falkordb-haystack 1.0.0__py3-none-any.whl

falkordb_haystack-1.0.0.dist-info/METADATA

@@ -0,0 +1,51 @@
+Metadata-Version: 2.4
+Name: falkordb-haystack
+Version: 1.0.0
+Summary: FalkorDB graph database integration for Haystack — enables GraphRAG with OpenCypher
+Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/falkordb#readme
+Project-URL: Issues, https://github.com/deepset-ai/haystack-core-integrations/issues
+Project-URL: Source, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/falkordb
+Author-email: deepset GmbH <info@deepset.ai>
+License-Expression: Apache-2.0
+License-File: LICENSE.txt
+Keywords: document-store,falkordb,graph,graph-database,graphrag,haystack,openCypher,rag,vector-search
+Classifier: Development Status :: 4 - Beta
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: Implementation :: CPython
+Classifier: Programming Language :: Python :: Implementation :: PyPy
+Classifier: Topic :: Database
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Requires-Dist: falkordb<2,>=1.1
+Requires-Dist: haystack-ai>=2.26.1
+Description-Content-Type: text/markdown
+
+# falkordb-haystack
+
+[![PyPI - Version](https://img.shields.io/pypi/v/falkordb-haystack.svg)](https://pypi.org/project/falkordb-haystack)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/falkordb-haystack.svg)](https://pypi.org/project/falkordb-haystack)
+
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/falkordb/CHANGELOG.md)
+
+---
+
+
+## Contributing
+
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).
+
+
+
+## Testing
+
+To run the integration tests locally, you need a running FalkorDB instance. You can start one using the provided docker-compose file:
+
+```bash
+docker-compose up -d
+```

falkordb_haystack-1.0.0.dist-info/RECORD

@@ -0,0 +1,11 @@
+haystack_integrations/components/retrievers/py.typed,sha256=4W8VliAYUP1KY2gLJ_YDy2TmcXYVm-PY7XikQD_bFwA,2
+haystack_integrations/components/retrievers/falkordb/__init__.py,sha256=g681NER0VaB-xo14jFzOi66YQRh9oLxnIvmGI2ydJQs,294
+haystack_integrations/components/retrievers/falkordb/cypher_retriever.py,sha256=yk5xghWjyHDVya57PD6Q8EADqh_UDoMB6MxTQgI8f_Q,4199
+haystack_integrations/components/retrievers/falkordb/embedding_retriever.py,sha256=kV3Z5gdVjHT1Tfx0b2Lq9fAqiFaahkpW-bPJAE2r_78,4701
+haystack_integrations/document_stores/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+haystack_integrations/document_stores/falkordb/__init__.py,sha256=v7rOS3LAZ6EYknSYtXmjkBD8O5eFIuLmV6TqFHb1arY,244
+haystack_integrations/document_stores/falkordb/document_store.py,sha256=AvsWDtJeydLd0kGC-2HspaGu7ADDxUtUxK-gIInuV8g,28587
+falkordb_haystack-1.0.0.dist-info/METADATA,sha256=oYpoJM43rIwrorG7PcTN0HjNjZSG7ku9SDdyF-erjvs,2184
+falkordb_haystack-1.0.0.dist-info/WHEEL,sha256=QccIxa26bgl1E6uMy58deGWi-0aeIkkangHcxk2kWfw,87
+falkordb_haystack-1.0.0.dist-info/licenses/LICENSE.txt,sha256=OrAtTu4aPmrTjKDSTzkSPgo_ji6NOMWO2N9JOTFyAzM,11350
+falkordb_haystack-1.0.0.dist-info/RECORD,,

falkordb_haystack-1.0.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.29.0
+Root-Is-Purelib: true
+Tag: py3-none-any

falkordb_haystack-1.0.0.dist-info/licenses/LICENSE.txt

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2023-present deepset GmbH
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

haystack_integrations/components/retrievers/falkordb/__init__.py

@@ -0,0 +1,8 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from .cypher_retriever import FalkorDBCypherRetriever
+from .embedding_retriever import FalkorDBEmbeddingRetriever
+
+__all__ = ["FalkorDBCypherRetriever", "FalkorDBEmbeddingRetriever"]

haystack_integrations/components/retrievers/falkordb/cypher_retriever.py

@@ -0,0 +1,113 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from typing import Any
+
+from haystack import component, default_from_dict, default_to_dict
+from haystack.dataclasses import Document
+
+from haystack_integrations.document_stores.falkordb import FalkorDBDocumentStore
+
+
+@component
+class FalkorDBCypherRetriever:
+    """
+    A power-user retriever for executing arbitrary OpenCypher queries against FalkorDB.
+
+    This retriever allows you to leverage graph traversal and multi-hop queries in
+    GraphRAG pipelines. The query must return nodes or dictionaries that can be
+    mapped exactly to a Haystack `Document`.
+
+    **Security Warning:** Raw Cypher queries must only come from trusted sources. Do
+    not use un-sanitised user input directly in query strings. Use `parameters` instead.
+
+    Usage example:
+    ```python
+    from haystack_integrations.document_stores.falkordb import FalkorDBDocumentStore
+    from haystack_integrations.components.retrievers.falkordb import FalkorDBCypherRetriever
+
+    store = FalkorDBDocumentStore(host="localhost", port=6379)
+    retriever = FalkorDBCypherRetriever(
+        document_store=store,
+        custom_cypher_query="MATCH (d:Document)-[:RELATES_TO]->(:Concept {name: $concept}) RETURN d"
+    )
+
+    res = retriever.run(parameters={"concept": "GraphRAG"})
+    print(res["documents"])
+    ```
+    """
+
+    def __init__(
+        self,
+        document_store: FalkorDBDocumentStore,
+        custom_cypher_query: str | None = None,
+    ) -> None:
+        """
+        Create a new FalkorDBCypherRetriever.
+
+        :param document_store: The FalkorDBDocumentStore instance.
+        :param custom_cypher_query: A static OpenCypher query to execute. Can be
+            overridden at runtime by passing `query` to `run()`.
+        :raises ValueError: If the provided `document_store` is not a `FalkorDBDocumentStore`.
+        """
+        if not isinstance(document_store, FalkorDBDocumentStore):
+            msg = "document_store must be an instance of FalkorDBDocumentStore"
+            raise ValueError(msg)
+
+        self.document_store = document_store
+        self.custom_cypher_query = custom_cypher_query
+
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Serialise the retriever to a dictionary.
+
+        :returns: Dictionary representation of the retriever.
+        """
+        return default_to_dict(
+            self,
+            document_store=self.document_store.to_dict(),
+            custom_cypher_query=self.custom_cypher_query,
+        )
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> "FalkorDBCypherRetriever":
+        """
+        Deserialise a `FalkorDBCypherRetriever` produced by `to_dict`.
+
+        :param data: Serialised retriever dictionary.
+        :returns: Reconstructed `FalkorDBCypherRetriever` instance.
+        """
+        init_params = data["init_parameters"]
+        init_params["document_store"] = FalkorDBDocumentStore.from_dict(init_params["document_store"])
+        return default_from_dict(cls, data)
+
+    @component.output_types(documents=list[Document])
+    def run(
+        self,
+        query: str | None = None,
+        parameters: dict[str, Any] | None = None,
+    ) -> dict[str, list[Document]]:
+        """
+        Retrieve documents by executing an OpenCypher query.
+
+        If a `query` is provided here, it overrides the `custom_cypher_query`
+        set during initialisation.
+
+        :param query: Optional OpenCypher query string.
+        :param parameters: Optional dictionary of query parameters (referenced as
+            `$param_name` in the Cypher string).
+        :raises ValueError: If no query string is provided (both here and at init).
+        :returns: Dictionary containing a `"documents"` key with the retrieved documents.
+        """
+        cypher = query or self.custom_cypher_query
+        if not cypher:
+            msg = "A Cypher query string must be provided either at init or at runtime."
+            raise ValueError(msg)
+
+        docs = self.document_store._cypher_retrieval(
+            cypher_query=cypher,
+            parameters=parameters,
+        )
+
+        return {"documents": docs}

haystack_integrations/components/retrievers/falkordb/embedding_retriever.py

@@ -0,0 +1,120 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from typing import Any
+
+from haystack import component, default_from_dict, default_to_dict
+from haystack.dataclasses import Document
+from haystack.document_stores.types import FilterPolicy, apply_filter_policy
+
+from haystack_integrations.document_stores.falkordb import FalkorDBDocumentStore
+
+
+@component
+class FalkorDBEmbeddingRetriever:
+    """
+    A component for retrieving documents from a FalkorDBDocumentStore using vector similarity.
+
+    The retriever uses FalkorDB's native vector search index to find documents whose embeddings
+    are most similar to the provided query embedding.
+
+    Usage example:
+    ```python
+    from haystack.dataclasses import Document
+    from haystack_integrations.document_stores.falkordb import FalkorDBDocumentStore
+    from haystack_integrations.components.retrievers.falkordb import FalkorDBEmbeddingRetriever
+
+    store = FalkorDBDocumentStore(host="localhost", port=6379)
+    store.write_documents([
+        Document(content="GraphRAG is powerful.", embedding=[0.1, 0.2, 0.3]),
+        Document(content="FalkorDB is fast.", embedding=[0.8, 0.9, 0.1]),
+    ])
+
+    retriever = FalkorDBEmbeddingRetriever(document_store=store)
+    res = retriever.run(query_embedding=[0.1, 0.2, 0.3])
+    print(res["documents"][0].content)  # "GraphRAG is powerful."
+    ```
+    """
+
+    def __init__(
+        self,
+        document_store: FalkorDBDocumentStore,
+        filters: dict[str, Any] | None = None,
+        top_k: int = 10,
+        filter_policy: FilterPolicy = FilterPolicy.REPLACE,
+    ) -> None:
+        """
+        Create a new FalkorDBEmbeddingRetriever.
+
+        :param document_store: The FalkorDBDocumentStore instance.
+        :param filters: Optional Haystack filters to narrow down the search space.
+        :param top_k: Maximum number of documents to retrieve.
+        :param filter_policy: Policy to determine how runtime filters are combined with
+            initialization filters.
+        :raises ValueError: If the provided `document_store` is not a `FalkorDBDocumentStore`.
+        """
+        if not isinstance(document_store, FalkorDBDocumentStore):
+            msg = "document_store must be an instance of FalkorDBDocumentStore"
+            raise ValueError(msg)
+
+        self.document_store = document_store
+        self.filters = filters
+        self.top_k = top_k
+        self.filter_policy = FilterPolicy(filter_policy) if isinstance(filter_policy, str) else filter_policy
+
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Serialise the retriever to a dictionary.
+
+        :returns: Dictionary representation of the retriever.
+        """
+        return default_to_dict(
+            self,
+            document_store=self.document_store.to_dict(),
+            filters=self.filters,
+            top_k=self.top_k,
+            filter_policy=self.filter_policy.value,
+        )
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> "FalkorDBEmbeddingRetriever":
+        """
+        Deserialise a `FalkorDBEmbeddingRetriever` produced by `to_dict`.
+
+        :param data: Serialised retriever dictionary.
+        :returns: Reconstructed `FalkorDBEmbeddingRetriever` instance.
+        """
+        init_params = data["init_parameters"]
+        init_params["document_store"] = FalkorDBDocumentStore.from_dict(init_params["document_store"])
+        if "filter_policy" in init_params:
+            init_params["filter_policy"] = FilterPolicy(init_params["filter_policy"])
+        return default_from_dict(cls, data)
+
+    @component.output_types(documents=list[Document])
+    def run(
+        self,
+        query_embedding: list[float],
+        filters: dict[str, Any] | None = None,
+        top_k: int | None = None,
+    ) -> dict[str, list[Document]]:
+        """
+        Retrieve documents by vector similarity.
+
+        :param query_embedding: Query embedding vector.
+        :param filters: Optional Haystack filters to be combined with the init filters based
+            on the configured filter policy.
+        :param top_k: Maximum number of documents to return. If not provided, the default
+            top_k from initialization is used.
+        :returns: Dictionary containing a `"documents"` key with the retrieved documents.
+        """
+        final_filters = apply_filter_policy(self.filter_policy, self.filters, filters)
+        final_top_k = top_k if top_k is not None else self.top_k
+
+        docs = self.document_store._embedding_retrieval(
+            query_embedding=query_embedding,
+            top_k=final_top_k,
+            filters=final_filters,
+        )
+
+        return {"documents": docs}

haystack_integrations/components/retrievers/py.typed

@@ -0,0 +1 @@
+ 

haystack_integrations/document_stores/falkordb/__init__.py

@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from haystack_integrations.document_stores.falkordb.document_store import FalkorDBDocumentStore
+
+__all__ = ["FalkorDBDocumentStore"]

haystack_integrations/document_stores/falkordb/document_store.py

@@ -0,0 +1,728 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+import logging
+from dataclasses import replace
+from datetime import datetime
+from typing import Any, Literal
+
+from haystack import default_from_dict, default_to_dict
+from haystack.dataclasses import Document
+from haystack.document_stores.errors import DocumentStoreError, DuplicateDocumentError
+from haystack.document_stores.types import DocumentStore, DuplicatePolicy
+from haystack.errors import FilterError
+from haystack.utils import Secret, deserialize_secrets_inplace
+from redis.exceptions import ResponseError
+
+import falkordb  # type: ignore[import-untyped,import-not-found]
+
+logger = logging.getLogger(__name__)
+
+# Haystack filter operators → Cypher comparison operators.
+_COMPARISON_OPS: dict[str, str] = {
+    "==": "=",
+    "!=": "<>",
+    ">": ">",
+    ">=": ">=",
+    "<": "<",
+    "<=": "<=",
+}
+
+SimilarityFunction = Literal["cosine", "euclidean"]
+
+
+class FalkorDBDocumentStore(DocumentStore):
+    """
+    A Haystack DocumentStore backed by FalkorDB — a high-performance graph database.
+
+    Optimised for GraphRAG workloads.
+
+    Documents are stored as graph nodes (labelled `Document` by default) in a named
+    FalkorDB graph.  Document properties, including `meta` fields, are stored
+    **flat** at the same level as `id` and `content` — exactly the same layout as
+    the `neo4j-haystack` reference integration.
+
+    Vector search is performed via FalkorDB's native vector index —
+    **no APOC is required**.  All bulk writes use `UNWIND` + `MERGE` for safe,
+    idiomatic OpenCypher upserts.
+
+    Usage example:
+
+    ```python
+    from haystack_integrations.document_stores.falkordb import FalkorDBDocumentStore
+    from haystack.dataclasses import Document
+
+    store = FalkorDBDocumentStore(host="localhost", port=6379)
+    store.write_documents([
+        Document(content="Hello, GraphRAG!", meta={"year": 2024}),
+    ])
+    print(store.count_documents())  # 1
+    ```
+    """
+
+    def __init__(
+        self,
+        *,
+        host: str = "localhost",
+        port: int = 6379,
+        graph_name: str = "haystack",
+        username: str | None = None,
+        password: Secret | None = None,
+        node_label: str = "Document",
+        embedding_dim: int = 768,
+        embedding_field: str = "embedding",
+        similarity: SimilarityFunction = "cosine",
+        write_batch_size: int = 100,
+        recreate_graph: bool = False,
+        verify_connectivity: bool = False,
+    ) -> None:
+        """
+        Create a new FalkorDBDocumentStore.
+
+        :param host: Hostname of the FalkorDB server.
+        :param port: Port the FalkorDB server listens on.
+        :param graph_name: Name of the FalkorDB graph to use. Each graph is an isolated
+            namespace.
+        :param username: Optional username for FalkorDB authentication.
+        :param password: Optional :class:`haystack.utils.Secret` holding the FalkorDB
+            password. The secret value is resolved lazily on first connection.
+        :param node_label: Label used for document nodes in the graph.
+        :param embedding_dim: Dimensionality of the vector embeddings. Used when
+            creating the vector index.
+        :param embedding_field: Name of the node property that stores the embedding
+            vector.
+        :param similarity: Similarity function for the vector index.  Accepted values
+            are `"cosine"` and `"euclidean"`.
+        :param write_batch_size: Number of documents written per `UNWIND` batch.
+        :param recreate_graph: When `True` the existing graph (and all its data) is
+            dropped and recreated on initialisation. Useful for tests.
+        :param verify_connectivity: When `True` a connectivity probe is run
+            immediately in `__init__` — raises if the server is unreachable.
+        :raises ValueError: If `similarity` is not `"cosine"` or `"euclidean"`.
+        """
+        if similarity not in ("cosine", "euclidean"):
+            msg = (
+                f"Provided similarity '{similarity}' is not supported by FalkorDBDocumentStore. "
+                "Please choose one of: 'cosine', 'euclidean'."
+            )
+            raise ValueError(msg)
+
+        self.host = host
+        self.port = port
+        self.graph_name = graph_name
+        self.username = username
+        self.password = password
+        self.node_label = node_label
+        self.embedding_dim = embedding_dim
+        self.embedding_field = embedding_field
+        self.similarity: SimilarityFunction = similarity
+        self.write_batch_size = write_batch_size
+        self.recreate_graph = recreate_graph
+        self.verify_connectivity = verify_connectivity
+
+        # Lazy — populated on first use via ensure_connected().
+        self.client: Any = None
+        self.graph: Any = None
+        self.initialized: bool = False
+
+        if verify_connectivity:
+            self._ensure_connected()
+
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Serialise the store to a dictionary suitable for `from_dict`.
+
+        :returns: Dictionary representation of the store.
+        """
+        return default_to_dict(
+            self,
+            host=self.host,
+            port=self.port,
+            graph_name=self.graph_name,
+            username=self.username,
+            password=self.password.to_dict() if self.password is not None else None,
+            node_label=self.node_label,
+            embedding_dim=self.embedding_dim,
+            embedding_field=self.embedding_field,
+            similarity=self.similarity,
+            write_batch_size=self.write_batch_size,
+            recreate_graph=self.recreate_graph,
+            verify_connectivity=self.verify_connectivity,
+        )
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> FalkorDBDocumentStore:
+        """
+        Deserialise a `FalkorDBDocumentStore` produced by `to_dict`.
+
+        :param data: Serialised store dictionary.
+        :returns: Reconstructed `FalkorDBDocumentStore` instance.
+        """
+        deserialize_secrets_inplace(data["init_parameters"], keys=["password"])
+        return default_from_dict(cls, data)
+
+    # ------------------------------------------------------------------
+    # Internal connection helpers
+    # ------------------------------------------------------------------
+
+    def _ensure_connected(self) -> None:
+        """
+        Lazily open the FalkorDB connection and set up the graph schema.
+
+        Called at the start of every public method so the store remains
+        serialisable without an active database connection.
+        """
+        if self.initialized:
+            return
+
+        password_value = self.password.resolve_value() if self.password is not None else None
+
+        self.client = falkordb.FalkorDB(
+            host=self.host,
+            port=self.port,
+            username=self.username,
+            password=password_value,
+        )
+
+        if self.recreate_graph:
+            try:
+                # In falkordb-py, delete() is a method of the Graph object
+                self.client.select_graph(self.graph_name).delete()
+            except Exception:
+                logger.debug("Graph '%s' could not be deleted (may not exist yet).", self.graph_name)
+
+        self.graph = self.client.select_graph(self.graph_name)
+        self._ensure_schema()
+        self.initialized = True
+
+    def _ensure_schema(self) -> None:
+        """
+        Create the property index and vector index if they do not already exist.
+
+        Uses only standard OpenCypher / FalkorDB-native syntax — **no APOC**.
+        """
+        # Property index on (:node_label {id}) for fast MERGE lookups.
+        try:
+            self.graph.query(f"CREATE INDEX FOR (d:{self.node_label}) ON (d.id)")
+        except ResponseError as e:
+            if "already indexed" in str(e).lower() or "already exists" in str(e).lower():
+                logger.debug("Property index on %s(id) already exists — skipping creation.", self.node_label)
+            else:
+                raise e
+
+        # FalkorDB-native vector index syntax
+        try:
+            cypher = (
+                f"CREATE VECTOR INDEX FOR (d:{self.node_label}) "
+                f"ON (d.{self.embedding_field}) "
+                f"OPTIONS {{dimension: {self.embedding_dim}, similarityFunction: '{self.similarity}'}}"
+            )
+            self.graph.query(cypher)
+        except ResponseError as e:
+            if "already indexed" in str(e).lower() or "already exists" in str(e).lower():
+                logger.debug(
+                    "Vector index on %s(%s) already exists — skipping creation.",
+                    self.node_label,
+                    self.embedding_field,
+                )
+            else:
+                raise e
+
+    # ------------------------------------------------------------------
+    # Haystack DocumentStore protocol
+    # ------------------------------------------------------------------
+
+    def count_documents(self) -> int:
+        """
+        Return the number of documents currently stored in the graph.
+
+        :returns: Integer count of document nodes.
+        """
+        self._ensure_connected()
+        result = self.graph.query(f"MATCH (d:{self.node_label}) RETURN count(d) AS n")
+        rows = result.result_set
+        return int(rows[0][0]) if rows else 0
+
+    def filter_documents(self, filters: dict[str, Any] | None = None) -> list[Document]:
+        """
+        Retrieve all documents that match the provided Haystack filters.
+
+        :param filters: Optional Haystack filter dict. When `None` all documents are
+            returned. For filter syntax see
+            [Metadata filtering](https://docs.haystack.deepset.ai/docs/metadata-filtering)
+        :returns: List of matching :class:`haystack.dataclasses.Document` objects.
+        :raises ValueError: If the filter dict is malformed.
+        """
+        self._ensure_connected()
+        if not filters:
+            result = self.graph.query(f"MATCH (d:{self.node_label}) RETURN d ORDER BY d.id")
+            return [_node_to_document(row[0]) for row in result.result_set]
+
+        if "operator" not in filters:
+            msg = "Invalid filter syntax. See https://docs.haystack.deepset.ai/docs/metadata-filtering"
+            raise FilterError(msg)
+
+        where_clause, params = _convert_filters(filters)
+        cypher = f"MATCH (d:{self.node_label}) WHERE {where_clause} RETURN d ORDER BY d.id"
+
+        result = self.graph.query(cypher, params)
+        return [_node_to_document(row[0]) for row in result.result_set]
+
+    def write_documents(
+        self,
+        documents: list[Document],
+        policy: DuplicatePolicy = DuplicatePolicy.NONE,
+    ) -> int:
+        """
+        Write documents to the FalkorDB graph using `UNWIND` + `MERGE` for batching.
+
+        Document `meta` fields are stored **flat** at the same level as `id` and
+        `content` — no prefix is added.  This matches the layout used by the
+        `neo4j-haystack` reference integration.
+
+        :param documents: List of :class:`haystack.dataclasses.Document` objects.
+        :param policy: How to handle documents whose `id` already exists.
+            Defaults to :attr:`DuplicatePolicy.NONE` (treated as FAIL).
+        :raises ValueError: If `documents` contains non-Document elements.
+        :raises DuplicateDocumentError: If `policy` is FAIL / NONE and a duplicate
+            ID is encountered.
+        :raises DocumentStoreError: If any other DB error occurs.
+        :returns: Number of documents written or updated.
+        """
+        self._ensure_connected()
+
+        for doc in documents:
+            if not isinstance(doc, Document):
+                msg = f"write_documents() expects a list of Documents but got an element of type {type(doc)}."
+                raise ValueError(msg)
+
+        if not documents:
+            logger.warning("Calling FalkorDBDocumentStore.write_documents() with an empty list.")
+            return 0
+
+        if policy == DuplicatePolicy.NONE:
+            policy = DuplicatePolicy.FAIL
+
+        document_objects = self._handle_duplicate_documents(documents, policy)
+
+        written = 0
+        for batch_start in range(0, len(document_objects), self.write_batch_size):
+            batch = document_objects[batch_start : batch_start + self.write_batch_size]
+            written += self._write_batch(batch, policy)
+
+        return written
+
+    def _handle_duplicate_documents(
+        self,
+        documents: list[Document],
+        policy: DuplicatePolicy,
+    ) -> list[Document]:
+        """
+        Checks for IDs that already exist in the database.
+
+        :param documents: All documents to write.
+        :param policy: Duplicate handling policy.
+        :returns: Filtered list ready for batch writing.
+        :raises DuplicateDocumentError: When `policy` is FAIL and existing IDs found.
+        """
+        if policy in (DuplicatePolicy.SKIP, DuplicatePolicy.FAIL):
+            # Step 1: deduplicate within the incoming list itself.
+            documents = self._drop_duplicate_documents(documents)
+
+            # Step 2: find which IDs already exist in the DB.
+            ids = [doc.id for doc in documents]
+            existing = self.graph.query(
+                f"UNWIND $ids AS id MATCH (d:{self.node_label} {{id: id}}) RETURN d.id",
+                {"ids": ids},
+            )
+            ids_exist_in_db: list[str] = [row[0] for row in existing.result_set]
+
+            if ids_exist_in_db and policy == DuplicatePolicy.FAIL:
+                msg = f"Document with ids '{', '.join(ids_exist_in_db)}' already exists in graph '{self.graph_name}'."
+                raise DuplicateDocumentError(msg)
+
+            # For SKIP: remove those that already exist.
+            if ids_exist_in_db:
+                existing_set = set(ids_exist_in_db)
+                documents = [d for d in documents if d.id not in existing_set]
+
+        return documents
+
+    def _drop_duplicate_documents(self, documents: list[Document]) -> list[Document]:
+        """
+        Drop duplicate documents (by ID) within the provided list.
+
+        :param documents: Input list — may contain repeated IDs.
+        :returns: Deduplicated list preserving first-occurrence order.
+        """
+        seen_ids: set[str] = set()
+        unique: list[Document] = []
+        for doc in documents:
+            if doc.id in seen_ids:
+                logger.info(
+                    "Duplicate Documents: Document with id '%s' already present in the batch — skipping.",
+                    doc.id,
+                )
+                continue
+            unique.append(doc)
+            seen_ids.add(doc.id)
+        return unique
+
+    def _write_batch(self, documents: list[Document], policy: DuplicatePolicy) -> int:
+        """
+        Write a single batch of documents using a single UNWIND query.
+
+        By the time this is called, duplicate handling has already been performed by
+        :meth:`_handle_duplicate_documents`.
+
+        :param documents: Batch of Documents (≤ `write_batch_size`).
+        :param policy: Duplicate policy — only OVERWRITE needs a different Cypher template.
+        :returns: Number of nodes created or updated.
+        """
+        records = [_document_to_falkordb_record(doc) for doc in documents]
+
+        if policy == DuplicatePolicy.OVERWRITE:
+            # ON MATCH SET applies the full map (including updated fields).
+            cypher = f"""
+UNWIND $docs AS doc
+MERGE (d:{self.node_label} {{id: doc.id}})
+ON CREATE SET d += doc
+ON MATCH SET d = doc
+RETURN count(d) AS n
+"""
+        else:
+            # FAIL already filtered duplicates above; SKIP excluded them.
+            # In both remaining cases we only write truly-new nodes.
+            cypher = f"""
+UNWIND $docs AS doc
+MERGE (d:{self.node_label} {{id: doc.id}})
+ON CREATE SET d += doc
+RETURN count(d) AS n
+"""
+
+        try:
+            result = self.graph.query(cypher, {"docs": records})
+            rows = result.result_set
+            written = int(rows[0][0]) if rows else 0
+        except Exception as exc:
+            msg = f"Failed to write documents to FalkorDB: {exc}"
+            raise DocumentStoreError(msg) from exc
+
+        # Second pass: store embeddings as vecf32 so FalkorDB's vector index
+        # recognises them as VALUE_VECTORF32 rather than a plain float array.
+        # vecf32() cannot be used inline in the MERGE query above because
+        # falkordb-py >=1.1 is required to parse VALUE_VECTORF32 results, and
+        # mixing it with RETURN count() triggers a client-side parsing error on
+        # older clients.
+        docs_with_emb = [doc for doc in documents if doc.embedding is not None]
+        if docs_with_emb:
+            emb_rows = [{"id": doc.id, "emb": doc.embedding} for doc in docs_with_emb]
+            try:
+                self.graph.query(
+                    f"""
+UNWIND $docs AS doc
+MATCH (d:{self.node_label} {{id: doc.id}})
+SET d.{self.embedding_field} = vecf32(doc.emb)
+""",
+                    {"docs": emb_rows},
+                )
+            except Exception as exc:
+                msg = f"Failed to set embeddings in FalkorDB: {exc}"
+                raise DocumentStoreError(msg) from exc
+
+        return written
+
+    def delete_documents(self, document_ids: list[str]) -> None:
+        """
+        Delete documents by their IDs using a single `UNWIND`-based query.
+
+        :param document_ids: List of document IDs to remove from the graph.
+        """
+        self._ensure_connected()
+        if not document_ids:
+            return
+        self.graph.query(
+            f"UNWIND $ids AS id MATCH (d:{self.node_label} {{id: id}}) DETACH DELETE d",
+            {"ids": document_ids},
+        )
+
+    # ------------------------------------------------------------------
+    # Internal retrieval helpers (called by retriever components)
+    # ------------------------------------------------------------------
+
+    def _embedding_retrieval(
+        self,
+        query_embedding: list[float],
+        top_k: int = 10,
+        filters: dict[str, Any] | None = None,
+        scale_score: bool = True,
+    ) -> list[Document]:
+        """
+        Retrieve documents by vector similarity using FalkorDB's native vector index.
+
+        Uses `CALL db.idx.vector.queryNodes` — FalkorDB's OpenCypher extension for
+        ANN search.  **No APOC is required.**
+
+        Cosine scores are returned as distance in `[0, 2]`; when `scale_score=True` they are
+        scaled to `[0, 1]` using the formula:
+        `1 - (score / 2)`.  Euclidean scores are transformed with `1 / (1 + score)`.
+
+        :param query_embedding: Query vector as a plain Python list of floats.
+        :param top_k: Maximum number of results to return.
+        :param filters: Optional Haystack filters applied as a `WHERE` predicate
+            on the vector search result set (post-filter).
+        :param scale_score: Whether to scale the raw similarity score to `[0, 1]`.
+        :returns: List of :class:`Document` objects ordered by similarity (best first).
+        """
+        self._ensure_connected()
+
+        if filters:
+            where_clause, filter_params = _convert_filters(filters)
+            cypher = f"""
+CALL db.idx.vector.queryNodes('{self.node_label}', '{self.embedding_field}', $top_k, vecf32($query_embedding))
+YIELD node AS d, score
+WHERE {where_clause}
+RETURN d, score
+ORDER BY score ASC, d.id ASC
+"""
+            params: dict[str, Any] = {
+                "top_k": top_k,
+                "query_embedding": query_embedding,
+                **filter_params,
+            }
+        else:
+            cypher = f"""
+CALL db.idx.vector.queryNodes('{self.node_label}', '{self.embedding_field}', $top_k, vecf32($query_embedding))
+YIELD node AS d, score
+RETURN d, score
+ORDER BY score ASC, d.id ASC
+"""
+            params = {"top_k": top_k, "query_embedding": query_embedding}
+
+        result = self.graph.query(cypher, params)
+        documents = []
+        for row in result.result_set:
+            node, score = row[0], row[1]
+            doc = _node_to_document(node)
+            final_score = self._scale_to_unit_interval(float(score)) if scale_score else float(score)
+            doc = replace(doc, score=final_score)
+            documents.append(doc)
+        return documents
+
+    def _cypher_retrieval(
+        self,
+        cypher_query: str,
+        parameters: dict[str, Any] | None = None,
+    ) -> list[Document]:
+        """
+        Execute an arbitrary OpenCypher query and map the results to Documents.
+
+        The first element of each result row is converted to a
+        :class:`haystack.dataclasses.Document`.
+
+        :param cypher_query: A valid OpenCypher query string.
+        :param parameters: Optional query parameters (`$param` placeholders).
+        :returns: List of :class:`Document` objects built from the query results.
+        :raises DocumentStoreError: If the query fails.
+        """
+        self._ensure_connected()
+
+        try:
+            # We don't force ORDER BY here as the query is custom,
+            # but we ensured everything else is stable.
+            result = self.graph.query(cypher_query, parameters or {})
+            return [_node_to_document(row[0]) for row in result.result_set]
+        except Exception as exc:
+            msg = f"Cypher query failed: {exc}"
+            raise DocumentStoreError(msg) from exc
+
+    def _scale_to_unit_interval(self, score: float) -> float:
+        """
+        Scale a raw similarity score to the unit interval `[0, 1]`.
+
+        Uses the following formulas:
+        - Cosine: `1 - (score / 2)`
+        - Euclidean: `1 / (1 + score)`
+
+        :param score: Raw score returned by the vector index.
+        :returns: Scaled score in `[0, 1]`.
+        """
+        if self.similarity == "cosine":
+            return 1 - (score / 2)
+        return 1 / (1 + score)
+
+
+# ---------------------------------------------------------------------------
+# Module-level helpers
+# ---------------------------------------------------------------------------
+
+
+def _document_to_falkordb_record(doc: Document) -> dict[str, Any]:
+    """
+    Convert a Haystack Document to a flat dict of non-embedding node properties.
+
+    - `meta` fields are stored **at the same level** as `id` and `content`.
+    - `embedding` is excluded; `_write_batch` stores it separately via `vecf32()`
+      so FalkorDB's vector index recognises the property as VALUE_VECTORF32.
+
+    :param doc: The document to convert.
+    :returns: Flat dictionary of non-embedding node properties.
+    """
+    record: dict[str, Any] = {}
+    if doc.meta:
+        record.update(doc.meta)
+    record["id"] = doc.id
+    record["content"] = doc.content
+
+    # Filter out None values — FalkorDB nodes don't need null properties stored.
+    return {k: v for k, v in record.items() if v is not None}
+
+
+def _node_to_document(node: Any) -> Document:
+    """
+    Convert a FalkorDB graph node back to a Haystack Document.
+
+    Properties that are not part of the standard Document schema are moved
+    into the `meta` dictionary.
+
+    :param node: A FalkorDB `Node` object or a plain `dict`.
+    :returns: Reconstructed :class:`haystack.dataclasses.Document`.
+    """
+    if hasattr(node, "properties"):
+        record: dict[str, Any] = dict(node.properties)
+    elif isinstance(node, dict):
+        record = node
+    else:
+        record = {}
+
+    # Standard Document fields
+    doc_id = record.pop("id", None)
+    content = record.pop("content", None)
+    embedding = record.pop("embedding", None)
+    score = record.pop("score", None)
+
+    # Everything else is metadata
+    # sparse_embedding is also popped if present (not supported by falkordb yet)
+    record.pop("sparse_embedding", None)
+
+    return Document(id=doc_id, content=content, embedding=embedding, meta=record, score=score)
+
+
+def _convert_filters(filters: dict[str, Any]) -> tuple[str, dict[str, Any]]:
+    """
+    Translate a Haystack filter dict into an OpenCypher `WHERE` sub-expression.
+
+    Supports the full Haystack filter DSL:
+
+    - Logical: `AND`, `OR`, `NOT`
+    - Comparison: `==`, `!=`, `>`, `>=`, `<`, `<=`
+    - Membership: `in`, `not in`
+
+    All values are passed as named query parameters to prevent injection.
+
+    :param filters: A Haystack filter dictionary.
+    :returns: Tuple of `(where_clause_string, params_dict)`.
+    :raises ValueError: If an unsupported operator or malformed filter is provided.
+    """
+    params: dict[str, Any] = {}
+    clause = _build_clause(filters, params, counter=[0])
+    return clause, params
+
+
+def _build_clause(node: dict[str, Any], params: dict[str, Any], counter: list[int]) -> str:
+    """
+    Recursively build a Cypher WHERE sub-expression from a Haystack filter node.
+
+    :param node: A filter node (logical group or comparison leaf).
+    :param params: Accumulating query parameter dict (mutated in place).
+    :param counter: Single-element list used as a mutable integer for unique param names.
+    :returns: Cypher expression string.
+    """
+    operator = node.get("operator", "")
+
+    # ------------------------------------------------------------------
+    # Logical / grouping operators
+    # ------------------------------------------------------------------
+    if operator.upper() in ("AND", "OR"):
+        if "conditions" not in node:
+            msg = f"Logical operator '{operator}' requires a 'conditions' key"
+            raise FilterError(msg)
+        sub_clauses = [_build_clause(c, params, counter) for c in node["conditions"]]
+        joiner = f" {operator.upper()} "
+        return f"({joiner.join(sub_clauses)})"
+
+    if operator.upper() == "NOT":
+        if "conditions" not in node:
+            msg = "Logical operator 'NOT' requires a 'conditions' key"
+            raise FilterError(msg)
+        sub_clauses = [_build_clause(c, params, counter) for c in node["conditions"]]
+        inner = " AND ".join(sub_clauses)
+        return f"NOT ({inner})"
+
+    # ------------------------------------------------------------------
+    # Leaf (comparison / membership) operators
+    # ------------------------------------------------------------------
+    if "field" not in node:
+        msg = f"Comparison operator '{operator}' requires a 'field' key"
+        raise FilterError(msg)
+    if "value" not in node:
+        msg = f"Comparison operator '{operator}' requires a 'value' key"
+        raise FilterError(msg)
+
+    field: str = node["field"]
+    value: Any = node["value"]
+
+    # Because meta fields are stored flat (no prefix), all fields map to d.<field>.
+    # We strip 'meta.' from the field name if Haystack adds it.
+    actual_field = field[5:] if field.startswith("meta.") else field
+    cypher_field = f"d.{actual_field}"
+
+    param_name = f"p{counter[0]}"
+    counter[0] += 1
+
+    if operator == "==":
+        if value is None:
+            return f"{cypher_field} IS NULL"
+        params[param_name] = value
+        return f"coalesce({cypher_field} = ${param_name}, false)"
+
+    if operator == "!=":
+        if value is None:
+            return f"{cypher_field} IS NOT NULL"
+        params[param_name] = value
+        return f"coalesce({cypher_field} <> ${param_name}, true)"
+
+    if operator in _COMPARISON_OPS:
+        if value is None:
+            return "false"
+        if isinstance(value, list):
+            msg = f"Operator '{operator}' does not support list values"
+            raise FilterError(msg)
+        if isinstance(value, str):
+            try:
+                datetime.fromisoformat(value)
+            except ValueError:
+                msg = f"Operator '{operator}' requires a numeric or ISO date value, got non-ISO string: '{value}'"
+                raise FilterError(msg) from None
+        params[param_name] = value
+        return f"coalesce({cypher_field} {_COMPARISON_OPS[operator]} ${param_name}, false)"
+
+    if operator == "in":
+        if not isinstance(value, list):
+            msg = f"Operator 'in' requires a list value, got {type(value).__name__}"
+            raise FilterError(msg)
+        params[param_name] = value
+        return f"coalesce({cypher_field} IN ${param_name}, false)"
+
+    if operator == "not in":
+        if not isinstance(value, list):
+            msg = f"Operator 'not in' requires a list value, got {type(value).__name__}"
+            raise FilterError(msg)
+        params[param_name] = value
+        return f"coalesce(NOT ({cypher_field} IN ${param_name}), true)"
+
+    msg = f"Unsupported filter operator: '{operator}'"
+    raise FilterError(msg)