qdrant-haystack 4.1.1__tar.gz → 4.1.2__tar.gz

{qdrant_haystack-4.1.1 → qdrant_haystack-4.1.2}/CHANGELOG.md

@@ -1,5 +1,15 @@
 # Changelog
 
+## [integrations/qdrant-v4.1.1] - 2024-07-10
+
+### 🚀 Features
+
+- Add filter_policy to qdrant integration (#819)
+
+### 🐛 Bug Fixes
+
+- Errors in convert_filters_to_qdrant (#870)
+
 ## [integrations/qdrant-v4.1.0] - 2024-07-03
 
 ### 🚀 Features

{qdrant_haystack-4.1.1 → qdrant_haystack-4.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: qdrant-haystack
-Version: 4.1.1
+Version: 4.1.2
 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-4.1.1 → qdrant_haystack-4.1.2}/src/haystack_integrations/components/retrievers/qdrant/retriever.py

@@ -108,7 +108,10 @@ class QdrantEmbeddingRetriever:
         """
         document_store = QdrantDocumentStore.from_dict(data["init_parameters"]["document_store"])
         data["init_parameters"]["document_store"] = document_store
-        data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(data["init_parameters"]["filter_policy"])
+        # Pipelines serialized with old versions of the component might not
+        # have the filter_policy field.
+        if filter_policy := data["init_parameters"].get("filter_policy"):
+            data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(filter_policy)
         return default_from_dict(cls, data)
 
     @component.output_types(documents=List[Document])
@@ -249,7 +252,10 @@ class QdrantSparseEmbeddingRetriever:
         """
         document_store = QdrantDocumentStore.from_dict(data["init_parameters"]["document_store"])
         data["init_parameters"]["document_store"] = document_store
-        data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(data["init_parameters"]["filter_policy"])
+        # Pipelines serialized with old versions of the component might not
+        # have the filter_policy field.
+        if filter_policy := data["init_parameters"].get("filter_policy"):
+            data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(filter_policy)
         return default_from_dict(cls, data)
 
     @component.output_types(documents=List[Document])
@@ -394,7 +400,10 @@ class QdrantHybridRetriever:
         """
         document_store = QdrantDocumentStore.from_dict(data["init_parameters"]["document_store"])
         data["init_parameters"]["document_store"] = document_store
-        data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(data["init_parameters"]["filter_policy"])
+        # Pipelines serialized with old versions of the component might not
+        # have the filter_policy field.
+        if filter_policy := data["init_parameters"].get("filter_policy"):
+            data["init_parameters"]["filter_policy"] = FilterPolicy.from_str(filter_policy)
         return default_from_dict(cls, data)
 
     @component.output_types(documents=List[Document])

{qdrant_haystack-4.1.1 → qdrant_haystack-4.1.2}/tests/test_retriever.py

@@ -296,6 +296,31 @@ class TestQdrantSparseEmbeddingRetriever(FilterableDocsFixtureMixin):
         assert retriever._return_embedding is True
         assert retriever._score_threshold is None
 
+    def test_from_dict_no_filter_policy(self):
+        data = {
+            "type": "haystack_integrations.components.retrievers.qdrant.retriever.QdrantSparseEmbeddingRetriever",
+            "init_parameters": {
+                "document_store": {
+                    "init_parameters": {"location": ":memory:", "index": "test"},
+                    "type": "haystack_integrations.document_stores.qdrant.document_store.QdrantDocumentStore",
+                },
+                "filters": None,
+                "top_k": 5,
+                "scale_score": False,
+                "return_embedding": True,
+                "score_threshold": None,
+            },
+        }
+        retriever = QdrantSparseEmbeddingRetriever.from_dict(data)
+        assert isinstance(retriever._document_store, QdrantDocumentStore)
+        assert retriever._document_store.index == "test"
+        assert retriever._filters is None
+        assert retriever._top_k == 5
+        assert retriever._filter_policy == FilterPolicy.REPLACE  # defaults to REPLACE
+        assert retriever._scale_score is False
+        assert retriever._return_embedding is True
+        assert retriever._score_threshold is None
+
     def test_run(self, filterable_docs: List[Document], generate_sparse_embedding):
         document_store = QdrantDocumentStore(location=":memory:", index="Boi", use_sparse_embeddings=True)
 
@@ -414,6 +439,29 @@ class TestQdrantHybridRetriever:
         assert retriever._return_embedding
         assert retriever._score_threshold is None
 
+    def test_from_dict_no_filter_policy(self):
+        data = {
+            "type": "haystack_integrations.components.retrievers.qdrant.retriever.QdrantHybridRetriever",
+            "init_parameters": {
+                "document_store": {
+                    "init_parameters": {"location": ":memory:", "index": "test"},
+                    "type": "haystack_integrations.document_stores.qdrant.document_store.QdrantDocumentStore",
+                },
+                "filters": None,
+                "top_k": 5,
+                "return_embedding": True,
+                "score_threshold": None,
+            },
+        }
+        retriever = QdrantHybridRetriever.from_dict(data)
+        assert isinstance(retriever._document_store, QdrantDocumentStore)
+        assert retriever._document_store.index == "test"
+        assert retriever._filters is None
+        assert retriever._top_k == 5
+        assert retriever._filter_policy == FilterPolicy.REPLACE  # defaults to REPLACE
+        assert retriever._return_embedding
+        assert retriever._score_threshold is None
+
     def test_run(self):
         mock_store = Mock(spec=QdrantDocumentStore)
         sparse_embedding = SparseEmbedding(indices=[0, 1, 2, 3], values=[0.1, 0.8, 0.05, 0.33])