fix: empty delete bug (#12339)

Co-authored-by: huangzhuo <[email protected]>

api/core/rag/datasource/vdb/baidu/baidu_vector.py

@@ -113,6 +113,8 @@ def text_exists(self, id: str) -> bool:
         return False
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         quoted_ids = [f"'{id}'" for id in ids]
         self._db.table(self._collection_name).delete(filter=f"id IN({', '.join(quoted_ids)})")
 

api/core/rag/datasource/vdb/chroma/chroma_vector.py

@@ -83,6 +83,8 @@ def delete(self):
         self._client.delete_collection(self._collection_name)
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         collection = self._client.get_or_create_collection(self._collection_name)
         collection.delete(ids=ids)
 

api/core/rag/datasource/vdb/elasticsearch/elasticsearch_vector.py

@@ -98,6 +98,8 @@ def text_exists(self, id: str) -> bool:
         return bool(self._client.exists(index=self._collection_name, id=id))
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         for id in ids:
             self._client.delete(index=self._collection_name, id=id)
 

api/core/rag/datasource/vdb/myscale/myscale_vector.py

@@ -100,6 +100,8 @@ def text_exists(self, id: str) -> bool:
         return results.row_count > 0
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         self._client.command(
             f"DELETE FROM {self._config.database}.{self._collection_name} WHERE id IN {str(tuple(ids))}"
         )

api/core/rag/datasource/vdb/oceanbase/oceanbase_vector.py

@@ -134,6 +134,8 @@ def text_exists(self, id: str) -> bool:
         return bool(cur.rowcount != 0)
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         self._client.delete(table_name=self._collection_name, ids=ids)
 
     def get_ids_by_metadata_field(self, key: str, value: str) -> list[str]:

api/core/rag/datasource/vdb/oracle/oraclevector.py

@@ -167,6 +167,8 @@ def get_by_ids(self, ids: list[str]) -> list[Document]:
         return docs
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         with self._get_cursor() as cur:
             cur.execute(f"DELETE FROM {self.table_name} WHERE id IN %s" % (tuple(ids),))
 

api/core/rag/datasource/vdb/pgvector/pgvector.py

@@ -129,6 +129,11 @@ def get_by_ids(self, ids: list[str]) -> list[Document]:
         return docs
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        # Avoiding crashes caused by performing delete operations on empty lists in certain scenarios
+        # Scenario 1: extract a document fails, resulting in a table not being created.
+        # Then clicking the retry button triggers a delete operation on an empty list.
+        if not ids:
+            return
         with self._get_cursor() as cur:
             cur.execute(f"DELETE FROM {self.table_name} WHERE id IN %s", (tuple(ids),))
 

api/core/rag/datasource/vdb/tencent/tencent_vector.py

@@ -140,6 +140,8 @@ def text_exists(self, id: str) -> bool:
         return False
 
     def delete_by_ids(self, ids: list[str]) -> None:
+        if not ids:
+            return
         self._db.collection(self._collection_name).delete(document_ids=ids)
 
     def delete_by_metadata_field(self, key: str, value: str) -> None: