refactor(pinecone): make get_metadata_field_min_max more consistent; use async DocumentStore mixin tests (#3231)

Co-authored-by: anakin87 <stefanofiorucci@gmail.com>

Author: Aftabbs

integrations/pinecone/pyproject.toml

@@ -24,7 +24,7 @@ classifiers = [
   "Programming Language :: Python :: Implementation :: PyPy",
 ]
 dependencies = [
-  "haystack-ai>=2.26.1",
+  "haystack-ai>=2.28.0",
   "pinecone[asyncio]>=7.0.0",
 ]
 

integrations/pinecone/src/haystack_integrations/document_stores/pinecone/document_store.py

@@ -873,8 +873,7 @@ def _get_metadata_field_min_max_impl(documents: list[Document], metadata_field:
                     values.append(value)
 
         if not values:
-            msg = f"No values found for metadata field '{metadata_field}'"
-            raise ValueError(msg)
+            return {"min": None, "max": None}
 
         result = {"min": min(values), "max": max(values)}
 
@@ -1041,8 +1040,8 @@ def get_metadata_field_min_max(self, metadata_field: str) -> dict[str, Any]:
         Subject to Pinecone's TOP_K_LIMIT of 1000 documents.
 
         :param metadata_field: The metadata field name to analyze.
-        :returns: Dictionary with 'min' and 'max' keys.
-        :raises ValueError: If the field doesn't exist or has no values.
+        :returns: Dictionary with 'min' and 'max' keys. Both values are None if the field has no
+            values (empty store, field absent, or unsupported field type).
         """
         documents = self.filter_documents(filters=None)
         return self._get_metadata_field_min_max_impl(documents, metadata_field)
@@ -1060,8 +1059,8 @@ async def get_metadata_field_min_max_async(self, metadata_field: str) -> dict[st
         Subject to Pinecone's TOP_K_LIMIT of 1000 documents.
 
         :param metadata_field: The metadata field name to analyze.
-        :returns: Dictionary with 'min' and 'max' keys.
-        :raises ValueError: If the field doesn't exist or has no values.
+        :returns: Dictionary with 'min' and 'max' keys. Both values are None if the field has no
+            values (empty store, field absent, or unsupported field type).
         """
         documents = await self.filter_documents_async(filters=None)
         return self._get_metadata_field_min_max_impl(documents, metadata_field)

integrations/pinecone/tests/test_document_store.py

@@ -279,15 +279,13 @@ def test_get_metadata_fields_info_impl_type_inference(documents, expected, warni
         assert warning_fragment in caplog.text
 
 
-def test_get_metadata_field_min_max_impl_strips_meta_prefix_and_errors():
+def test_get_metadata_field_min_max_impl_strips_meta_prefix_and_handles_missing():
     docs = [
         Document(content="a", meta={"priority": 1}),
         Document(content="b", meta={"priority": 5}),
     ]
     assert PineconeDocumentStore._get_metadata_field_min_max_impl(docs, "meta.priority") == {"min": 1, "max": 5}
-
-    with pytest.raises(ValueError, match="No values found"):
-        PineconeDocumentStore._get_metadata_field_min_max_impl(docs, "missing")
+    assert PineconeDocumentStore._get_metadata_field_min_max_impl(docs, "missing") == {"min": None, "max": None}
 
 
 def test_get_metadata_field_unique_values_impl_pagination_search_and_lists():
@@ -523,25 +521,18 @@ def test_get_metadata_field_min_max_boolean_and_string(self, document_store: Pin
         assert min_max["min"] == "Alpha"
         assert min_max["max"] == "Zebra"
 
-    def test_get_metadata_field_min_max_empty_collection(self, document_store: PineconeDocumentStore):
-        assert document_store.count_documents() == 0
-        with pytest.raises(ValueError, match="No values found"):
-            document_store.get_metadata_field_min_max("priority")
-
     def test_get_metadata_field_min_max_no_values(self, document_store: PineconeDocumentStore):
         docs = [
             Document(content="Doc 1", meta={"tags": ["tag1", "tag2"]}),
             Document(content="Doc 2", meta={"tags": ["tag3", "tag4"]}),
         ]
         document_store.write_documents(docs)
 
-        # Try to get min/max for unsupported field type (list)
-        with pytest.raises(ValueError, match="No values found"):
-            document_store.get_metadata_field_min_max("tags")
+        # Unsupported field type (list) — no comparable values collected
+        assert document_store.get_metadata_field_min_max("tags") == {"min": None, "max": None}
 
-        # Try to get min/max for non-existent field
-        with pytest.raises(ValueError, match="No values found"):
-            document_store.get_metadata_field_min_max("nonexistent")
+        # Non-existent field
+        assert document_store.get_metadata_field_min_max("nonexistent") == {"min": None, "max": None}
 
     def test_get_metadata_field_unique_values(self, document_store: PineconeDocumentStore):
         docs = [