修改了document exists调用的判断逻辑，从doc_id变成了doc_id_chunk_0

sync/folder_sync.py

@@ -14,7 +14,7 @@ from llama_index.core import Document
 class FolderSync(BaseSync):
     """Handle synchronization between folder (local or remote) and ChromaDB"""
     
-    def __init__(self, config: BaseDataSourceConfig):
+    def __init__(self, config: BaseDataSourceConfig, vector_store_manager=None):
         """
         Initialize folder sync with configuration
         
@@ -25,6 +25,7 @@ class FolderSync(BaseSync):
         self.file_parser = FileParser()
         self._ssh_client = None
         self._sftp_client = None
+        self.vector_store_manager = vector_store_manager
     
     def fetch_all_documents(self, last_sync_time=None) -> List[Dict[str, Any]]:
         """
@@ -55,7 +56,7 @@ class FolderSync(BaseSync):
                     
                     if last_sync_time is not None:
                         # Check if document has already been synced
-                        if self.vector_store_manager and self.vector_store_manager.document_exists(doc_id):
+                        if self.vector_store_manager and self.vector_store_manager.document_exists(doc_id + "_chunk_0"):
                             # If document is already synced, check if it's been modified since last sync
                             if last_sync_time:
                                 file_stat = self._sftp_client.stat(file_path)

sync_service.py

@@ -51,7 +51,7 @@ class SyncService:
             sync_class = get_sync_class(self.source_config.type)
             
             # Create a new syncer instance for this data source
-            self.syncer = sync_class(self.source_config, vector_store_manager=self.vector_store_manager)
+            self.syncer = sync_class(self.source_config, self.vector_store_manager)
             
             # Initialize sync tracking data
             self.last_sync_time = None