chroma-haystack 4.1.0__tar.gz → 4.2.0__tar.gz

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/CHANGELOG.md

@@ -1,5 +1,18 @@
 # Changelog
 
+## [integrations/chroma-v4.1.1] - 2026-02-10
+
+### 📚 Documentation
+
+- Fixing curly brackets (#2824)
+
+
+## [integrations/chroma-v4.1.0] - 2026-02-10
+
+### 🌀 Miscellaneous
+
+- Add metadata query operations to ChromaDocumentStore (#2819)
+
 ## [integrations/chroma-v4.0.0] - 2026-01-12
 
 ### 🚀 Features

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: chroma-haystack
-Version: 4.1.0
+Version: 4.2.0
 Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/chroma#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/chroma
-Author-email: John Doe <jd@example.com>
+Author-email: deepset GmbH <info@deepset.ai>
 License-Expression: Apache-2.0
 License-File: LICENSE
 Classifier: Development Status :: 4 - Beta
@@ -17,8 +17,8 @@ Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
 Requires-Python: >=3.10
-Requires-Dist: chromadb>=1.0.2
-Requires-Dist: haystack-ai>=2.22.0
+Requires-Dist: chromadb>=1.5.0
+Requires-Dist: haystack-ai>=2.24.0
 Description-Content-Type: text/markdown
 
 # chroma-haystack

chroma_haystack-4.2.0/pydoc/config_docusaurus.yml

@@ -0,0 +1,16 @@
+loaders:
+  - modules:
+      - haystack_integrations.components.retrievers.chroma.retriever
+      - haystack_integrations.document_stores.chroma.document_store
+      - haystack_integrations.document_stores.chroma.errors
+      - haystack_integrations.document_stores.chroma.utils
+    search_path: [../src]
+processors:
+  - type: filter
+    documented_only: true
+    skip_empty_modules: true
+renderer:
+  description: Chroma integration for Haystack
+  id: integrations-chroma
+  filename: chroma.md
+  title: Chroma

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/pyproject.toml

@@ -10,7 +10,7 @@ readme = "README.md"
 requires-python = ">=3.10"
 license = "Apache-2.0"
 keywords = []
-authors = [{ name = "John Doe", email = "jd@example.com" }]
+authors = [{ name = "deepset GmbH", email = "info@deepset.ai" }]
 classifiers = [
   "License :: OSI Approved :: Apache Software License",
   "Development Status :: 4 - Beta",
@@ -22,7 +22,10 @@ classifiers = [
   "Programming Language :: Python :: Implementation :: CPython",
   "Programming Language :: Python :: Implementation :: PyPy",
 ]
-dependencies = ["haystack-ai>=2.22.0", "chromadb>=1.0.2"]
+dependencies = [
+  "haystack-ai>=2.24.0",
+  "chromadb>=1.5.0"
+]
 
 [project.urls]
 Documentation = "https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/chroma#readme"
@@ -45,7 +48,7 @@ installer = "uv"
 dependencies = ["haystack-pydoc-tools", "ruff"]
 
 [tool.hatch.envs.default.scripts]
-docs = ["pydoc-markdown pydoc/config_docusaurus.yml"]
+docs = ["haystack-pydoc pydoc/config_docusaurus.yml"]
 fmt = "ruff check --fix {args}; ruff format {args}"
 fmt-check = "ruff check {args} && ruff format --check {args}"
 
@@ -73,8 +76,6 @@ non_interactive = true
 check_untyped_defs = true
 disallow_incomplete_defs = true
 
-[tool.hatch.metadata]
-allow-direct-references = true
 
 [tool.ruff]
 line-length = 120

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/src/haystack_integrations/document_stores/chroma/document_store.py

@@ -21,7 +21,9 @@ logger = logging.getLogger(__name__)
 
 
 VALID_DISTANCE_FUNCTIONS = "l2", "cosine", "ip"
+# Supported ChromaDB metadata values can be found here: https://cookbook.chromadb.dev/core/concepts/#metadata
 SUPPORTED_TYPES_FOR_METADATA_VALUES = str, int, float, bool
+SUPPORTED_METADATA_VALUES_TYPE = str | int | float | bool
 
 
 class ChromaDocumentStore:
@@ -533,12 +535,21 @@ class ChromaDocumentStore:
         data: dict[str, Any] = {"ids": [doc.id], "documents": [doc.content]}
 
         if doc.meta:
-            valid_meta = {}
+            valid_meta: dict[str, SUPPORTED_METADATA_VALUES_TYPE | list[SUPPORTED_METADATA_VALUES_TYPE]] = {}
             discarded_keys = []
 
             for k, v in doc.meta.items():
                 if isinstance(v, SUPPORTED_TYPES_FOR_METADATA_VALUES):
                     valid_meta[k] = v
+                # For list metadata values, list must not be empty, must be of all same type
+                # and be one of `SUPPORTED_TYPES_FOR_METADATA_VALUES`
+                elif (
+                    isinstance(v, list)
+                    and len(v) > 0
+                    and (first_type := type(v[0])) in SUPPORTED_TYPES_FOR_METADATA_VALUES
+                    and all(isinstance(item, first_type) for item in v)
+                ):
+                    valid_meta[k] = v
                 else:
                     discarded_keys.append(k)
 
@@ -856,12 +867,13 @@ class ChromaDocumentStore:
             else:
                 collection = self._collection.get()
                 ids = collection.get("ids", [])
-                self._collection.delete(ids=ids)  # type: ignore
-                logger.info(
-                    "Deleted all the {n_docs} documents from the collection '{name}'.",
-                    name=self._collection_name,
-                    n_docs=len(ids),
-                )
+                if ids:
+                    self._collection.delete(ids=ids)  # type: ignore
+                    logger.info(
+                        "Deleted all the {n_docs} documents from the collection '{name}'.",
+                        name=self._collection_name,
+                        n_docs=len(ids),
+                    )
         except Exception as e:
             msg = f"Failed to delete all documents from ChromaDB: {e!s}"
             raise DocumentStoreError(msg) from e
@@ -895,12 +907,13 @@ class ChromaDocumentStore:
             else:
                 collection = await self._async_collection.get()
                 ids = collection.get("ids", [])
-                await self._async_collection.delete(ids=ids)  # type: ignore
-                logger.info(
-                    "Deleted all the {n_docs} documents from the collection '{name}'.",
-                    name=self._collection_name,
-                    n_docs=len(ids),
-                )
+                if ids:
+                    await self._async_collection.delete(ids=ids)  # type: ignore
+                    logger.info(
+                        "Deleted all the {n_docs} documents from the collection '{name}'.",
+                        name=self._collection_name,
+                        n_docs=len(ids),
+                    )
 
         except Exception as e:
             msg = f"Failed to delete all documents from ChromaDB: {e!s}"
@@ -1183,7 +1196,11 @@ class ChromaDocumentStore:
             Can include or omit the "meta." prefix.
         :returns: A dictionary with the keys "min" and "max", where each value is
                   the minimum or maximum value of the metadata field across all documents.
-                  Returns {"min": None, "max": None} if field doesn't exist or has no values.
+                  Returns:
+                  ```python
+                    {"min": None, "max": None}
+                  ```
+                  if field doesn't exist or has no values.
         """
         self._ensure_initialized()
         assert self._collection is not None
@@ -1203,7 +1220,11 @@ class ChromaDocumentStore:
             Can include or omit the "meta." prefix.
         :returns: A dictionary with the keys "min" and "max", where each value is
                   the minimum or maximum value of the metadata field across all documents.
-                  Returns {"min": None, "max": None} if field doesn't exist or has no values.
+                  Returns:
+                  ```python
+                    {"min": None, "max": None}
+                  ```
+                  if field doesn't exist or has no values.
         """
         await self._ensure_initialized_async()
         assert self._async_collection is not None

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/tests/test_document_store.py

@@ -14,8 +14,12 @@ from haystack.dataclasses import ByteStream, Document
 from haystack.testing.document_store import (
     TEST_EMBEDDING_1,
     CountDocumentsTest,
+    DeleteAllTest,
+    DeleteByFilterTest,
     DeleteDocumentsTest,
+    FilterableDocsFixtureMixin,
     FilterDocumentsTest,
+    UpdateByFilterTest,
 )
 
 from haystack_integrations.document_stores.chroma import ChromaDocumentStore
@@ -34,7 +38,15 @@ def clear_chroma_system_cache():
     SharedSystemClient.clear_system_cache()
 
 
-class TestDocumentStore(CountDocumentsTest, DeleteDocumentsTest, FilterDocumentsTest):
+class TestDocumentStore(
+    CountDocumentsTest,
+    DeleteDocumentsTest,
+    FilterDocumentsTest,
+    FilterableDocsFixtureMixin,
+    UpdateByFilterTest,
+    DeleteAllTest,
+    DeleteByFilterTest,
+):
     """
     Common test cases will be provided by `DocumentStoreBaseTests` but
     you can add more to this class.
@@ -112,16 +124,6 @@ class TestDocumentStore(CountDocumentsTest, DeleteDocumentsTest, FilterDocuments
         store._ensure_initialized()
         assert store._client.get_settings().anonymized_telemetry is False
 
-    def test_invalid_client_settings(self, clear_chroma_system_cache):
-        store = ChromaDocumentStore(
-            client_settings={
-                "invalid_setting_name": "some_value",
-                "another_fake_setting": 123,
-            }
-        )
-        with pytest.raises(ValueError, match="Invalid client_settings"):
-            store._ensure_initialized()
-
     def test_to_dict(self, request):
         ds = ChromaDocumentStore(
             collection_name=request.node.name,
@@ -253,15 +255,17 @@ class TestDocumentStore(CountDocumentsTest, DeleteDocumentsTest, FilterDocuments
 
         assert document_store.filter_documents()[0].embedding == pytest.approx(TEST_EMBEDDING_1)
 
-    def test_write_documents_unsupported_meta_values(self, document_store: ChromaDocumentStore):
+    def test_write_documents_unsupported_meta_values(self, document_store: ChromaDocumentStore) -> None:
         """
         Unsupported meta values should be removed from the documents before writing them to the database
         """
 
         docs = [
             Document(content="test doc 1", meta={"invalid": {"dict": "value"}}),
-            Document(content="test doc 2", meta={"invalid": ["list", "value"]}),
-            Document(content="test doc 3", meta={"ok": 123}),
+            Document(content="test doc 2", meta={"invalid": [["list", "value"]]}),
+            Document(content="test doc 3", meta={"invalid": []}),
+            Document(content="test doc 4", meta={"invalid": ["list", 2]}),
+            Document(content="test doc 5", meta={"ok": 123}),
         ]
 
         document_store.write_documents(docs)
@@ -269,11 +273,41 @@ class TestDocumentStore(CountDocumentsTest, DeleteDocumentsTest, FilterDocuments
         written_docs = document_store.filter_documents()
         written_docs.sort(key=lambda x: x.content)
 
-        assert len(written_docs) == 3
+        assert len(written_docs) == 5
         assert [doc.id for doc in written_docs] == [doc.id for doc in docs]
         assert written_docs[0].meta == {}
         assert written_docs[1].meta == {}
-        assert written_docs[2].meta == {"ok": 123}
+        assert written_docs[2].meta == {}
+        assert written_docs[3].meta == {}
+        assert written_docs[4].meta == {"ok": 123}
+
+    def test_write_documents_supported_meta_values(self, document_store: ChromaDocumentStore) -> None:
+        """
+        Unsupported meta values should be removed from the documents before writing them to the database
+        """
+
+        docs = [
+            Document(content="test doc 1", meta={"ok": "test"}),
+            Document(content="test doc 2", meta={"ok": ["list", "value"]}),
+            Document(content="test doc 3", meta={"ok": [True, False, True]}),
+            Document(content="test doc 4", meta={"ok": [1, 2, 3]}),
+            Document(content="test doc 5", meta={"ok": [1.1, 2.2, 3.3]}),
+            Document(content="test doc 6", meta={"ok": 123}),
+        ]
+
+        document_store.write_documents(docs)
+
+        written_docs = document_store.filter_documents()
+        written_docs.sort(key=lambda x: x.content)
+
+        assert len(written_docs) == 6
+        assert [doc.id for doc in written_docs] == [doc.id for doc in docs]
+        assert written_docs[0].meta == {"ok": "test"}
+        assert written_docs[1].meta == {"ok": ["list", "value"]}
+        assert written_docs[2].meta == {"ok": [True, False, True]}
+        assert written_docs[3].meta == {"ok": [1, 2, 3]}
+        assert written_docs[4].meta == {"ok": [1.1, 2.2, 3.3]}
+        assert written_docs[5].meta == {"ok": 123}
 
     def test_documents_with_content_none_are_not_stored(self, document_store: ChromaDocumentStore):
         document_store.write_documents([Document(content=None)])
@@ -465,88 +499,6 @@ class TestDocumentStore(CountDocumentsTest, DeleteDocumentsTest, FilterDocuments
         assert len(results) == 1
         assert results[0].content == "New document after delete all"
 
-    def test_delete_by_filter(self, document_store: ChromaDocumentStore):
-        docs = [
-            Document(content="Doc 1", meta={"category": "A"}),
-            Document(content="Doc 2", meta={"category": "B"}),
-            Document(content="Doc 3", meta={"category": "A"}),
-        ]
-        document_store.write_documents(docs)
-        assert document_store.count_documents() == 3
-
-        # delete documents with category="A"
-        deleted_count = document_store.delete_by_filter(
-            filters={"field": "meta.category", "operator": "==", "value": "A"}
-        )
-        assert deleted_count == 2
-        assert document_store.count_documents() == 1
-
-        # verify only category B remains
-        remaining_docs = document_store.filter_documents()
-        assert len(remaining_docs) == 1
-        assert remaining_docs[0].meta["category"] == "B"
-
-    def test_delete_by_filter_no_matches(self, document_store: ChromaDocumentStore):
-        docs = [
-            Document(content="Doc 1", meta={"category": "A"}),
-            Document(content="Doc 2", meta={"category": "B"}),
-        ]
-        document_store.write_documents(docs)
-        assert document_store.count_documents() == 2
-
-        # delete documents with category="C" (no matches)
-        deleted_count = document_store.delete_by_filter(
-            filters={"field": "meta.category", "operator": "==", "value": "C"}
-        )
-        assert deleted_count == 0
-        assert document_store.count_documents() == 2
-
-    def test_update_by_filter(self, document_store: ChromaDocumentStore):
-        docs = [
-            Document(content="Doc 1", meta={"category": "A", "status": "draft"}),
-            Document(content="Doc 2", meta={"category": "B", "status": "draft"}),
-            Document(content="Doc 3", meta={"category": "A", "status": "draft"}),
-        ]
-        document_store.write_documents(docs)
-        assert document_store.count_documents() == 3
-
-        # update status for category="A" documents
-        updated_count = document_store.update_by_filter(
-            filters={"field": "meta.category", "operator": "==", "value": "A"}, meta={"status": "published"}
-        )
-        assert updated_count == 2
-
-        # verify the updated documents have the new metadata
-        published_docs = document_store.filter_documents(
-            filters={"field": "meta.status", "operator": "==", "value": "published"}
-        )
-        assert len(published_docs) == 2
-        for doc in published_docs:
-            assert doc.meta["status"] == "published"
-            assert doc.meta["category"] == "A"
-
-        # Verify documents with category="B" were not updated
-        unpublished_docs = document_store.filter_documents(
-            filters={"field": "meta.category", "operator": "==", "value": "B"}
-        )
-        assert len(unpublished_docs) == 1
-        assert unpublished_docs[0].meta["status"] == "draft"
-
-    def test_update_by_filter_no_matches(self, document_store: ChromaDocumentStore):
-        docs = [
-            Document(content="Doc 1", meta={"category": "A"}),
-            Document(content="Doc 2", meta={"category": "B"}),
-        ]
-        document_store.write_documents(docs)
-        assert document_store.count_documents() == 2
-
-        # Try to update documents with category="C" (no matches)
-        updated_count = document_store.update_by_filter(
-            filters={"field": "meta.category", "operator": "==", "value": "C"}, meta={"status": "published"}
-        )
-        assert updated_count == 0
-        assert document_store.count_documents() == 2
-
     @pytest.mark.integration
     def test_search_embeddings(self, document_store: ChromaDocumentStore):
         query_embedding = TEST_EMBEDDING_1

{chroma_haystack-4.1.0 → chroma_haystack-4.2.0}/tests/test_document_store_async.py

@@ -107,19 +107,6 @@ class TestDocumentStoreAsync:
         await store._ensure_initialized_async()
         assert store._async_client.get_settings().anonymized_telemetry is False
 
-    async def test_invalid_client_settings_async(self):
-        store = ChromaDocumentStore(
-            host="localhost",
-            port=8000,
-            client_settings={
-                "invalid_setting_name": "some_value",
-                "another_fake_setting": 123,
-            },
-            collection_name=f"{uuid.uuid1()}-async-invalid",
-        )
-        with pytest.raises(ValueError, match="Invalid client_settings"):
-            await store._ensure_initialized_async()
-
     async def test_search_async(self):
         document_store = ChromaDocumentStore(host="localhost", port=8000, collection_name="my_custom_collection")
 

chroma_haystack-4.1.0/pydoc/config_docusaurus.yml

@@ -1,31 +0,0 @@
-loaders:
-- ignore_when_discovered:
-  - __init__
-  modules:
-  - haystack_integrations.components.retrievers.chroma.retriever
-  - haystack_integrations.document_stores.chroma.document_store
-  - haystack_integrations.document_stores.chroma.errors
-  - haystack_integrations.document_stores.chroma.utils
-  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: Chroma integration for Haystack
-  id: integrations-chroma
-  markdown:
-    add_member_class_prefix: false
-    add_method_class_prefix: true
-    classdef_code_block: false
-    descriptive_class_title: false
-    descriptive_module_title: true
-    filename: chroma.md
-  title: Chroma
-  type: haystack_pydoc_tools.renderers.DocusaurusRenderer