improved testing

libs/astradb/langchain_astradb/graph_vectorstores.py

@@ -22,7 +22,7 @@
 from langchain_core.documents import Document
 from typing_extensions import override
 
-from langchain_astradb.utils.astradb import COMPONENT_NAME_GRAPHVECTORSTORE
+from langchain_astradb.utils.astradb import COMPONENT_NAME_GRAPHVECTORSTORE, SetupMode
 from langchain_astradb.utils.mmr_helper import MmrHelper
 from langchain_astradb.vectorstores import AstraDBVectorStore
 
@@ -33,8 +33,6 @@
     from astrapy.info import CollectionVectorServiceOptions
     from langchain_core.embeddings import Embeddings
 
-    from langchain_astradb.utils.astradb import SetupMode
-
 DEFAULT_INDEXING_OPTIONS = {"allow": ["metadata"]}
 
 
@@ -307,11 +305,43 @@ def __init__(
                 async_astra_db_client=async_astra_db_client,
             )
 
-            # # attempt a query to see if the table is setup correctly
+            # for the test search, if setup_mode is ASYNC,
+            # create a temp store with SYNC
+            if setup_mode == SetupMode.ASYNC:
+                test_vs = AstraDBVectorStore(
+                    collection_name=collection_name,
+                    embedding=embedding,
+                    token=token,
+                    api_endpoint=api_endpoint,
+                    environment=environment,
+                    namespace=namespace,
+                    metric=metric,
+                    batch_size=batch_size,
+                    bulk_insert_batch_concurrency=bulk_insert_batch_concurrency,
+                    bulk_insert_overwrite_concurrency=bulk_insert_overwrite_concurrency,
+                    bulk_delete_concurrency=bulk_delete_concurrency,
+                    setup_mode=SetupMode.SYNC,
+                    pre_delete_collection=pre_delete_collection,
+                    metadata_indexing_include=metadata_indexing_include,
+                    metadata_indexing_exclude=metadata_indexing_exclude,
+                    collection_indexing_policy=collection_indexing_policy,
+                    collection_vector_service_options=collection_vector_service_options,
+                    collection_embedding_api_key=collection_embedding_api_key,
+                    content_field=content_field,
+                    ignore_invalid_documents=ignore_invalid_documents,
+                    autodetect_collection=autodetect_collection,
+                    ext_callers=ext_callers,
+                    component_name=component_name,
+                    astra_db_client=astra_db_client,
+                    async_astra_db_client=async_astra_db_client,
+                )
+            else:
+                test_vs = self.vector_store
 
-            # self.metadata_search(filter = {
-            #     self.metadata_incoming_links_key : "test"
-            # }, n=1)
+            # try a simple search to ensure that the indexes are setup properly
+            test_vs.metadata_search(
+                filter={self.metadata_incoming_links_key: "test"}, n=1
+            )
         except BaseException as exp:
             # determine if error is because of a un-indexed column. Ref:
             # https://docs.datastax.com/en/astra-db-serverless/api-reference/collections.html#considerations-for-selective-indexing

libs/astradb/tests/conftest.py

@@ -33,7 +33,6 @@ def embed_query(self, text: str) -> list[float]:
         try:
             vals = json.loads(text)
         except json.JSONDecodeError:
-            print(f'[ParserEmbeddings] Returning a moot vector for "{text}"')
             return [0.0] * self.dimension
         else:
             assert len(vals) == self.dimension

libs/astradb/tests/integration_tests/conftest.py

@@ -66,8 +66,10 @@
 # for KMS (aka shared_secret) vectorize setup (vectorstores)
 EPHEMERAL_COLLECTION_NAME_VZ_KMS = "lc_test_vz_kms_short"
 # indexing-related collection names (function-lived) (vectorstores)
+EPHEMERAL_ALLOW_IDX_NAME_D2 = "lc_test_allow_idx_d2_short"
 EPHEMERAL_CUSTOM_IDX_NAME_D2 = "lc_test_custom_idx_d2_short"
 EPHEMERAL_DEFAULT_IDX_NAME_D2 = "lc_test_default_idx_d2_short"
+EPHEMERAL_DENY_IDX_NAME_D2 = "lc_test_deny_idx_d2_short"
 EPHEMERAL_LEGACY_IDX_NAME_D2 = "lc_test_legacy_idx_d2_short"
 # indexing-related collection names (function-lived) (storage)
 EPHEMERAL_CUSTOM_IDX_NAME = "lc_test_custom_idx_short"
@@ -515,8 +517,10 @@ def ephemeral_indexing_collections_cleaner(
     """
 
     collection_names = [
+        EPHEMERAL_ALLOW_IDX_NAME_D2,
         EPHEMERAL_CUSTOM_IDX_NAME_D2,
         EPHEMERAL_DEFAULT_IDX_NAME_D2,
+        EPHEMERAL_DENY_IDX_NAME_D2,
         EPHEMERAL_LEGACY_IDX_NAME_D2,
         EPHEMERAL_CUSTOM_IDX_NAME,
         EPHEMERAL_LEGACY_IDX_NAME,