Update app.py

app.py

@@ -1,39 +1,48 @@
 import gradio as gr
 import requests
-import feedparser
+from bs4 import BeautifulSoup
 from datetime import datetime, timedelta
+import pytz
 
 SUPPORTED_COUNTRIES = {
-    'United States': 'https://news.google.com/rss?hl=en-US&gl=US&ceid=US:en',
-    'United Kingdom': 'https://news.google.com/rss?hl=en-GB&gl=GB&ceid=GB:en',
-    'Australia': 'https://news.google.com/rss?hl=en-AU&gl=AU&ceid=AU:en',
-    'India': 'https://news.google.com/rss?hl=en-IN&gl=IN&ceid=IN:en',
-    'Canada': 'https://news.google.com/rss?hl=en-CA&gl=CA&ceid=CA:en'
+    'United States': 'US',
+    'United Kingdom': 'GB',
+    'Australia': 'AU',
+    'India': 'IN',
+    'Canada': 'CA'
 }
 
 def get_news(country, keyword):
-    if country not in SUPPORTED_COUNTRIES:
+    country_code = SUPPORTED_COUNTRIES.get(country)
+    if not country_code:
         return "Selected country is not supported."
     
-    url = SUPPORTED_COUNTRIES[country]
+    base_url = f"https://news.google.com/search?q={keyword}&hl=en-{country_code}&gl={country_code}&ceid={country_code}:en"
     
     try:
-        response = requests.get(url, timeout=10)
+        response = requests.get(base_url, timeout=10)
         response.raise_for_status()
     except requests.RequestException as e:
         return f"Error fetching news: {str(e)}"
 
-    feed = feedparser.parse(response.content)
+    soup = BeautifulSoup(response.text, 'html.parser')
+    articles = soup.find_all('article')
     
-    time_threshold = datetime.now() - timedelta(hours=48)  # 48시간으로 확장
+    time_threshold = datetime.now(pytz.utc) - timedelta(hours=48)
     
     filtered_news = []
-    for entry in feed.entries:
-        if (keyword.lower() in entry.title.lower() or 
-            ('summary' in entry and keyword.lower() in entry.summary.lower())):
-            pub_date = datetime(*entry.published_parsed[:6])
-            if pub_date > time_threshold:
-                filtered_news.append(f"Title: {entry.title}\nLink: {entry.link}\nDate: {pub_date}\n")
+    for article in articles:
+        title_elem = article.find('h3', class_='ipQwMb ekueJc RD0gLb')
+        link_elem = article.find('a', class_='VDXfz')
+        time_elem = article.find('time', class_='WW6dff uQIVzc Sksgp')
+        
+        if title_elem and link_elem and time_elem:
+            title = title_elem.text
+            link = f"https://news.google.com{link_elem['href'][1:]}"
+            pub_time = datetime.fromtimestamp(int(time_elem['datetime'])/1000, pytz.utc)
+            
+            if pub_time > time_threshold:
+                filtered_news.append(f"Title: {title}\nLink: {link}\nDate: {pub_time}\n")
     
     if filtered_news:
         return "\n".join(filtered_news)
@@ -49,7 +58,7 @@ iface = gr.Interface(
     ],
     outputs="text",
     title="Google News Search",
-    description="Search for news articles from the last 48 hours using Google News RSS feeds."
+    description="Search for news articles from the last 48 hours using Google News."
 )
 
 iface.launch()