Update app.py

app.py

@@ -8,51 +8,42 @@ from bs4 import BeautifulSoup  # for web scraping
 
 from Gradio_UI import GradioUI
 
-
-@tool
-def fetch_marathon_plan(goal_time: str) -> str:
-    """Fetches a marathon training plan based on a goal time.
-    
-    Args:
-        goal_time: Desired marathon completion time (e.g., '4:00', '3:30').
-    
-    Returns:
-        A link to a training plan or summary of key details.
-    """
-    search_tool = DuckDuckGoSearchTool()
-    query = f"marathon training plan {goal_time} site:runnersworld.com"
-    results = search_tool.search(query)
-    
-    if results:
-        return f"Here is a recommended training plan for {goal_time}: {results[0]['url']}"
-    else:
-        return f"Could not find a specific plan for {goal_time}, but you can check Runner's World for more details."
-
 # Fetch marathons in Europe for 2025
 @tool
 def fetch_european_marathons_2025() -> str:
     """Fetches a list of marathons in Europe in 2025 from MarathonRunnersDiary."""
-    url = "http://www.marathonrunnersdiary.com/races/europe-marathons-list.php"
-    response = requests.get(url)
-    
-    if response.status_code != 200:
-        return "Failed to fetch marathon data."
-    
-    soup = BeautifulSoup(response.text, 'html.parser')
+    # Fetch the webpage
+URL = "http://www.marathonrunnersdiary.com/races/europe-marathons-list.php"
+page = requests.get(URL)
+soup = BeautifulSoup(page.content, "html.parser")
+
+# Find the section containing marathon events
+results = soup.find(id="holder")
+marathons = results.find_all(itemtype="http://schema.org/Event")
+
+# Extract relevant details
+marathon_list = []
+for marathon in marathons:
+    # Extract name
+    name_tag = marathon.find(itemprop="name")
+    name = name_tag.text.strip() if name_tag else "N/A"
     
-    # Locate the marathon table (this is based on the structure of the website)
-    marathons = []
-    table = soup.find('table', {'class': 'racedatatable'})  # Adjust the class if necessary
+    # Extract date
+    date_meta = marathon.find("meta", itemprop="startDate")
+    date = date_meta["content"] if date_meta else marathon.find("div", class_="leftdate").text.strip()
+
+    # Extract city and country
+    city_tag = marathon.find("div", class_="leftcity")
+    city = city_tag.text.strip() if city_tag else "N/A"
+
+    country_tag = marathon.find("div", class_="leftregion")
+    country = country_tag.text.strip() if country_tag else "N/A"
     
-    if table:
-        rows = table.find_all('tr')[1:]  # Skip the header row
-        for row in rows:
-            columns = row.find_all('td')
-            if len(columns) >= 4:
-                marathon_name = columns[0].get_text(strip=True)
-                marathon_date = columns[1].get_text(strip=True)
-                marathon_location = columns[2].get_text(strip=True)
-                marathons.append(f"{marathon_name} on {marathon_date} in {marathon_location}")
+    # Extract URL
+    link_tag = marathon.find("a")
+    link = link_tag["href"] if link_tag else "N/A"
+
+    marathon_list.append([name, date, city, country, link])
     
     if marathons:
         return "\n".join(marathons)
@@ -73,7 +64,7 @@ with open("prompts.yaml", 'r') as stream:
 
 agent = CodeAgent(
     model=model,
-    tools=[final_answer, fetch_marathon_plan, fetch_european_marathons_2025],
+    tools=[final_answer, fetch_european_marathons_2025],
     max_steps=6,
     verbosity_level=1,
     grammar=None,