Xu ly SSO

core/logging.py

@@ -0,0 +1,62 @@
+import logging
+import json
+import sys
+from datetime import datetime
+from typing import Optional
+
+
+class StructuredFormatter(logging.Formatter):
+    """JSON structured log formatter for production."""
+    
+    def format(self, record):
+        log_entry = {
+            "timestamp": datetime.utcnow().isoformat() + "Z",
+            "level": record.levelname,
+            "logger": record.name,
+            "message": record.getMessage(),
+        }
+        
+        if record.exc_info and record.exc_info[0]:
+            log_entry["exception"] = self.formatException(record.exc_info)
+        
+        if hasattr(record, "extra_data"):
+            log_entry["data"] = record.extra_data
+        
+        return json.dumps(log_entry, ensure_ascii=False)
+
+
+class HumanFormatter(logging.Formatter):
+    """Human-readable log formatter for development."""
+    
+    def format(self, record):
+        return f"{datetime.now().strftime('%H:%M:%S')} [{record.levelname}] {record.name}: {record.getMessage()}"
+
+
+def setup_logging(level: str = "INFO", structured: bool = False):
+    """
+    Setup logging cho toàn bộ ứng dụng.
+    
+    Args:
+        level: Log level (DEBUG, INFO, WARNING, ERROR)
+        structured: True = JSON format (production), False = human readable (development)
+    """
+    root_logger = logging.getLogger()
+    root_logger.setLevel(getattr(logging, level.upper(), logging.INFO))
+    
+    # Xóa existing handlers
+    root_logger.handlers.clear()
+    
+    handler = logging.StreamHandler(sys.stdout)
+    handler.setFormatter(StructuredFormatter() if structured else HumanFormatter())
+    root_logger.addHandler(handler)
+
+
+def log_event(logger: logging.Logger, level: str, message: str, **kwargs):
+    """
+    Ghi log với structured data.
+    
+    Usage:
+        log_event(logger, "info", "File processed", file_name="test.pdf", pages=5)
+    """
+    extra = {"extra_data": kwargs} if kwargs else {}
+    getattr(logger, level.lower(), logger.info)(message, extra=extra)

core/models.py

@@ -64,6 +64,7 @@ class DocumentChunk(BaseModel):
     embedding: Optional[list[float]] = None
     page_from: int
     page_to: int
-    source_url: str
+    source_url: Optional[str] = None
+    download_url: Optional[str] = None
     permissions: list[str] = []
     site_id: str = ""