Update app.py

app.py

@@ -129,19 +129,36 @@ Classify as "knowledge_base" if the query:
         logger.error(f'Error determining query type: {e}. Defaulting to knowledge_base')
         return QueryType.KNOWLEDGE_BASE
 
+def extract_urls(text: str) -> List[str]:
+    """Extract URLs from text using an improved regex pattern."""
+    # Updated regex pattern to better handle complex URLs with query parameters and paths
+    url_pattern = r'https?://(?:[-\w.]|(?:%[\da-fA-F]{2}))+(?:/[^)\s]*)?'
+    urls = re.findall(url_pattern, text)
+    
+    # Clean and validate found URLs
+    valid_urls = []
+    for url in urls:
+        # Remove trailing punctuation or artifacts that might have been captured
+        url = url.rstrip('.,;:)')
+        if is_valid_url(url):
+            valid_urls.append(url)
+    
+    return valid_urls
+
 def is_valid_url(url: str) -> bool:
-    """Check if the provided string is a valid URL."""
+    """Check if the provided string is a valid URL with enhanced validation."""
     try:
         result = urlparse(url)
-        return all([result.scheme, result.netloc])
-    except:
+        # Check for both scheme and netloc (domain)
+        has_valid_scheme = result.scheme in ('http', 'https')
+        has_valid_domain = bool(result.netloc)
+        # Additional validation to ensure complete URL structure
+        is_complete = all([has_valid_scheme, has_valid_domain])
+        return is_complete
+    except Exception as e:
+        logger.error(f'URL validation error: {e}')
         return False
 
-def extract_urls(text: str) -> List[str]:
-    """Extract URLs from text using regex pattern."""
-    url_pattern = r'https?://(?:[-\w.]|(?:%[\da-fA-F]{2}))+'
-    return re.findall(url_pattern, text)
-
 async def process_knowledge_base_query(query: str, chat_history: List[List[str]], temperature: float = 0.7) -> str:
     """Handle queries that can be answered from the knowledge base, with context."""
     logger.info(f'Processing knowledge base query: {query}')