Skip to content

armgit / dify

Public Imported from GitHub
0

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.

test(api): bring the dataset model tests under strict pyrefly (#45695)

yetuge6a8b81a52863
diff --git a/api/tests/unit_tests/models/test_dataset.py b/api/tests/unit_tests/models/test_dataset.py
index fe6467dbd60..5222ed7105a 100644
--- a/api/tests/unit_tests/models/test_dataset.py
+++ b/api/tests/unit_tests/models/test_dataset.py
@@ -165,7 +165,7 @@ def _make_segments(document: Document, hit_counts: list[int]) -> list[DocumentSe
class TestDatasetModelValidation:
"""Test suite for Dataset model validation and basic operations."""
- def test_dataset_uses_typebase_registry_and_generates_ids(self):
+ def test_dataset_uses_typebase_registry_and_generates_ids(self) -> None:
generated_dataset = Dataset()
another_dataset = Dataset()
explicit_id = str(uuid4())
@@ -175,7 +175,7 @@ class TestDatasetModelValidation:
assert generated_dataset.id != another_dataset.id
assert Dataset(id=explicit_id).id == explicit_id
- def test_dataset_creation_with_required_fields(self):
+ def test_dataset_creation_with_required_fields(self) -> None:
"""Test creating a dataset with all required fields."""
# Arrange
tenant_id = str(uuid4())
@@ -196,7 +196,7 @@ class TestDatasetModelValidation:
assert dataset.created_by == created_by
# Note: Default values are set by database, not by model instantiation
- def test_dataset_creation_with_optional_fields(self):
+ def test_dataset_creation_with_optional_fields(self) -> None:
"""Test creating a dataset with optional fields."""
# Arrange & Act
dataset = Dataset(
@@ -217,7 +217,7 @@ class TestDatasetModelValidation:
assert dataset.embedding_model_provider == "openai"
@pytest.mark.parametrize("sqlite_session", [(Dataset, Account, DatasetProcessRule, Document)], indirect=True)
- def test_session_aware_dataset_getters_use_caller_session(self, sqlite_session: Session):
+ def test_session_aware_dataset_getters_use_caller_session(self, sqlite_session: Session) -> None:
account = Account(name="Ada", email="[email protected]")
account.id = "account-1"
dataset = _make_dataset(created_by=account.id)
@@ -250,7 +250,7 @@ class TestDatasetModelValidation:
assert get_dataset_doc_form(dataset, session=sqlite_session) is None
@pytest.mark.parametrize("sqlite_session", [(Dataset, DatasetKeywordTable)], indirect=True)
- def test_get_dataset_keyword_table_uses_caller_session(self, sqlite_session: Session):
+ def test_get_dataset_keyword_table_uses_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset()
keyword_table = DatasetKeywordTable(
dataset_id=dataset.id,
@@ -264,7 +264,7 @@ class TestDatasetModelValidation:
assert result is keyword_table
@pytest.mark.parametrize("sqlite_session", [(Dataset, Account, App, AppDatasetJoin, Document)], indirect=True)
- def test_dataset_detail_getters_use_caller_session(self, sqlite_session: Session):
+ def test_dataset_detail_getters_use_caller_session(self, sqlite_session: Session) -> None:
account = Account(name="Ada", email="[email protected]")
account.id = "account-1"
dataset = _make_dataset(created_by=account.id)
@@ -308,7 +308,7 @@ class TestDatasetModelValidation:
assert get_dataset_doc_metadata(dataset, session=sqlite_session) == []
assert get_dataset_is_published(dataset, session=sqlite_session) is False
- def test_dataset_indexing_technique_validation(self):
+ def test_dataset_indexing_technique_validation(self) -> None:
"""Test dataset indexing technique values."""
# Arrange & Act
dataset_high_quality = Dataset(
@@ -332,7 +332,7 @@ class TestDatasetModelValidation:
assert IndexTechniqueType.HIGH_QUALITY in Dataset.INDEXING_TECHNIQUE_LIST
assert IndexTechniqueType.ECONOMY in Dataset.INDEXING_TECHNIQUE_LIST
- def test_dataset_provider_validation(self):
+ def test_dataset_provider_validation(self) -> None:
"""Test dataset provider values."""
# Arrange & Act
dataset_vendor = Dataset(
@@ -356,7 +356,7 @@ class TestDatasetModelValidation:
assert "vendor" in Dataset.PROVIDER_LIST
assert "external" in Dataset.PROVIDER_LIST
- def test_dataset_index_struct_dict_property(self):
+ def test_dataset_index_struct_dict_property(self) -> None:
"""Test index_struct_dict property parsing."""
# Arrange
index_struct_data = {"type": "vector", "dimension": 1536}
@@ -373,10 +373,11 @@ class TestDatasetModelValidation:
# Assert
assert result == index_struct_data
+ assert result is not None
assert result["type"] == "vector"
assert result["dimension"] == 1536
- def test_dataset_index_struct_dict_property_none(self):
+ def test_dataset_index_struct_dict_property_none(self) -> None:
"""Test index_struct_dict property when index_struct is None."""
# Arrange
dataset = Dataset(
@@ -392,7 +393,7 @@ class TestDatasetModelValidation:
# Assert
assert result is None
- def test_dataset_external_retrieval_model_property(self):
+ def test_dataset_external_retrieval_model_property(self) -> None:
"""Test external_retrieval_model property with default values."""
# Arrange
dataset = Dataset(
@@ -412,7 +413,9 @@ class TestDatasetModelValidation:
@pytest.mark.parametrize(
"sqlite_session", [(Dataset, ExternalKnowledgeBindings, ExternalKnowledgeApis)], indirect=True
)
- def test_dataset_external_knowledge_info_returns_none_for_cross_tenant_template(self, sqlite_session: Session):
+ def test_dataset_external_knowledge_info_returns_none_for_cross_tenant_template(
+ self, sqlite_session: Session
+ ) -> None:
"""Test external datasets fail closed when the bound template is outside the tenant."""
dataset = _make_dataset(provider="external")
external_api = ExternalKnowledgeApis(
@@ -438,7 +441,7 @@ class TestDatasetModelValidation:
@pytest.mark.parametrize(
"sqlite_session", [(ExternalKnowledgeApis, ExternalKnowledgeBindings, Dataset)], indirect=True
)
- def test_external_knowledge_api_dataset_bindings_use_caller_session(self, sqlite_session: Session):
+ def test_external_knowledge_api_dataset_bindings_use_caller_session(self, sqlite_session: Session) -> None:
external_api = ExternalKnowledgeApis(
tenant_id="tenant-1",
created_by=str(uuid4()),
@@ -484,7 +487,7 @@ class TestDatasetModelValidation:
assert result == [{"id": dataset.id, "name": dataset.name}]
@pytest.mark.parametrize("sqlite_session", [(DatasetQuery, UploadFile)], indirect=True)
- def test_dataset_query_get_queries_uses_caller_session(self, sqlite_session: Session):
+ def test_dataset_query_get_queries_uses_caller_session(self, sqlite_session: Session) -> None:
dataset_query = DatasetQuery(
dataset_id=str(uuid4()),
content=json.dumps([{"content_type": "image_query", "content": "file-1"}]),
@@ -528,7 +531,7 @@ class TestDatasetModelValidation:
}
]
- def test_dataset_retrieval_model_dict_property(self):
+ def test_dataset_retrieval_model_dict_property(self) -> None:
"""Test retrieval_model_dict property with default values."""
# Arrange
dataset = Dataset(
@@ -546,7 +549,7 @@ class TestDatasetModelValidation:
assert result["reranking_enable"] is False
assert result["score_threshold_enabled"] is False
- def test_dataset_retrieval_model_dict_property_merges_partial_values(self):
+ def test_dataset_retrieval_model_dict_property_merges_partial_values(self) -> None:
"""Test retrieval_model_dict property fills in missing legacy keys."""
# Arrange
dataset = Dataset(
@@ -623,7 +626,7 @@ class TestDatasetModelValidation:
assert error.value.errors()[0]["loc"] == (field,)
- def test_dataset_gen_collection_name_by_id(self):
+ def test_dataset_gen_collection_name_by_id(self) -> None:
"""Test static method for generating collection name."""
# Arrange
dataset_id = "12345678-1234-1234-1234-123456789abc"
@@ -639,7 +642,7 @@ class TestDatasetModelValidation:
class TestDocumentModelRelationships:
"""Test suite for Document model relationships and properties."""
- def test_document_creation_with_required_fields(self):
+ def test_document_creation_with_required_fields(self) -> None:
"""Test creating a document with all required fields."""
# Arrange
tenant_id = str(uuid4())
@@ -669,7 +672,7 @@ class TestDocumentModelRelationships:
assert document.created_by == created_by
# Note: Default values are set by database, not by model instantiation
- def test_document_data_source_types(self):
+ def test_document_data_source_types(self) -> None:
"""Test document data source type validation."""
# Assert
assert "upload_file" in Document.DATA_SOURCES
@@ -677,7 +680,7 @@ class TestDocumentModelRelationships:
assert "website_crawl" in Document.DATA_SOURCES
@pytest.mark.parametrize("sqlite_session", [(Document, DatasetProcessRule, DocumentSegment)], indirect=True)
- def test_session_aware_document_getters_use_caller_session(self, sqlite_session: Session):
+ def test_session_aware_document_getters_use_caller_session(self, sqlite_session: Session) -> None:
process_rule = DatasetProcessRule(
dataset_id="dataset-1",
mode=ProcessRuleMode.CUSTOM,
@@ -706,7 +709,7 @@ class TestDocumentModelRelationships:
assert get_document_segment_count(document, session=sqlite_session) == 3
assert get_document_hit_count(document, session=sqlite_session) == 7
- def test_document_display_status_queuing(self):
+ def test_document_display_status_queuing(self) -> None:
"""Test document display_status property for queuing state."""
# Arrange
document = Document(
@@ -727,7 +730,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "queuing"
- def test_document_display_status_paused(self):
+ def test_document_display_status_paused(self) -> None:
"""Test document display_status property for paused state."""
# Arrange
document = Document(
@@ -749,7 +752,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "paused"
- def test_document_display_status_indexing(self):
+ def test_document_display_status_indexing(self) -> None:
"""Test document display_status property for indexing state."""
# Arrange
for indexing_status in [
@@ -776,7 +779,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "indexing"
- def test_document_display_status_error(self):
+ def test_document_display_status_error(self) -> None:
"""Test document display_status property for error state."""
# Arrange
document = Document(
@@ -797,7 +800,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "error"
- def test_document_display_status_available(self):
+ def test_document_display_status_available(self) -> None:
"""Test document display_status property for available state."""
# Arrange
document = Document(
@@ -820,7 +823,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "available"
- def test_document_display_status_disabled(self):
+ def test_document_display_status_disabled(self) -> None:
"""Test document display_status property for disabled state."""
# Arrange
document = Document(
@@ -843,7 +846,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "disabled"
- def test_document_display_status_archived(self):
+ def test_document_display_status_archived(self) -> None:
"""Test document display_status property for archived state."""
# Arrange
document = Document(
@@ -865,7 +868,7 @@ class TestDocumentModelRelationships:
# Assert
assert status == "archived"
- def test_document_data_source_info_dict_property(self):
+ def test_document_data_source_info_dict_property(self) -> None:
"""Test data_source_info_dict property parsing."""
# Arrange
data_source_info = {"upload_file_id": str(uuid4()), "file_name": "test.pdf"}
@@ -889,7 +892,7 @@ class TestDocumentModelRelationships:
assert "upload_file_id" in result
assert "file_name" in result
- def test_document_data_source_info_dict_property_empty(self):
+ def test_document_data_source_info_dict_property_empty(self) -> None:
"""Test data_source_info_dict property when data_source_info is None."""
# Arrange
document = Document(
@@ -910,7 +913,7 @@ class TestDocumentModelRelationships:
assert result == {}
@pytest.mark.parametrize("sqlite_session", [(Document, Dataset)], indirect=True)
- def test_document_get_dataset_uses_caller_session(self, sqlite_session: Session):
+ def test_document_get_dataset_uses_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset()
document = _make_document(dataset_id=dataset.id)
sqlite_session.add_all([dataset, document])
@@ -922,7 +925,7 @@ class TestDocumentModelRelationships:
def test_document_average_segment_length(
self,
sqlite_session: Session,
- ):
+ ) -> None:
"""Test average_segment_length property calculation."""
# Arrange
document = Document(
@@ -949,7 +952,7 @@ class TestDocumentModelRelationships:
# Assert
assert result == 100
- def test_document_average_segment_length_zero(self):
+ def test_document_average_segment_length_zero(self) -> None:
"""Test average_segment_length property when word_count is zero."""
# Arrange
document = Document(
@@ -977,7 +980,7 @@ class TestDocumentSegmentIndexing:
@pytest.mark.parametrize(
"sqlite_session", [(DocumentSegment, Document, DatasetProcessRule, ChildChunk)], indirect=True
)
- def test_get_child_chunks_uses_caller_session(self, sqlite_session: Session):
+ def test_get_child_chunks_uses_caller_session(self, sqlite_session: Session) -> None:
process_rule = DatasetProcessRule(
dataset_id="dataset-1",
mode=ProcessRuleMode.HIERARCHICAL,
@@ -1014,7 +1017,7 @@ class TestDocumentSegmentIndexing:
@pytest.mark.parametrize(
"sqlite_session", [(DocumentSegment, Document, DatasetProcessRule, ChildChunk)], indirect=True
)
- def test_get_child_chunks_includes_full_doc_unless_explicitly_hidden(self, sqlite_session: Session):
+ def test_get_child_chunks_includes_full_doc_unless_explicitly_hidden(self, sqlite_session: Session) -> None:
process_rule = DatasetProcessRule(
dataset_id="dataset-1",
mode=ProcessRuleMode.HIERARCHICAL,
@@ -1051,7 +1054,7 @@ class TestDocumentSegmentIndexing:
assert response_result == []
@pytest.mark.parametrize("sqlite_session", [(Dataset, Document, DocumentSegment)], indirect=True)
- def test_relationship_getters_use_caller_session(self, sqlite_session: Session):
+ def test_relationship_getters_use_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset()
document = _make_document(dataset_id=dataset.id)
segment = DocumentSegment(
@@ -1070,7 +1073,7 @@ class TestDocumentSegmentIndexing:
assert get_segment_dataset(segment, session=sqlite_session) is dataset
assert get_segment_document(segment, session=sqlite_session) is document
- def test_document_segment_creation_with_required_fields(self):
+ def test_document_segment_creation_with_required_fields(self) -> None:
"""Test creating a document segment with all required fields."""
# Arrange
tenant_id = str(uuid4())
@@ -1101,7 +1104,7 @@ class TestDocumentSegmentIndexing:
assert segment.created_by == created_by
# Note: Default values are set by database, not by model instantiation
- def test_document_segment_with_indexing_fields(self):
+ def test_document_segment_with_indexing_fields(self) -> None:
"""Test creating a document segment with indexing fields."""
# Arrange
index_node_id = str(uuid4())
@@ -1128,7 +1131,7 @@ class TestDocumentSegmentIndexing:
assert segment.index_node_hash == index_node_hash
assert segment.keywords == keywords
- def test_document_segment_with_answer_field(self):
+ def test_document_segment_with_answer_field(self) -> None:
"""Test creating a document segment with answer field for QA model."""
# Arrange
content = "What is AI?"
@@ -1151,7 +1154,7 @@ class TestDocumentSegmentIndexing:
assert segment.content == content
assert segment.answer == answer
- def test_document_segment_status_transitions(self):
+ def test_document_segment_status_transitions(self) -> None:
"""Test document segment status field values."""
# Arrange & Act
segment_waiting = DocumentSegment(
@@ -1181,7 +1184,7 @@ class TestDocumentSegmentIndexing:
assert segment_waiting.status == SegmentStatus.WAITING
assert segment_completed.status == SegmentStatus.COMPLETED
- def test_document_segment_enabled_disabled_tracking(self):
+ def test_document_segment_enabled_disabled_tracking(self) -> None:
"""Test document segment enabled/disabled state tracking."""
# Arrange
disabled_by = str(uuid4())
@@ -1207,7 +1210,7 @@ class TestDocumentSegmentIndexing:
assert segment.disabled_by == disabled_by
assert segment.disabled_at == disabled_at
- def test_document_segment_hit_count_tracking(self):
+ def test_document_segment_hit_count_tracking(self) -> None:
"""Test document segment hit count tracking."""
# Arrange & Act
segment = DocumentSegment(
@@ -1228,7 +1231,7 @@ class TestDocumentSegmentIndexing:
@pytest.mark.parametrize("sqlite_session", [(DocumentSegment, UploadFile, SegmentAttachmentBinding)], indirect=True)
def test_document_segment_attachments_prefers_files_url_for_source_url(
self, sqlite_session: Session, monkeypatch: pytest.MonkeyPatch
- ):
+ ) -> None:
"""Test attachment source URLs use FILES_URL before falling back to CONSOLE_API_URL."""
# Arrange
segment = DocumentSegment(
@@ -1289,7 +1292,7 @@ class TestDocumentSegmentIndexing:
assert query["nonce"] == ["01010101010101010101010101010101"]
assert query["sign"][0]
- def test_document_segment_error_tracking(self):
+ def test_document_segment_error_tracking(self) -> None:
"""Test document segment error tracking."""
# Arrange
error_message = "Indexing failed due to timeout"
@@ -1317,7 +1320,7 @@ class TestDocumentSegmentIndexing:
class TestEmbeddingStorage:
"""Test suite for Embedding model storage and retrieval."""
- def test_embedding_creation_with_required_fields(self):
+ def test_embedding_creation_with_required_fields(self) -> None:
"""Test creating an embedding with required fields."""
# Arrange
model_name = "text-embedding-ada-002"
@@ -1338,7 +1341,7 @@ class TestEmbeddingStorage:
assert embedding.provider_name == provider_name
assert embedding.embedding == b"binary_data"
- def test_embedding_set_and_get_embedding(self):
+ def test_embedding_set_and_get_embedding(self) -> None:
"""Test setting and getting embedding data."""
# Arrange
embedding_data = [0.1, 0.2, 0.3, 0.4, 0.5]
@@ -1359,7 +1362,7 @@ class TestEmbeddingStorage:
assert retrieved_data[0] == 0.1
assert retrieved_data[4] == 0.5
- def test_embedding_pickle_serialization(self):
+ def test_embedding_pickle_serialization(self) -> None:
"""Test embedding data is properly pickled."""
# Arrange
embedding_data = [0.1, 0.2, 0.3]
@@ -1380,7 +1383,7 @@ class TestEmbeddingStorage:
unpickled_data = pickle.loads(embedding.embedding) # noqa: S301
assert unpickled_data == embedding_data
- def test_embedding_with_large_vector(self):
+ def test_embedding_with_large_vector(self) -> None:
"""Test embedding with large dimension vector."""
# Arrange
# Simulate a 1536-dimension vector (OpenAI ada-002 size)
@@ -1405,7 +1408,7 @@ class TestEmbeddingStorage:
class TestDatasetProcessRule:
"""Test suite for DatasetProcessRule model."""
- def test_dataset_process_rule_creation(self):
+ def test_dataset_process_rule_creation(self) -> None:
"""Test creating a dataset process rule."""
# Arrange
dataset_id = str(uuid4())
@@ -1421,14 +1424,14 @@ class TestDatasetProcessRule:
assert process_rule.mode == ProcessRuleMode.AUTOMATIC
assert process_rule.created_by == created_by
- def test_dataset_process_rule_modes(self):
+ def test_dataset_process_rule_modes(self) -> None:
"""Test dataset process rule mode validation."""
# Assert
assert "automatic" in DatasetProcessRule.MODES
assert "custom" in DatasetProcessRule.MODES
assert "hierarchical" in DatasetProcessRule.MODES
- def test_dataset_process_rule_with_rules_dict(self):
+ def test_dataset_process_rule_with_rules_dict(self) -> None:
"""Test dataset process rule with rules dictionary."""
# Arrange
rules_data = {
@@ -1450,10 +1453,11 @@ class TestDatasetProcessRule:
# Assert
assert result == rules_data
+ assert result is not None
assert "pre_processing_rules" in result
assert "segmentation" in result
- def test_dataset_process_rule_to_dict(self):
+ def test_dataset_process_rule_to_dict(self) -> None:
"""Test dataset process rule to_dict method."""
# Arrange
dataset_id = str(uuid4())
@@ -1473,7 +1477,7 @@ class TestDatasetProcessRule:
assert result["mode"] == ProcessRuleMode.AUTOMATIC
assert result["rules"] == rules_data
- def test_dataset_process_rule_automatic_rules(self):
+ def test_dataset_process_rule_automatic_rules(self) -> None:
"""Test dataset process rule automatic rules constant."""
# Act
automatic_rules = DatasetProcessRule.AUTOMATIC_RULES
@@ -1487,7 +1491,7 @@ class TestDatasetProcessRule:
class TestDatasetKeywordTable:
"""Test suite for DatasetKeywordTable model."""
- def test_dataset_keyword_table_creation(self):
+ def test_dataset_keyword_table_creation(self) -> None:
"""Test creating a dataset keyword table."""
# Arrange
dataset_id = str(uuid4())
@@ -1503,7 +1507,7 @@ class TestDatasetKeywordTable:
assert keyword_table.dataset_id == dataset_id
assert keyword_table.data_source_type == "database" # Default value
- def test_dataset_keyword_table_data_source_type(self):
+ def test_dataset_keyword_table_data_source_type(self) -> None:
"""Test dataset keyword table data source type."""
# Arrange & Act
keyword_table = DatasetKeywordTable(
@@ -1516,7 +1520,7 @@ class TestDatasetKeywordTable:
assert keyword_table.data_source_type == "file"
@pytest.mark.parametrize("sqlite_session", [(Dataset, DatasetKeywordTable)], indirect=True)
- def test_get_keyword_table_dict_from_database_uses_caller_session(self, sqlite_session: Session):
+ def test_get_keyword_table_dict_from_database_uses_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset()
keyword_table = DatasetKeywordTable(
dataset_id=dataset.id,
@@ -1540,7 +1544,7 @@ class TestDatasetKeywordTable:
class TestAppDatasetJoin:
"""Test suite for AppDatasetJoin model."""
- def test_app_dataset_join_creation(self):
+ def test_app_dataset_join_creation(self) -> None:
"""Test creating an app-dataset join relationship."""
# Arrange
app_id = str(uuid4())
@@ -1561,7 +1565,7 @@ class TestAppDatasetJoin:
class TestChildChunk:
"""Test suite for ChildChunk model."""
- def test_child_chunk_creation(self):
+ def test_child_chunk_creation(self) -> None:
"""Test creating a child chunk."""
# Arrange
tenant_id = str(uuid4())
@@ -1593,7 +1597,7 @@ class TestChildChunk:
assert child_chunk.created_by == created_by
# Note: Default values are set by database, not by model instantiation
- def test_child_chunk_with_indexing_fields(self):
+ def test_child_chunk_with_indexing_fields(self) -> None:
"""Test creating a child chunk with indexing fields."""
# Arrange
index_node_id = str(uuid4())
@@ -1621,7 +1625,7 @@ class TestChildChunk:
class TestModelIntegration:
"""Test suite for model integration scenarios."""
- def test_complete_dataset_document_segment_hierarchy(self):
+ def test_complete_dataset_document_segment_hierarchy(self) -> None:
"""Test complete hierarchy from dataset to segment."""
# Arrange
tenant_id = str(uuid4())
@@ -1679,7 +1683,7 @@ class TestModelIntegration:
def test_document_to_dict_serialization(
self,
sqlite_session: Session,
- ):
+ ) -> None:
"""Test document to_dict method for serialization."""
# Arrange
tenant_id = str(uuid4())
@@ -1727,7 +1731,7 @@ class TestChildChunkSessionAccessors:
global ``db.session`` internally; they are now plain methods accepting a ``Session`` explicitly.
"""
- def test_accessors_resolve_related_rows_via_caller_session(self, sqlite_session: Session):
+ def test_accessors_resolve_related_rows_via_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset(dataset_id=str(uuid4()), tenant_id=str(uuid4()))
document = _make_document(document_id=str(uuid4()), dataset_id=dataset.id, tenant_id=dataset.tenant_id)
segment = _make_segments(document, [0])[0]
@@ -1748,7 +1752,7 @@ class TestChildChunkSessionAccessors:
assert get_child_document(child_chunk, session=sqlite_session) is document
assert get_child_segment(child_chunk, session=sqlite_session) is segment
- def test_accessors_return_none_when_related_rows_are_absent(self, sqlite_session: Session):
+ def test_accessors_return_none_when_related_rows_are_absent(self, sqlite_session: Session) -> None:
child_chunk = ChildChunk(
tenant_id=str(uuid4()),
dataset_id=str(uuid4()),
@@ -1772,7 +1776,7 @@ class TestDatasetAvailableDocumentCount:
caller-provided session, matching the other ``get_*`` accessors on ``Dataset``.
"""
- def test_counts_only_completed_enabled_documents(self, sqlite_session: Session):
+ def test_counts_only_completed_enabled_documents(self, sqlite_session: Session) -> None:
dataset = _make_dataset(dataset_id=str(uuid4()), tenant_id=str(uuid4()))
completed = _make_document(
document_id=str(uuid4()),
@@ -1801,7 +1805,7 @@ class TestDocumentSegmentNeighborAccessors:
plain methods accepting a ``Session`` explicitly.
"""
- def test_accessors_resolve_neighboring_segments_via_caller_session(self, sqlite_session: Session):
+ def test_accessors_resolve_neighboring_segments_via_caller_session(self, sqlite_session: Session) -> None:
dataset = _make_dataset(dataset_id=str(uuid4()), tenant_id=str(uuid4()))
document = _make_document(document_id=str(uuid4()), dataset_id=dataset.id, tenant_id=dataset.tenant_id)
segments = _make_segments(document, [0, 0, 0])
@@ -1812,7 +1816,7 @@ class TestDocumentSegmentNeighborAccessors:
assert get_previous_segment(middle_segment, session=sqlite_session) is segments[0]
assert get_next_segment(middle_segment, session=sqlite_session) is segments[2]
- def test_accessors_return_none_when_neighbors_are_absent(self, sqlite_session: Session):
+ def test_accessors_return_none_when_neighbors_are_absent(self, sqlite_session: Session) -> None:
dataset = _make_dataset(dataset_id=str(uuid4()), tenant_id=str(uuid4()))
document = _make_document(document_id=str(uuid4()), dataset_id=dataset.id, tenant_id=dataset.tenant_id)
segment = _make_segments(document, [0])[0]
@@ -1830,7 +1834,7 @@ class TestAppDatasetJoinSessionAccessors:
it is now a plain method accepting a ``Session`` explicitly.
"""
- def test_app_accessor_resolves_app_via_caller_session(self, sqlite_session: Session):
+ def test_app_accessor_resolves_app_via_caller_session(self, sqlite_session: Session) -> None:
app = _make_app(app_id=str(uuid4()))
join = AppDatasetJoin(
app_id=app.id,
@@ -1841,7 +1845,7 @@ class TestAppDatasetJoinSessionAccessors:
assert get_joined_app(join, session=sqlite_session) is app
- def test_app_accessor_returns_none_when_app_absent(self, sqlite_session: Session):
+ def test_app_accessor_returns_none_when_app_absent(self, sqlite_session: Session) -> None:
join = AppDatasetJoin(
app_id=str(uuid4()),
dataset_id=str(uuid4()),
diff --git a/api/tests/unit_tests/pyrefly.toml b/api/tests/unit_tests/pyrefly.toml
index 017191be198..345cbdad1a0 100644
--- a/api/tests/unit_tests/pyrefly.toml
+++ b/api/tests/unit_tests/pyrefly.toml
@@ -267,7 +267,6 @@ project-excludes = [
"migrations/versions/test_2025_07_02_2332-1c9ba48be8e4_add_uuidv7_function_in_sql.py",
"models/test_account.py",
"models/test_model.py",
- "models/test_dataset.py",
"models/test_tools.py",
"services/agent/test_dsl_service.py",
"services/agent/test_roster_service.py",