qdrant-haystack 9.2.0__tar.gz → 9.3.0__tar.gz

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/CHANGELOG.md

@@ -1,6 +1,16 @@
 # Changelog
 
-## [integrations/qdrant-v9.1.3] - 2025-05-28
+## [integrations/qdrant-v9.2.0] - 2025-06-12
+
+### 🐛 Bug Fixes
+
+- Fix Qdrant types + add py.typed (#1919)
+
+
+### 🧹 Chores
+
+- Align core-integrations Hatch scripts (#1898)
+- Update md files for new hatch scripts (#1911)
 
 ### 🌀 Miscellaneous
 

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: qdrant-haystack
-Version: 9.2.0
+Version: 9.3.0
 Summary: An integration of Qdrant ANN vector database backend with Haystack
 Project-URL: Source, https://github.com/deepset-ai/haystack-core-integrations
 Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/qdrant/README.md
@@ -28,26 +28,11 @@ Description-Content-Type: text/markdown
 [![PyPI - Version](https://img.shields.io/pypi/v/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
 
------
+- [Integration page](https://haystack.deepset.ai/integrations/qdrant-document-store)
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/qdrant/CHANGELOG.md)
 
-**Table of Contents**
+---
 
-- [Installation](#installation)
-- [License](#license)
+## Contributing
 
-## Installation
-
-```console
-pip install qdrant-haystack
-```
-
-## Testing
-The test suites use Qdrant's in-memory instance. No additional steps required.
-
-```console
-hatch run test:all
-```
-
-## License
-
-`qdrant-haystack` is distributed under the terms of the [Apache-2.0](https://spdx.org/licenses/Apache-2.0.html) license.
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).

qdrant_haystack-9.3.0/README.md

@@ -0,0 +1,13 @@
+# qdrant-haystack
+
+[![PyPI - Version](https://img.shields.io/pypi/v/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
+
+- [Integration page](https://haystack.deepset.ai/integrations/qdrant-document-store)
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/qdrant/CHANGELOG.md)
+
+---
+
+## Contributing
+
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).

qdrant_haystack-9.3.0/pydoc/config_docusaurus.yml

@@ -0,0 +1,30 @@
+loaders:
+- ignore_when_discovered:
+  - __init__
+  modules:
+  - haystack_integrations.components.retrievers.qdrant.retriever
+  - haystack_integrations.document_stores.qdrant.document_store
+  - haystack_integrations.document_stores.qdrant.migrate_to_sparse
+  search_path:
+  - ../src
+  type: haystack_pydoc_tools.loaders.CustomPythonLoader
+processors:
+- do_not_filter_modules: false
+  documented_only: true
+  expression: null
+  skip_empty_modules: true
+  type: filter
+- type: smart
+- type: crossref
+renderer:
+  description: Qdrant integration for Haystack
+  id: integrations-qdrant
+  markdown:
+    add_member_class_prefix: false
+    add_method_class_prefix: true
+    classdef_code_block: false
+    descriptive_class_title: false
+    descriptive_module_title: true
+    filename: qdrant.md
+  title: Qdrant
+  type: haystack_pydoc_tools.renderers.DocusaurusRenderer

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/pyproject.toml

@@ -67,7 +67,7 @@ dependencies = [
 unit = 'pytest -m "not integration" {args:tests}'
 integration = 'pytest -m "integration" {args:tests}'
 all = 'pytest {args:tests}'
-cov-retry = 'all --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x'
+cov-retry = 'pytest --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x {args:tests}'
 
 types = """mypy -p haystack_integrations.document_stores.qdrant \
 -p haystack_integrations.components.retrievers.qdrant {args}"""
@@ -78,10 +78,6 @@ non_interactive = true
 check_untyped_defs = true
 disallow_incomplete_defs = true
 
-[tool.black]
-target-version = ["py38"]
-line-length = 120
-skip-string-normalization = true
 
 [tool.ruff]
 target-version = "py38"

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/src/haystack_integrations/document_stores/qdrant/document_store.py

@@ -53,9 +53,8 @@ def get_batches_from_generator(iterable: List, n: int) -> Generator:
 
 class QdrantDocumentStore:
     """
-    A QdrantDocumentStore implementation that you
-    can use with any Qdrant instance: in-memory, disk-persisted, Docker-based,
-    and Qdrant Cloud Cluster deployments.
+    A QdrantDocumentStore implementation that you can use with any Qdrant instance: in-memory, disk-persisted,
+    Docker-based, and Qdrant Cloud Cluster deployments.
 
     Usage example by creating an in-memory instance:
 
@@ -65,7 +64,8 @@ class QdrantDocumentStore:
 
     document_store = QdrantDocumentStore(
         ":memory:",
-        recreate_index=True
+        recreate_index=True,
+        embedding_dim=5
     )
     document_store.write_documents([
         Document(content="This is first", embedding=[0.0]*5),
@@ -135,6 +135,8 @@ class QdrantDocumentStore:
         payload_fields_to_index: Optional[List[dict]] = None,
     ) -> None:
         """
+        Initializes a QdrantDocumentStore.
+
         :param location:
             If `":memory:"` - use in-memory Qdrant instance.
             If `str` - use it as a URL parameter.
@@ -350,7 +352,7 @@ class QdrantDocumentStore:
         # No need to initialize client here as _get_documents_generator
         # will handle client initialization internally
 
-        self._validate_filters(filters)
+        QdrantDocumentStore._validate_filters(filters)
         return list(
             self._get_documents_generator(
                 filters,
@@ -367,7 +369,7 @@ class QdrantDocumentStore:
         # No need to initialize client here as _get_documents_generator_async
         # will handle client initialization internally
 
-        self._validate_filters(filters)
+        QdrantDocumentStore._validate_filters(filters)
         return [doc async for doc in self._get_documents_generator_async(filters)]
 
     def write_documents(
@@ -521,6 +523,108 @@ class QdrantDocumentStore:
                 "Called QdrantDocumentStore.delete_documents_async() on a non-existing ID",
             )
 
+    def delete_all_documents(self, recreate_index: bool = False) -> None:
+        """
+        Deletes all documents from the document store.
+
+        :param recreate_index: Whether to recreate the index after deleting all documents.
+        """
+
+        self._initialize_client()
+        assert self._client is not None
+
+        if recreate_index:
+            # get current collection config as json
+            collection_info = self._client.get_collection(collection_name=self.index)
+            info_json = collection_info.model_dump()
+
+            # deal with the Optional use_sparse_embeddings
+            sparse_vectors = info_json["config"]["params"]["sparse_vectors"]
+            use_sparse_embeddings = True if sparse_vectors else False
+
+            # deal with the Optional sparse_idf
+            hnsw_config = info_json["config"]["params"]["vectors"].get("config", {}).get("hnsw_config", None)
+            sparse_idf = True if use_sparse_embeddings and hnsw_config else False
+
+            # recreate collection
+            self._set_up_collection(
+                collection_name=self.index,
+                embedding_dim=info_json["config"]["params"]["vectors"]["size"],
+                recreate_collection=True,
+                similarity=info_json["config"]["params"]["vectors"]["distance"].lower(),
+                use_sparse_embeddings=use_sparse_embeddings,
+                sparse_idf=sparse_idf,
+                on_disk=info_json["config"]["hnsw_config"]["on_disk"],
+                payload_fields_to_index=info_json["payload_schema"],
+            )
+
+        else:
+            try:
+                self._client.delete(
+                    collection_name=self.index,
+                    points_selector=rest.FilterSelector(
+                        filter=rest.Filter(
+                            must=[],
+                        )
+                    ),
+                    wait=self.wait_result_from_api,
+                )
+            except Exception as e:
+                logger.warning(
+                    f"Error {e} when calling QdrantDocumentStore.delete_all_documents()",
+                )
+
+    async def delete_all_documents_async(self, recreate_index: bool = False) -> None:
+        """
+        Asynchronously deletes all documents from the document store.
+
+        :param recreate_index: Whether to recreate the index after deleting all documents.
+        """
+
+        await self._initialize_async_client()
+        assert self._async_client is not None
+
+        if recreate_index:
+            # get current collection config as json
+            collection_info = await self._async_client.get_collection(collection_name=self.index)
+            info_json = collection_info.model_dump()
+
+            # deal with the Optional use_sparse_embeddings
+            sparse_vectors = info_json["config"]["params"]["sparse_vectors"]
+            use_sparse_embeddings = True if sparse_vectors else False
+
+            # deal with the Optional sparse_idf
+            hnsw_config = info_json["config"]["params"]["vectors"].get("config", {}).get("hnsw_config", None)
+            sparse_idf = True if use_sparse_embeddings and hnsw_config else False
+
+            # recreate collection
+            await self._set_up_collection_async(
+                collection_name=self.index,
+                embedding_dim=info_json["config"]["params"]["vectors"]["size"],
+                recreate_collection=True,
+                similarity=info_json["config"]["params"]["vectors"]["distance"].lower(),
+                use_sparse_embeddings=use_sparse_embeddings,
+                sparse_idf=sparse_idf,
+                on_disk=info_json["config"]["hnsw_config"]["on_disk"],
+                payload_fields_to_index=info_json["payload_schema"],
+            )
+
+        else:
+            try:
+                await self._async_client.delete(
+                    collection_name=self.index,
+                    points_selector=rest.FilterSelector(
+                        filter=rest.Filter(
+                            must=[],
+                        )
+                    ),
+                    wait=self.wait_result_from_api,
+                )
+            except Exception as e:
+                logger.warning(
+                    f"Error {e} when calling QdrantDocumentStore.delete_all_documents_async()",
+                )
+
     @classmethod
     def from_dict(cls, data: Dict[str, Any]) -> "QdrantDocumentStore":
         """
@@ -1214,7 +1318,8 @@ class QdrantDocumentStore:
 
     def _create_payload_index(self, collection_name: str, payload_fields_to_index: Optional[List[dict]] = None) -> None:
         """
-        Create payload index for the collection if payload_fields_to_index is provided
+        Create payload index for the collection if payload_fields_to_index is provided.
+
         See: https://qdrant.tech/documentation/concepts/indexing/#payload-index
         """
         if payload_fields_to_index is not None:
@@ -1233,7 +1338,8 @@ class QdrantDocumentStore:
         self, collection_name: str, payload_fields_to_index: Optional[List[dict]] = None
     ) -> None:
         """
-        Asynchronously create payload index for the collection if payload_fields_to_index is provided
+        Asynchronously create payload index for the collection if payload_fields_to_index is provided.
+
         See: https://qdrant.tech/documentation/concepts/indexing/#payload-index
         """
         if payload_fields_to_index is not None:
@@ -1261,6 +1367,7 @@ class QdrantDocumentStore:
     ) -> None:
         """
         Sets up the Qdrant collection with the specified parameters.
+
         :param collection_name:
             The name of the collection to set up.
         :param embedding_dim:
@@ -1317,6 +1424,7 @@ class QdrantDocumentStore:
     ) -> None:
         """
         Asynchronously sets up the Qdrant collection with the specified parameters.
+
         :param collection_name:
             The name of the collection to set up.
         :param embedding_dim:
@@ -1601,7 +1709,8 @@ class QdrantDocumentStore:
 
         return vectors_config, sparse_vectors_config
 
-    def _validate_filters(self, filters: Optional[Union[Dict[str, Any], rest.Filter]] = None) -> None:
+    @staticmethod
+    def _validate_filters(filters: Optional[Union[Dict[str, Any], rest.Filter]] = None) -> None:
         """
         Validates the filters provided for querying.
 

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/tests/test_document_store.py

@@ -297,3 +297,41 @@ class TestQdrantDocumentStore(CountDocumentsTest, WriteDocumentsTest, DeleteDocu
         ):
             with pytest.raises(ValueError, match="different vector size"):
                 document_store._set_up_collection("test_collection", 768, False, "cosine", False, False)
+
+    def test_delete_all_documents_no_index_recreation(self, document_store):
+        document_store._initialize_client()
+
+        # write some documents
+        docs = [Document(id=str(i)) for i in range(5)]
+        document_store.write_documents(docs)
+
+        # delete all documents without recreating the index
+        document_store.delete_all_documents(recreate_index=False)
+        assert document_store.count_documents() == 0
+
+        # ensure the collection still exists by writing documents again
+        document_store.write_documents(docs)
+        assert document_store.count_documents() == 5
+
+    def test_delete_all_documents_index_recreation(self, document_store):
+        document_store._initialize_client()
+
+        # write some documents
+        docs = [Document(id=str(i)) for i in range(5)]
+        document_store.write_documents(docs)
+
+        # get the current document_store config
+        config_before = document_store._client.get_collection(document_store.index)
+
+        # delete all documents with recreating the index
+        document_store.delete_all_documents(recreate_index=True)
+        assert document_store.count_documents() == 0
+
+        # assure that with the same config
+        config_after = document_store._client.get_collection(document_store.index)
+
+        assert config_before.config == config_after.config
+
+        # ensure the collection still exists by writing documents again
+        document_store.write_documents(docs)
+        assert document_store.count_documents() == 5

{qdrant_haystack-9.2.0 → qdrant_haystack-9.3.0}/tests/test_document_store_async.py

@@ -218,3 +218,43 @@ class TestQdrantDocumentStore:
         ):
             with pytest.raises(ValueError, match="different similarity"):
                 await document_store._set_up_collection_async("test_collection", 768, False, "cosine", False, False)
+
+    @pytest.mark.asyncio
+    async def test_delete_all_documents_async_no_index_recreation(self, document_store):
+        await document_store._initialize_async_client()
+
+        # write some documents
+        docs = [Document(id=str(i)) for i in range(5)]
+        await document_store.write_documents_async(docs)
+
+        # delete all documents without recreating the index
+        await document_store.delete_all_documents_async(recreate_index=False)
+        assert await document_store.count_documents_async() == 0
+
+        # ensure the collection still exists by writing documents again
+        await document_store.write_documents_async(docs)
+        assert await document_store.count_documents_async() == 5
+
+    @pytest.mark.asyncio
+    async def test_delete_all_documents_async_index_recreation(self, document_store):
+        await document_store._initialize_async_client()
+
+        # write some documents
+        docs = [Document(id=str(i)) for i in range(5)]
+        await document_store.write_documents_async(docs)
+
+        # get the current document_store config
+        config_before = await document_store._async_client.get_collection(document_store.index)
+
+        # delete all documents with recreating the index
+        await document_store.delete_all_documents_async(recreate_index=True)
+        assert await document_store.count_documents_async() == 0
+
+        # assure that with the same config
+        config_after = await document_store._async_client.get_collection(document_store.index)
+
+        assert config_before.config == config_after.config
+
+        # ensure the collection still exists by writing documents again
+        await document_store.write_documents_async(docs)
+        assert await document_store.count_documents_async() == 5

qdrant_haystack-9.2.0/README.md

@@ -1,28 +0,0 @@
-# qdrant-haystack
-
-[![PyPI - Version](https://img.shields.io/pypi/v/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
-[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/qdrant-haystack.svg)](https://pypi.org/project/qdrant-haystack)
-
------
-
-**Table of Contents**
-
-- [Installation](#installation)
-- [License](#license)
-
-## Installation
-
-```console
-pip install qdrant-haystack
-```
-
-## Testing
-The test suites use Qdrant's in-memory instance. No additional steps required.
-
-```console
-hatch run test:all
-```
-
-## License
-
-`qdrant-haystack` is distributed under the terms of the [Apache-2.0](https://spdx.org/licenses/Apache-2.0.html) license.