """Searches for recent news headlines from specified source using DuckDuckGo.
     
     Args:
-        source: Name of news source from supported list (BBC News, CNN, Reuters, AP, 
-               Fox News, CBS News, Wall Street Journal, Daily Mail)
+        source: News source from supported list (BBC, CNN, etc.)
     
     Returns:
-        String containing latest headlines or error message
+        Formatted news results or error message
     """
-    news_urls = {
-        "BBC News": "https://www.bbc.com/news",
-        "CNN": "https://www.cnn.com",
-        "Reuters": "https://www.reuters.com",
-        "AP": "https://apnews.com",
-        "Fox News": "https://www.foxnews.com",
-        "CBS News": "https://www.cbsnews.com",
-        "Wall Street Journal": "https://www.wsj.com",
-        "Daily Mail": "https://www.dailymail.co.uk/home/index.html"
+    domain_map = {
+        "BBC News": "bbc.com",
+        "CNN": "cnn.com",
+        "Reuters": "reuters.com",
+        "AP": "apnews.com",
+        "Fox News": "foxnews.com",
+        "CBS News": "cbsnews.com",
+        "Wall Street Journal": "wsj.com",
+        "Daily Mail": "dailymail.co.uk"
     }
-
+    
     try:
-        # Get the URL for the requested source
-        url = news_urls.get(source)
-        if not url:
+        domain = domain_map.get(source)
+        if not domain:
             return f"Error: {source} is not a supported news source"
         
-        # Fetch and parse the page
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
-        }
-        response = requests.get(url, headers=headers, timeout=10)
-        response.raise_for_status()
-        soup = BeautifulSoup(response.text, 'html.parser')
+        # Create search query with site filter and time filter
+        search_query = f"site:{domain} latest news after:2023-01-01"
         
-        # Extract headlines based on source-specific patterns
-        headlines = []
-        if source == "BBC News":
-            headlines = [h3.text.strip() for h3 in soup.select('h3[class^="gs-c-promo-heading"]')][:5]
-        elif source == "CNN":
-            headlines = [div.text.strip() for div in soup.select('div[class^="container__headline"]')][:5]
-        elif source == "Reuters":
-            headlines = [a.text.strip() for a in soup.select('a[data-testid="Heading"]')][:5]
-        elif source == "AP":
-            headlines = [a.text.strip() for a in soup.select('a[class*="Component-headline"]')][:5]
-        elif source == "Fox News":
-            headlines = [h2.text.strip() for h2 in soup.select('h2[class*="title"]')][:5]
-        elif source == "CBS News":
-            headlines = [h3.text.strip() for h3 in soup.select('h3[class*="item__hed"]')][:5]
-        elif source == "Wall Street Journal":
-            headlines = [h3.text.strip() for h3 in soup.select('h3[class*="WSJTheme--headline"]')][:5]
-        elif source == "Daily Mail":
-            headlines = [h2.text.strip() for h2 in soup.select('h2[class*="linkro-darkred"]')][:5]
-        # Add other sources' parsing logic here...
-        
-        return f"Latest headlines from {source}:\n- " + "\n- ".join(headlines[:5])
+        # Use existing DuckDuckGoSearchTool implementation
+        ddg_tool = DuckDuckGoSearchTool(max_results=5)
+        results = ddg_tool.forward(search_query)
         
+        return f"Recent news from {source}:\n{results}"
+    
     except Exception as e:
-        return f"Error fetching news from {source}: {str(e)}"
+        return f"News search failed: {str(e)}"
 
 
 @tool
@@ -124,7 +100,7 @@ with open("prompts.yaml", 'r') as stream:
     
 agent = CodeAgent(
     model=model,
-    tools=[final_answer, get_latest_news], ## add your tools here (don't remove final answer)
+    tools=[final_answer, search_news_headlines], ## add your tools here (don't remove final answer)
     max_steps=6,
     verbosity_level=0,
     grammar=None,