qdrant-haystack 3.5.0__py3-none-any.whl → 3.7.0__py3-none-any.whl

haystack_integrations/components/retrievers/qdrant/retriever.py

@@ -1,8 +1,9 @@
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from haystack import Document, component, default_from_dict, default_to_dict
 from haystack.dataclasses.sparse_embedding import SparseEmbedding
 from haystack_integrations.document_stores.qdrant import QdrantDocumentStore
+from qdrant_client.http import models
 
 
 @component
@@ -12,6 +13,7 @@ class QdrantEmbeddingRetriever:
 
     Usage example:
     ```python
+    from haystack.dataclasses import Document
     from haystack_integrations.components.retrievers.qdrant import QdrantEmbeddingRetriever
     from haystack_integrations.document_stores.qdrant import QdrantDocumentStore
 
@@ -33,7 +35,7 @@ class QdrantEmbeddingRetriever:
     def __init__(
         self,
         document_store: QdrantDocumentStore,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: int = 10,
         scale_score: bool = True,
         return_embedding: bool = False,
@@ -42,12 +44,12 @@ class QdrantEmbeddingRetriever:
         Create a QdrantEmbeddingRetriever component.
 
         :param document_store: An instance of QdrantDocumentStore.
-        :param filters: A dictionary with filters to narrow down the search space. Default is None.
-        :param top_k: The maximum number of documents to retrieve. Default is 10.
-        :param scale_score: Whether to scale the scores of the retrieved documents or not. Default is True.
-        :param return_embedding: Whether to return the embedding of the retrieved Documents. Default is False.
+        :param filters: A dictionary with filters to narrow down the search space.
+        :param top_k: The maximum number of documents to retrieve.
+        :param scale_score: Whether to scale the scores of the retrieved documents or not.
+        :param return_embedding: Whether to return the embedding of the retrieved Documents.
 
-        :raises ValueError: If 'document_store' is not an instance of QdrantDocumentStore.
+        :raises ValueError: If `document_store` is not an instance of `QdrantDocumentStore`.
         """
 
         if not isinstance(document_store, QdrantDocumentStore):
@@ -97,7 +99,7 @@ class QdrantEmbeddingRetriever:
     def run(
         self,
         query_embedding: List[float],
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: Optional[int] = None,
         scale_score: Optional[bool] = None,
         return_embedding: Optional[bool] = None,
@@ -134,7 +136,7 @@ class QdrantSparseEmbeddingRetriever:
     ```python
     from haystack_integrations.components.retrievers.qdrant import QdrantSparseEmbeddingRetriever
     from haystack_integrations.document_stores.qdrant import QdrantDocumentStore
-    from haystack.dataclasses.sparse_embedding import SparseEmbedding
+    from haystack.dataclasses import Document, SparseEmbedding
 
     document_store = QdrantDocumentStore(
         ":memory:",
@@ -155,7 +157,7 @@ class QdrantSparseEmbeddingRetriever:
     def __init__(
         self,
         document_store: QdrantDocumentStore,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: int = 10,
         scale_score: bool = True,
         return_embedding: bool = False,
@@ -164,12 +166,12 @@ class QdrantSparseEmbeddingRetriever:
         Create a QdrantSparseEmbeddingRetriever component.
 
         :param document_store: An instance of QdrantDocumentStore.
-        :param filters: A dictionary with filters to narrow down the search space. Default is None.
-        :param top_k: The maximum number of documents to retrieve. Default is 10.
-        :param scale_score: Whether to scale the scores of the retrieved documents or not. Default is True.
-        :param return_embedding: Whether to return the sparse embedding of the retrieved Documents. Default is False.
+        :param filters: A dictionary with filters to narrow down the search space.
+        :param top_k: The maximum number of documents to retrieve.
+        :param scale_score: Whether to scale the scores of the retrieved documents or not.
+        :param return_embedding: Whether to return the sparse embedding of the retrieved Documents.
 
-        :raises ValueError: If 'document_store' is not an instance of QdrantDocumentStore.
+        :raises ValueError: If `document_store` is not an instance of `QdrantDocumentStore`.
         """
 
         if not isinstance(document_store, QdrantDocumentStore):
@@ -219,7 +221,7 @@ class QdrantSparseEmbeddingRetriever:
     def run(
         self,
         query_sparse_embedding: SparseEmbedding,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: Optional[int] = None,
         scale_score: Optional[bool] = None,
         return_embedding: Optional[bool] = None,
@@ -257,7 +259,7 @@ class QdrantHybridRetriever:
     ```python
     from haystack_integrations.components.retrievers.qdrant import QdrantHybridRetriever
     from haystack_integrations.document_stores.qdrant import QdrantDocumentStore
-    from haystack.dataclasses.sparse_embedding import SparseEmbedding
+    from haystack.dataclasses import Document, SparseEmbedding
 
     document_store = QdrantDocumentStore(
         ":memory:",
@@ -283,7 +285,7 @@ class QdrantHybridRetriever:
     def __init__(
         self,
         document_store: QdrantDocumentStore,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: int = 10,
         return_embedding: bool = False,
     ):
@@ -341,7 +343,7 @@ class QdrantHybridRetriever:
         self,
         query_embedding: List[float],
         query_sparse_embedding: SparseEmbedding,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], models.Filter]] = None,
         top_k: Optional[int] = None,
         return_embedding: Optional[bool] = None,
     ):

haystack_integrations/document_stores/qdrant/document_store.py

@@ -5,7 +5,6 @@ from typing import Any, ClassVar, Dict, Generator, List, Optional, Set, Union
 
 import numpy as np
 import qdrant_client
-from grpc import RpcError
 from haystack import default_from_dict, default_to_dict
 from haystack.dataclasses import Document
 from haystack.dataclasses.sparse_embedding import SparseEmbedding
@@ -66,7 +65,7 @@ class QdrantDocumentStore:
         https: Optional[bool] = None,
         api_key: Optional[Secret] = None,
         prefix: Optional[str] = None,
-        timeout: Optional[float] = None,
+        timeout: Optional[int] = None,
         host: Optional[str] = None,
         path: Optional[str] = None,
         index: str = "Document",
@@ -96,23 +95,7 @@ class QdrantDocumentStore:
         scroll_size: int = 10_000,
         payload_fields_to_index: Optional[List[dict]] = None,
     ):
-        super().__init__()
-
-        metadata = metadata or {}
-        self.client = qdrant_client.QdrantClient(
-            location=location,
-            url=url,
-            port=port,
-            grpc_port=grpc_port,
-            prefer_grpc=prefer_grpc,
-            https=https,
-            api_key=api_key.resolve_value() if api_key else None,
-            prefix=prefix,
-            timeout=timeout,
-            host=host,
-            path=path,
-            metadata=metadata,
-        )
+        self._client = None
 
         # Store the Qdrant client specific attributes
         self.location = location
@@ -126,7 +109,7 @@ class QdrantDocumentStore:
         self.timeout = timeout
         self.host = host
         self.path = path
-        self.metadata = metadata
+        self.metadata = metadata or {}
         self.api_key = api_key
 
         # Store the Qdrant collection specific attributes
@@ -143,12 +126,6 @@ class QdrantDocumentStore:
         self.recreate_index = recreate_index
         self.payload_fields_to_index = payload_fields_to_index
         self.use_sparse_embeddings = use_sparse_embeddings
-
-        # Make sure the collection is properly set up
-        self._set_up_collection(
-            index, embedding_dim, recreate_index, similarity, use_sparse_embeddings, on_disk, payload_fields_to_index
-        )
-
         self.embedding_dim = embedding_dim
         self.on_disk = on_disk
         self.content_field = content_field
@@ -162,6 +139,35 @@ class QdrantDocumentStore:
         self.write_batch_size = write_batch_size
         self.scroll_size = scroll_size
 
+    @property
+    def client(self):
+        if not self._client:
+            self._client = qdrant_client.QdrantClient(
+                location=self.location,
+                url=self.url,
+                port=self.port,
+                grpc_port=self.grpc_port,
+                prefer_grpc=self.prefer_grpc,
+                https=self.https,
+                api_key=self.api_key.resolve_value() if self.api_key else None,
+                prefix=self.prefix,
+                timeout=self.timeout,
+                host=self.host,
+                path=self.path,
+                metadata=self.metadata,
+            )
+            # Make sure the collection is properly set up
+            self._set_up_collection(
+                self.index,
+                self.embedding_dim,
+                self.recreate_index,
+                self.similarity,
+                self.use_sparse_embeddings,
+                self.on_disk,
+                self.payload_fields_to_index,
+            )
+        return self._client
+
     def count_documents(self) -> int:
         try:
             response = self.client.count(
@@ -176,13 +182,13 @@ class QdrantDocumentStore:
 
     def filter_documents(
         self,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], rest.Filter]] = None,
     ) -> List[Document]:
-        if filters and not isinstance(filters, dict):
-            msg = "Filter must be a dictionary"
+        if filters and not isinstance(filters, dict) and not isinstance(filters, rest.Filter):
+            msg = "Filter must be a dictionary or an instance of `qdrant_client.http.models.Filter`"
             raise ValueError(msg)
 
-        if filters and "operator" not in filters:
+        if filters and not isinstance(filters, rest.Filter) and "operator" not in filters:
             filters = convert_legacy_filters(filters)
         return list(
             self.get_documents_generator(
@@ -260,7 +266,7 @@ class QdrantDocumentStore:
 
     def get_documents_generator(
         self,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], rest.Filter]] = None,
     ) -> Generator[Document, None, None]:
         index = self.index
         qdrant_filters = convert_filters_to_qdrant(filters)
@@ -311,7 +317,7 @@ class QdrantDocumentStore:
     def _query_by_sparse(
         self,
         query_sparse_embedding: SparseEmbedding,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], rest.Filter]] = None,
         top_k: int = 10,
         scale_score: bool = True,
         return_embedding: bool = False,
@@ -353,7 +359,7 @@ class QdrantDocumentStore:
     def _query_by_embedding(
         self,
         query_embedding: List[float],
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], rest.Filter]] = None,
         top_k: int = 10,
         scale_score: bool = True,
         return_embedding: bool = False,
@@ -388,7 +394,7 @@ class QdrantDocumentStore:
         self,
         query_embedding: List[float],
         query_sparse_embedding: SparseEmbedding,
-        filters: Optional[Dict[str, Any]] = None,
+        filters: Optional[Union[Dict[str, Any], rest.Filter]] = None,
         top_k: int = 10,
         return_embedding: bool = False,
     ) -> List[Document]:
@@ -464,7 +470,7 @@ class QdrantDocumentStore:
 
         return results
 
-    def _get_distance(self, similarity: str) -> rest.Distance:
+    def get_distance(self, similarity: str) -> rest.Distance:
         try:
             return self.SIMILARITY[similarity]
         except KeyError as ke:
@@ -498,31 +504,17 @@ class QdrantDocumentStore:
         on_disk: bool = False,
         payload_fields_to_index: Optional[List[dict]] = None,
     ):
-        distance = self._get_distance(similarity)
+        distance = self.get_distance(similarity)
 
-        if recreate_collection:
+        if recreate_collection or not self.client.collection_exists(collection_name):
             # There is no need to verify the current configuration of that
-            # collection. It might be just recreated again.
-            self._recreate_collection(collection_name, distance, embedding_dim, on_disk, use_sparse_embeddings)
+            # collection. It might be just recreated again or does not exist yet.
+            self.recreate_collection(collection_name, distance, embedding_dim, on_disk, use_sparse_embeddings)
             # Create Payload index if payload_fields_to_index is provided
             self._create_payload_index(collection_name, payload_fields_to_index)
             return
 
-        try:
-            # Check if the collection already exists and validate its
-            # current configuration with the parameters.
-            collection_info = self.client.get_collection(collection_name)
-        except (UnexpectedResponse, RpcError, ValueError):
-            # That indicates the collection does not exist, so it can be
-            # safely created with any configuration.
-            #
-            # Qdrant local raises ValueError if the collection is not found, but
-            # with the remote server UnexpectedResponse / RpcError is raised.
-            # Until that's unified, we need to catch both.
-            self._recreate_collection(collection_name, distance, embedding_dim, on_disk, use_sparse_embeddings)
-            # Create Payload index if payload_fields_to_index is provided
-            self._create_payload_index(collection_name, payload_fields_to_index)
-            return
+        collection_info = self.client.get_collection(collection_name)
 
         has_named_vectors = (
             isinstance(collection_info.config.params.vectors, dict)
@@ -573,14 +565,20 @@ class QdrantDocumentStore:
             )
             raise ValueError(msg)
 
-    def _recreate_collection(
+    def recreate_collection(
         self,
         collection_name: str,
         distance,
         embedding_dim: int,
-        on_disk: bool,
-        use_sparse_embeddings: bool,
+        on_disk: Optional[bool] = None,
+        use_sparse_embeddings: Optional[bool] = None,
     ):
+        if on_disk is None:
+            on_disk = self.on_disk
+
+        if use_sparse_embeddings is None:
+            use_sparse_embeddings = self.use_sparse_embeddings
+
         # dense vectors configuration
         vectors_config = rest.VectorParams(size=embedding_dim, on_disk=on_disk, distance=distance)
 
@@ -596,7 +594,10 @@ class QdrantDocumentStore:
                 ),
             }
 
-        self.client.recreate_collection(
+        if self.client.collection_exists(collection_name):
+            self.client.delete_collection(collection_name)
+
+        self.client.create_collection(
             collection_name=collection_name,
             vectors_config=vectors_config,
             sparse_vectors_config=sparse_vectors_config if use_sparse_embeddings else None,

haystack_integrations/document_stores/qdrant/filters.py

@@ -11,10 +11,11 @@ LOGICAL_OPERATORS = LOGICAL_OPERATORS.keys()
 
 
 def convert_filters_to_qdrant(
-    filter_term: Optional[Union[List[dict], dict]] = None,
+    filter_term: Optional[Union[List[dict], dict, models.Filter]] = None,
 ) -> Optional[models.Filter]:
     """Converts Haystack filters to the format used by Qdrant."""
-
+    if isinstance(filter_term, models.Filter):
+        return filter_term
     if not filter_term:
         return None
 

{qdrant_haystack-3.5.0.dist-info → qdrant_haystack-3.7.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: qdrant-haystack
-Version: 3.5.0
+Version: 3.7.0
 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

{qdrant_haystack-3.5.0.dist-info → qdrant_haystack-3.7.0.dist-info}/RECORD

@@ -1,11 +1,11 @@
 haystack_integrations/components/retrievers/qdrant/__init__.py,sha256=IRjcM4f8b5eKFEMn8tn6h6RrfslEGP3WafU7mrzNzQM,313
-haystack_integrations/components/retrievers/qdrant/retriever.py,sha256=cb2JxC3O-pH8iWUcLO4EduKUYiO7KoVzBOsrnirSb8w,13384
+haystack_integrations/components/retrievers/qdrant/retriever.py,sha256=r416_a7_6l7ehfPLdFtyHncsPrHj3mFCyROeJJF9rwM,13463
 haystack_integrations/document_stores/qdrant/__init__.py,sha256=kUGc5uewqArhmVR-JqB_NmJ4kNkTIQIvYDNSoO2ELn0,302
 haystack_integrations/document_stores/qdrant/converters.py,sha256=oSO2YlsWEQbcw9CPlWfSg_HoTZlnkAhZw_6VlYWzKKs,2525
-haystack_integrations/document_stores/qdrant/document_store.py,sha256=rzsDn8IkxyWisSR-jdgGJw6khk8Lqwer0QxCMHYLDxY,26260
-haystack_integrations/document_stores/qdrant/filters.py,sha256=iNWOqv1otUaXTURXd8e9QOYg8sx3Qm_LOqOaxAP2xJI,8249
+haystack_integrations/document_stores/qdrant/document_store.py,sha256=JOOBaE4nGWAp8MZmjsBgYIYUZzZCLrRUGa8KNwiVEd0,26347
+haystack_integrations/document_stores/qdrant/filters.py,sha256=0w70Wa3Za1fNdbJ5O95sZDIpXfblJG_sBBUv0JTQ0-o,8337
 haystack_integrations/document_stores/qdrant/migrate_to_sparse.py,sha256=i6wBC_9_JVzYZtqKm3dhHKTxhwNdcAdpgki8GABDp1c,4909
-qdrant_haystack-3.5.0.dist-info/METADATA,sha256=R-_leVdxwt--25K9DfScVtjyWgUiaHVZ3ax21iD6Vyw,1862
-qdrant_haystack-3.5.0.dist-info/WHEEL,sha256=zEMcRr9Kr03x1ozGwg5v9NQBKn3kndp6LSoSlVg-jhU,87
-qdrant_haystack-3.5.0.dist-info/licenses/LICENSE.txt,sha256=B05uMshqTA74s-0ltyHKI6yoPfJ3zYgQbvcXfDVGFf8,10280
-qdrant_haystack-3.5.0.dist-info/RECORD,,
+qdrant_haystack-3.7.0.dist-info/METADATA,sha256=Q2rwJIK4Cw96sSyHuOJY1w8az1kVk4jUMIBLbqSKJGI,1862
+qdrant_haystack-3.7.0.dist-info/WHEEL,sha256=zEMcRr9Kr03x1ozGwg5v9NQBKn3kndp6LSoSlVg-jhU,87
+qdrant_haystack-3.7.0.dist-info/licenses/LICENSE.txt,sha256=B05uMshqTA74s-0ltyHKI6yoPfJ3zYgQbvcXfDVGFf8,10280
+qdrant_haystack-3.7.0.dist-info/RECORD,,