fix news feed

app/components/news.py

@@ -81,7 +81,7 @@ def display_news_card(news_item: dict):
                 <!-- Title -->
                 <h3 style='color: #f3f4f6; margin: 0 0 12px 0; font-size: 17px;
                            line-height: 1.5; font-weight: 600;'>
-                    {html_module.escape(news_item['summary'])}
+                    {news_item.get('summary', '').strip()}
                 </h3>
 
                 <!-- Meta info -->
@@ -245,7 +245,7 @@ def display_breaking_news_banner(df: pd.DataFrame):
                         BREAKING NEWS • {latest['source'].upper()}
                     </div>
                     <div style='color: white; font-size: 18px; font-weight: 600; line-height: 1.4;'>
-                        {html_module.escape(latest['summary'])}
+                        {latest.get('summary', '').strip()}
                     </div>
                 </div>
                 <a href='{latest['url']}' target='_blank'

app/services/news_scraper.py

@@ -239,8 +239,13 @@ class FinanceNewsScraper:
                         continue
 
                     # Skip if title looks like it contains HTML comments or code
-                    if '<!--' in title or '-->' in title or 'style=' in title:
-                        logger.debug(f"Skipping malformed title from {source_name}: {title[:100]}")
+                    if any(marker in title for marker in ['<!--', '-->', 'style=', '<div', '</div>', '<span', '</span>', 'justify-content', 'flex:', 'padding:']):
+                        logger.warning(f"Skipping malformed title from {source_name} (contains HTML): {title[:100]}...")
+                        continue
+
+                    # Skip if title is suspiciously long (likely scraped wrong element)
+                    if len(title) > 500:
+                        logger.warning(f"Skipping suspiciously long title from {source_name}: {len(title)} chars")
                         continue
 
                     # Find associated link