refactor(qdrant): use async DocumentStore mixin tests

integrations/qdrant/pyproject.toml

@@ -28,7 +28,7 @@ classifiers = [
   "Programming Language :: Python :: Implementation :: PyPy",
 ]
 dependencies = [
-  "haystack-ai>=2.26.1",
+  "haystack-ai>=2.28.0",
   "qdrant-client>=1.12.0"
 ]
 

integrations/qdrant/src/haystack_integrations/document_stores/qdrant/document_store.py

@@ -269,6 +269,12 @@ def _initialize_client(self) -> None:
             client_params = self._prepare_client_params()
             # This step adds the api-key and User-Agent to metadata
             self._client = qdrant_client.QdrantClient(**client_params)
+            # For in-memory stores: if the async client was already initialised,
+            # share its underlying collections dict so both clients see the same data.
+            if self.location == ":memory:" and self._async_client is not None:
+                self._client._client.collections = self._async_client._client.collections  # type: ignore[attr-defined]
+                self._client._client.aliases = self._async_client._client.aliases  # type: ignore[attr-defined]
+                return
             # Make sure the collection is properly set up
             self._set_up_collection(
                 self.index,
@@ -290,6 +296,12 @@ async def _initialize_async_client(self) -> None:
             self._async_client = qdrant_client.AsyncQdrantClient(
                 **client_params,
             )
+            # For in-memory stores: if the sync client was already initialised,
+            # share its underlying collections dict so both clients see the same data.
+            if self.location == ":memory:" and self._client is not None:
+                self._async_client._client.collections = self._client._client.collections  # type: ignore[attr-defined]
+                self._async_client._client.aliases = self._client._client.aliases  # type: ignore[attr-defined]
+                return
             await self._set_up_collection_async(
                 self.index,
                 self.embedding_dim,