qxqx

libs/core/langchain_core/indexing/base_index.py

@@ -18,6 +18,7 @@ from typing import (
 from langchain_core._api import beta
 from langchain_core.documents.base import BaseMedia
 from langchain_core.indexing.base import DeleteResponse, UpsertResponse
+from langchain_core.runnables import run_in_executor
 from langchain_core.utils import abatch_iterate, batch_iterate
 
 
@@ -153,6 +154,12 @@ class BaseIndex(Generic[T]):
 
         .. versionadded:: 0.2.11
         """
+        return await run_in_executor(
+            None,
+            self.upsert,
+            items,
+            **kwargs,
+        )
 
     @abc.abstractmethod
     def get_by_ids(
@@ -169,31 +176,3 @@ class BaseIndex(Generic[T]):
         /,
     ) -> DeleteResponse:
         """Delete items by id."""
-
-    # Delete and get are part of the READ/WRITE interface.
-    # They do not take advantage of indexes on the content.
-    # However, all the indexers ARE assumed to have the capability to index
-    # on metadata if they implement the delete_by_query and get_by_query methods.
-    # @abc.abstractmethod
-    # def get_by_query(
-    #     self,
-    #     *,
-    #     filter: Optional[Union[List[Dict[str, Any], Dict[str, Any]]]] = None,
-    #     limit: Optional[int] = None,
-    #     offset: Optional[int] = None,
-    #     sort: Optional[Union[Sort, List[Sort]]] = None,
-    #     **kwargs: Any,
-    # ) -> Iterable[T]:
-    #     """Get items by query."""
-    #
-    # @abc.abstractmethod
-    # def delete_by_query(
-    #     self,
-    #     *,
-    #     filter: Optional[Union[List[Dict[str, Any], Dict[str, Any]]]] = None,
-    #     limit: Optional[int] = None,
-    #     offset: Optional[int] = None,
-    #     sort: Optional[Union[Sort, List[Sort]]] = None,
-    #     **kwargs: Any,
-    # ) -> Iterable[DeleteResponse]:
-    #     """Delete items by query."""

libs/core/tests/unit_tests/example_selectors/test_similarity.py

@@ -1,4 +1,4 @@
-from typing import Any, Iterable, List, Optional, cast
+from typing import Any, Iterable, List, Optional, Sequence, cast
 
 from langchain_core.documents import Document
 from langchain_core.embeddings import Embeddings, FakeEmbeddings
@@ -6,6 +6,7 @@ from langchain_core.example_selectors import (
     MaxMarginalRelevanceExampleSelector,
     SemanticSimilarityExampleSelector,
 )
+from langchain_core.indexing.base import DeleteResponse
 from langchain_core.vectorstores import VectorStore
 
 
@@ -31,6 +32,16 @@ class DummyVectorStore(VectorStore):
             self.metadatas.extend(metadatas)
         return ["dummy_id"]
 
+    def get_by_ids(self, ids: Sequence[str], /) -> List[Document]:
+        raise NotImplementedError()
+
+    def delete_by_ids(
+        self,
+        ids: Sequence[str],
+        /,
+    ) -> DeleteResponse:
+        raise NotImplementedError()
+
     def similarity_search(
         self, query: str, k: int = 4, **kwargs: Any
     ) -> List[Document]:

libs/core/tests/unit_tests/indexing/test_indexing.py

@@ -6,7 +6,7 @@ from typing import (
     Iterator,
     Sequence,
 )
-from unittest.mock import MagicMock, patch, AsyncMock
+from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
 import pytest_asyncio

libs/core/tests/unit_tests/vectorstores/test_vectorstore.py

@@ -7,7 +7,7 @@ from typing_extensions import TypedDict
 
 from langchain_core.documents import Document
 from langchain_core.embeddings import Embeddings
-from langchain_core.indexing.base import UpsertResponse
+from langchain_core.indexing.base import DeleteResponse, UpsertResponse
 from langchain_core.vectorstores import VectorStore
 
 
@@ -68,6 +68,18 @@ class CustomSyncVectorStore(VectorStore):
     def get_by_ids(self, ids: Sequence[str], /) -> List[Document]:
         return [self.store[id] for id in ids if id in self.store]
 
+    def delete_by_ids(
+        self,
+        ids: Sequence[str],
+        /,
+    ) -> DeleteResponse:
+        for id_ in ids:
+            self.store.pop(id_, None)
+        return {
+            "succeeded": ids,
+            "failed": [],
+        }
+
     def from_texts(  # type: ignore
         cls,
         texts: List[str],