chore: Enable Ruff checks in helper private modules (#2084)

pyproject.toml

@@ -151,9 +151,6 @@ xdoctest = ">=1.1.1"
 [tool.poetry.group.benchmark.dependencies]
 pytest-codspeed = ">=2.2.0"
 
-[tool.black]
-exclude = ".*simpleeval.*"
-
 [tool.pytest.ini_options]
 addopts = '--ignore=singer_sdk/helpers/_simpleeval.py -m "not external"'
 markers = [
@@ -247,8 +244,7 @@ target-version = "py37"
 [tool.ruff.lint]
 exclude = [
     "cookiecutter/*",
-    "singer_sdk/helpers/_simpleeval.py",
-    "tests/core/test_simpleeval.py",
+    "*simpleeval*",
 ]
 ignore = [
     "ANN101",  # Missing type annotation for `self` in method
@@ -318,8 +314,6 @@ unfixable = [
 "tests/*" = ["ANN", "D1", "D2", "FBT001", "FBT003", "PLR2004", "S101"]
 # Disabled some checks in samples code
 "samples/*" = ["ANN", "D"]
-# Don't require docstrings conventions or type annotations in private modules
-"singer_sdk/helpers/_*.py" = ["ANN", "D105"]
 # Templates support a generic resource of type Any.
 "singer_sdk/testing/*.py" = ["S101"]
 "singer_sdk/testing/templates.py" = ["ANN401"]

singer_sdk/helpers/_batch.py

@@ -92,7 +92,7 @@ class SDKBatchMessage(Message):
     manifest: list[str] = field(default_factory=list)
     """The manifest of files in the batch."""
 
-    def __post_init__(self):
+    def __post_init__(self) -> None:
         if isinstance(self.encoding, dict):
             self.encoding = BaseBatchFileEncoding.from_dict(self.encoding)
 
@@ -112,7 +112,7 @@ class StorageTarget:
     params: dict = field(default_factory=dict)
     """"The storage parameters."""
 
-    def asdict(self):
+    def asdict(self) -> dict[str, t.Any]:
         """Return a dictionary representation of the message.
 
         Returns:
@@ -144,7 +144,7 @@ def split_url(url: str) -> tuple[str, str]:
         """
         if platform.system() == "Windows" and "\\" in url:
             # Original code from pyFileSystem split
-            # Augemnted slitly to properly Windows paths
+            # Augmented slightly to properly handle Windows paths
             split = url.rsplit("\\", 1)
             return (split[0] or "\\", split[1])
 
@@ -224,7 +224,7 @@ class BatchConfig:
     batch_size: int = DEFAULT_BATCH_SIZE
     """The max number of records in a batch."""
 
-    def __post_init__(self):
+    def __post_init__(self) -> None:
         if isinstance(self.encoding, dict):
             self.encoding = BaseBatchFileEncoding.from_dict(self.encoding)
 
@@ -234,7 +234,7 @@ def __post_init__(self):
         if self.batch_size is None:
             self.batch_size = DEFAULT_BATCH_SIZE
 
-    def asdict(self):
+    def asdict(self) -> dict[str, t.Any]:
         """Return a dictionary representation of the message.
 
         Returns:

singer_sdk/helpers/_flattening.py

@@ -1,4 +1,4 @@
-"""Internal helper library for record flatteting functions."""
+"""Internal helper library for record flattening functions."""
 
 from __future__ import annotations
 
@@ -70,7 +70,7 @@ def flatten_key(key_name: str, parent_keys: list[str], separator: str = "__") ->
             inflection.camelize(inflected_key[reducer_index]),
         )
         inflected_key[reducer_index] = (
-            reduced_key if len(reduced_key) > 1 else inflected_key[reducer_index][0:3]
+            reduced_key if len(reduced_key) > 1 else inflected_key[reducer_index][:3]
         ).lower()
         reducer_index += 1
 
@@ -358,8 +358,8 @@ def _flatten_schema(  # noqa: C901, PLR0912
                 items.append((new_key, next(iter(field_schema.values()))[0]))
 
     # Sort and check for duplicates
-    def _key_func(item):
-        return item[0]  # first item is tuple is the key name.
+    def _key_func(item: tuple[str, dict]) -> str:
+        return item[0]  # first item in tuple is the key name.
 
     sorted_items = sorted(items, key=_key_func)
     for field_name, g in itertools.groupby(sorted_items, key=_key_func):
@@ -451,7 +451,11 @@ def _flatten_record(
     return dict(items)
 
 
-def _should_jsondump_value(key: str, value: t.Any, flattened_schema=None) -> bool:
+def _should_jsondump_value(
+    key: str,
+    value: t.Any,  # noqa: ANN401
+    flattened_schema: dict[str, t.Any] | None = None,
+) -> bool:
     """Return True if json.dump() should be used to serialize the value.
 
     Args:

singer_sdk/helpers/_secrets.py

@@ -28,7 +28,7 @@ def is_common_secret_key(key_name: str) -> bool:
 class SecretString(str):
     """For now, this class wraps a sensitive string to be identified as such later."""
 
-    def __init__(self, contents):
+    def __init__(self, contents: str) -> None:
         """Initialize secret string."""
         self.contents = contents
 

singer_sdk/helpers/_state.py

@@ -23,7 +23,7 @@ def get_state_if_exists(
     tap_stream_id: str,
     state_partition_context: dict | None = None,
     key: str | None = None,
-) -> t.Any | None:
+) -> t.Any | None:  # noqa: ANN401
     """Return the stream or partition state, creating a new one if it does not exist.
 
     Args:
@@ -135,10 +135,10 @@ def get_writeable_state_dict(
 
 
 def write_stream_state(
-    tap_state,
+    tap_state: dict,
     tap_stream_id: str,
-    key,
-    val,
+    key: str,
+    val: t.Any,  # noqa: ANN401
     *,
     state_partition_context: dict | None = None,
 ) -> None:
@@ -165,21 +165,21 @@ def reset_state_progress_markers(stream_or_partition_state: dict) -> dict | None
 
 def write_replication_key_signpost(
     stream_or_partition_state: dict,
-    new_signpost_value: t.Any,
+    new_signpost_value: t.Any,  # noqa: ANN401
 ) -> None:
     """Write signpost value."""
     stream_or_partition_state[SIGNPOST_MARKER] = to_json_compatible(new_signpost_value)
 
 
 def write_starting_replication_value(
     stream_or_partition_state: dict,
-    initial_value: t.Any,
+    initial_value: t.Any,  # noqa: ANN401
 ) -> None:
     """Write initial replication value to state."""
     stream_or_partition_state[STARTING_MARKER] = to_json_compatible(initial_value)
 
 
-def get_starting_replication_value(stream_or_partition_state: dict):
+def get_starting_replication_value(stream_or_partition_state: dict) -> t.Any | None:  # noqa: ANN401
     """Retrieve initial replication marker value from state."""
     if not stream_or_partition_state:
         return None

singer_sdk/helpers/_typing.py

@@ -39,7 +39,7 @@ def __init__(self, *args: object) -> None:
         super().__init__(msg, *args)
 
 
-def to_json_compatible(val: t.Any) -> t.Any:
+def to_json_compatible(val: t.Any) -> t.Any:  # noqa: ANN401
     """Return as string if datetime. JSON does not support proper datetime types.
 
     If given a naive datetime object, pendulum automatically makes it utc
@@ -185,7 +185,7 @@ def get_datelike_property_type(property_schema: dict) -> str | None:
     return None
 
 
-def _is_string_with_format(type_dict):
+def _is_string_with_format(type_dict: dict[str, t.Any]) -> bool | None:
     if "string" in type_dict.get("type", []) and type_dict.get("format") in {
         "date-time",
         "time",
@@ -196,14 +196,14 @@ def _is_string_with_format(type_dict):
 
 
 def handle_invalid_timestamp_in_record(
-    record,  # noqa: ARG001
+    record: dict[str, t.Any],  # noqa: ARG001
     key_breadcrumb: list[str],
     invalid_value: str,
     datelike_typename: str,
     ex: Exception,
     treatment: DatetimeErrorTreatmentEnum | None,
     logger: logging.Logger,
-) -> t.Any:
+) -> t.Any:  # noqa: ANN401
     """Apply treatment or raise an error for invalid time values."""
     treatment = treatment or DatetimeErrorTreatmentEnum.ERROR
     msg = (
@@ -331,7 +331,7 @@ def _warn_unmapped_properties(
     stream_name: str,
     property_names: tuple[str],
     logger: logging.Logger,
-):
+) -> None:
     logger.warning(
         "Properties %s were present in the '%s' stream but "
         "not found in catalog schema. Ignoring.",
@@ -470,9 +470,9 @@ def _conform_record_data_types(  # noqa: PLR0912
 
 
 def _conform_primitive_property(  # noqa: PLR0911
-    elem: t.Any,
+    elem: t.Any,  # noqa: ANN401
     property_schema: dict,
-) -> t.Any:
+) -> t.Any:  # noqa: ANN401
     """Converts a primitive (i.e. not object or array) to a json compatible type."""
     if isinstance(elem, (datetime.datetime, pendulum.DateTime)):
         return to_json_compatible(elem)