Checking properly source of the file for metadata, with instanceof

services/rag/langchain/enrichment.py

@@ -118,7 +118,7 @@ def try_guess_file_type(extension: str) -> str:
 
 
 def identify_adaptive_file_source(adaptive_file: _AdaptiveFile) -> str:
-    if adaptive_file is YandexDiskAdaptiveFile:
+    if isinstance(adaptive_file, YandexDiskAdaptiveFile):
         return "Яндекс Диск"
     else:
         return "Локальный Файл"
@@ -279,9 +279,11 @@ class DocumentEnricher:
             file_hash = self._get_file_hash(local_file_path)
             if self._is_document_hash_processed(file_hash):
                 logger.info(
-                    f"Skipping already processed document hash for: {source_identifier}"
+                    f"SKIPPING already processed document hash for: {source_identifier}"
                 )
                 return
+            else:
+                logger.info("Document is not processed! Doing it")
 
             loader = self._get_loader_for_extension(local_file_path)
             if loader is None:
@@ -326,6 +328,7 @@ class DocumentEnricher:
 
             self.ADAPTIVE_FILES_QUEUE.put(adaptive_file)
 
+        logger.debug("ADAPTIVE COLLECTION DEPLETED!")
         self.collection_finished.set()
 
     # Phase 13 API: reads adaptive files and writes processed docs into PROCESSED_DOCUMENTS_QUEUE
@@ -368,8 +371,9 @@ class DocumentEnricher:
                         processed_record[0], processed_record[1]
                     )
             except Exception as error:
-                logger.error(f"Error uploading processed documents: {error}")
-                raise
+                logger.error(
+                    f"Error uploading processed documents: {error}. But swallowing error. NOT raising."
+                )
             finally:
                 self.PROCESSED_DOCUMENTS_QUEUE.task_done()