add docstrings to all indexing pipeline tests

surfsense_backend/tests/unit/indexing_pipeline/test_connector_document.py

@@ -6,6 +6,7 @@ from app.indexing_pipeline.connector_document import ConnectorDocument
 
 
 def test_valid_document_created_with_required_fields():
+    """All optional fields default correctly when only required fields are supplied."""
     doc = ConnectorDocument(
         title="Task",
         source_markdown="## Task\n\nSome content.",
@@ -23,6 +24,7 @@ def test_valid_document_created_with_required_fields():
 
 
 def test_omitting_created_by_id_raises():
+    """Omitting created_by_id raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",
@@ -35,6 +37,7 @@ def test_omitting_created_by_id_raises():
 
 
 def test_empty_source_markdown_raises():
+    """Empty source_markdown raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",
@@ -46,6 +49,7 @@ def test_empty_source_markdown_raises():
 
 
 def test_whitespace_only_source_markdown_raises():
+    """Whitespace-only source_markdown raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",
@@ -57,6 +61,7 @@ def test_whitespace_only_source_markdown_raises():
 
 
 def test_empty_title_raises():
+    """Empty title raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="",
@@ -68,6 +73,7 @@ def test_empty_title_raises():
 
 
 def test_empty_created_by_id_raises():
+    """Empty created_by_id raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",
@@ -81,6 +87,7 @@ def test_empty_created_by_id_raises():
 
 
 def test_zero_search_space_id_raises():
+    """search_space_id of zero raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",
@@ -94,6 +101,7 @@ def test_zero_search_space_id_raises():
 
 
 def test_empty_unique_id_raises():
+    """Empty unique_id raises a validation error."""
     with pytest.raises(ValidationError):
         ConnectorDocument(
             title="Task",

surfsense_backend/tests/unit/indexing_pipeline/test_document_chunker.py

@@ -7,6 +7,7 @@ pytestmark = pytest.mark.unit
 
 @pytest.mark.usefixtures("patched_chunker_instance", "patched_code_chunker_instance")
 def test_uses_code_chunker_when_flag_is_true():
+    """Code chunker is selected when use_code_chunker=True."""
     result = chunk_text("def foo(): pass", use_code_chunker=True)
 
     assert result == ["code chunk"]
@@ -14,6 +15,7 @@ def test_uses_code_chunker_when_flag_is_true():
 
 @pytest.mark.usefixtures("patched_chunker_instance", "patched_code_chunker_instance")
 def test_uses_default_chunker_when_flag_is_false():
+    """Default prose chunker is selected when use_code_chunker=False."""
     result = chunk_text("Some prose text.", use_code_chunker=False)
 
     assert result == ["prose chunk"]

surfsense_backend/tests/unit/indexing_pipeline/test_document_hashing.py

@@ -7,36 +7,42 @@ pytestmark = pytest.mark.unit
 
 
 def test_different_unique_id_produces_different_hash(make_connector_document):
+    """Two documents with different unique_ids produce different identifier hashes."""
     doc_a = make_connector_document(unique_id="id-001")
     doc_b = make_connector_document(unique_id="id-002")
     assert compute_unique_identifier_hash(doc_a) != compute_unique_identifier_hash(doc_b)
 
 
 def test_different_search_space_produces_different_identifier_hash(make_connector_document):
+    """Same document in different search spaces produces different identifier hashes."""
     doc_a = make_connector_document(search_space_id=1)
     doc_b = make_connector_document(search_space_id=2)
     assert compute_unique_identifier_hash(doc_a) != compute_unique_identifier_hash(doc_b)
 
 
 def test_different_document_type_produces_different_identifier_hash(make_connector_document):
+    """Same unique_id with different document types produces different identifier hashes."""
     doc_a = make_connector_document(document_type=DocumentType.CLICKUP_CONNECTOR)
     doc_b = make_connector_document(document_type=DocumentType.NOTION_CONNECTOR)
     assert compute_unique_identifier_hash(doc_a) != compute_unique_identifier_hash(doc_b)
 
 
 def test_same_content_same_space_produces_same_content_hash(make_connector_document):
+    """Identical content in the same search space always produces the same content hash."""
     doc_a = make_connector_document(source_markdown="Hello world", search_space_id=1)
     doc_b = make_connector_document(source_markdown="Hello world", search_space_id=1)
     assert compute_content_hash(doc_a) == compute_content_hash(doc_b)
 
 
 def test_same_content_different_space_produces_different_content_hash(make_connector_document):
+    """Identical content in different search spaces produces different content hashes."""
     doc_a = make_connector_document(source_markdown="Hello world", search_space_id=1)
     doc_b = make_connector_document(source_markdown="Hello world", search_space_id=2)
     assert compute_content_hash(doc_a) != compute_content_hash(doc_b)
 
 
 def test_different_content_produces_different_content_hash(make_connector_document):
+    """Different source markdown produces different content hashes."""
     doc_a = make_connector_document(source_markdown="Original content")
     doc_b = make_connector_document(source_markdown="Updated content")
     assert compute_content_hash(doc_a) != compute_content_hash(doc_b)

surfsense_backend/tests/unit/indexing_pipeline/test_document_summarizer.py

@@ -8,6 +8,7 @@ pytestmark = pytest.mark.unit
 
 @pytest.mark.usefixtures("patched_summarizer_chain")
 async def test_without_metadata_returns_raw_summary():
+    """Summarizer returns the LLM output directly when no metadata is provided."""
     result = await summarize_document("# Content", llm=MagicMock(model="gpt-4"))
 
     assert result == "The summary."
@@ -15,6 +16,7 @@ async def test_without_metadata_returns_raw_summary():
 
 @pytest.mark.usefixtures("patched_summarizer_chain")
 async def test_with_metadata_includes_metadata_values_in_output():
+    """Non-empty metadata values are prepended to the summary output."""
     result = await summarize_document(
         "# Content",
         llm=MagicMock(model="gpt-4"),
@@ -27,6 +29,7 @@ async def test_with_metadata_includes_metadata_values_in_output():
 
 @pytest.mark.usefixtures("patched_summarizer_chain")
 async def test_with_metadata_omits_empty_fields_from_output():
+    """Empty metadata fields are omitted from the summary output."""
     result = await summarize_document(
         "# Content",
         llm=MagicMock(model="gpt-4"),