Ruff ANN201: fixed return value type annotations

Author: evertlammerts

duckdb/experimental/spark/sql/column.py

@@ -205,18 +205,18 @@ def __getattr__(self, item: Any) -> "Column":
             raise AttributeError(msg)
         return self[item]
 
-    def alias(self, alias: str):  # noqa: D102
+    def alias(self, alias: str) -> "Column":  # noqa: D102
         return Column(self.expr.alias(alias))
 
-    def when(self, condition: "Column", value: Any):  # noqa: D102
+    def when(self, condition: "Column", value: Any) -> "Column":  # noqa: D102
         if not isinstance(condition, Column):
             msg = "condition should be a Column"
             raise TypeError(msg)
         v = _get_expr(value)
         expr = self.expr.when(condition.expr, v)
         return Column(expr)
 
-    def otherwise(self, value: Any):  # noqa: D102
+    def otherwise(self, value: Any) -> "Column":  # noqa: D102
         v = _get_expr(value)
         expr = self.expr.otherwise(v)
         return Column(expr)

duckdb/experimental/spark/sql/dataframe.py

@@ -989,7 +989,7 @@ def groupBy(self, *cols: "ColumnOrName") -> "GroupedData":  # type: ignore[misc]
     def write(self) -> DataFrameWriter:  # noqa: D102
         return DataFrameWriter(self)
 
-    def printSchema(self):  # noqa: D102
+    def printSchema(self) -> None:  # noqa: D102
         raise ContributionsAcceptedError
 
     def union(self, other: "DataFrame") -> "DataFrame":

duckdb/experimental/spark/sql/functions.py

@@ -30,7 +30,7 @@ def _invoke_function_over_columns(name: str, *cols: "ColumnOrName") -> Column:
     return _invoke_function(name, *cols)
 
 
-def col(column: str):  # noqa: D103
+def col(column: str) -> Column:  # noqa: D103
     return Column(ColumnExpression(column))
 
 

duckdb/experimental/spark/sql/readwriter.py

@@ -55,7 +55,7 @@ def csv(  # noqa: D102
         encoding: Optional[str] = None,
         emptyValue: Optional[str] = None,
         lineSep: Optional[str] = None,
-    ):
+    ) -> None:
         if mode not in (None, "overwrite"):
             raise NotImplementedError
         if escapeQuotes:

duckdb/filesystem.py

@@ -7,7 +7,7 @@
 from .bytes_io_wrapper import BytesIOWrapper
 
 
-def is_file_like(obj):  # noqa: D103, ANN001
+def is_file_like(obj) -> bool:  # noqa: D103, ANN001
     # We only care that we can read from the file
     return hasattr(obj, "read") and hasattr(obj, "seek")
 
@@ -17,7 +17,7 @@ class ModifiedMemoryFileSystem(MemoryFileSystem):  # noqa: D101
     # defer to the original implementation that doesn't hardcode the protocol
     _strip_protocol = classmethod(AbstractFileSystem._strip_protocol.__func__)
 
-    def add_file(self, object: IO, path: str):  # noqa: D102
+    def add_file(self, object: IO, path: str) -> None:  # noqa: D102
         if not is_file_like(object):
             msg = "Can not read from a non file-like object"
             raise ValueError(msg)

duckdb/query_graph/__main__.py

@@ -57,15 +57,15 @@
   text-align: center;
   padding: 0px;
   border-radius: 1px;
-  
+
   /* Positioning */
   position: absolute;
   z-index: 1;
   bottom: 100%;
   left: 50%;
   transform: translateX(-50%);
   margin-bottom: 8px;
-  
+
   /* Tooltip Arrow */
   width: 400px;
 }
@@ -77,7 +77,7 @@
 
 
 class NodeTiming:  # noqa: D101
-    def __init__(self, phase: str, time: float) -> object:  # noqa: D107
+    def __init__(self, phase: str, time: float) -> None:  # noqa: D107
         self.phase = phase
         self.time = time
         # percentage is determined later.
@@ -86,7 +86,7 @@ def __init__(self, phase: str, time: float) -> object:  # noqa: D107
     def calculate_percentage(self, total_time: float) -> None:  # noqa: D102
         self.percentage = self.time / total_time
 
-    def combine_timing(l: object, r: object) -> object:  # noqa: D102
+    def combine_timing(l: "NodeTiming", r: "NodeTiming") -> "NodeTiming":  # noqa: D102
         # TODO: can only add timings for same-phase nodes
         total_time = l.time + r.time
         return NodeTiming(l.phase, total_time)
@@ -96,25 +96,25 @@ class AllTimings:  # noqa: D101
     def __init__(self) -> None:  # noqa: D107
         self.phase_to_timings = {}
 
-    def add_node_timing(self, node_timing: NodeTiming):  # noqa: D102
+    def add_node_timing(self, node_timing: NodeTiming) -> None:  # noqa: D102
         if node_timing.phase in self.phase_to_timings:
             self.phase_to_timings[node_timing.phase].append(node_timing)
-            return
-        self.phase_to_timings[node_timing.phase] = [node_timing]
+        else:
+            self.phase_to_timings[node_timing.phase] = [node_timing]
 
-    def get_phase_timings(self, phase: str):  # noqa: D102
+    def get_phase_timings(self, phase: str) -> list[NodeTiming]:  # noqa: D102
         return self.phase_to_timings[phase]
 
-    def get_summary_phase_timings(self, phase: str):  # noqa: D102
+    def get_summary_phase_timings(self, phase: str) -> NodeTiming:  # noqa: D102
         return reduce(NodeTiming.combine_timing, self.phase_to_timings[phase])
 
-    def get_phases(self):  # noqa: D102
+    def get_phases(self) -> list[NodeTiming]:  # noqa: D102
         phases = list(self.phase_to_timings.keys())
         phases.sort(key=lambda x: (self.get_summary_phase_timings(x)).time)
         phases.reverse()
         return phases
 
-    def get_sum_of_all_timings(self):  # noqa: D102
+    def get_sum_of_all_timings(self) -> float:  # noqa: D102
         total_timing_sum = 0
         for phase in self.phase_to_timings.keys():
             total_timing_sum += self.get_summary_phase_timings(phase).time
@@ -132,7 +132,7 @@ def get_child_timings(top_node: object, query_timings: object) -> str:  # noqa:
         get_child_timings(child, query_timings)
 
 
-def get_pink_shade_hex(fraction: float):  # noqa: D103
+def get_pink_shade_hex(fraction: float) -> str:  # noqa: D103
     fraction = max(0, min(1, fraction))
 
     # Define the RGB values for very light pink (almost white) and dark pink
@@ -211,7 +211,7 @@ def generate_timing_html(graph_json: object, query_timings: object) -> object:
     gather_timing_information(json_graph, query_timings)
     total_time = float(json_graph.get("operator_timing") or json_graph.get("latency"))
     table_head = """
-	<table class=\"styled-table\"> 
+	<table class=\"styled-table\">
 		<thead>
 			<tr>
 				<th>Phase</th>

duckdb/udf.py

@@ -1,7 +1,7 @@
 from typing import Callable
 
 
-def vectorized(func: Callable):
+def vectorized(func: Callable) -> Callable:
     """Decorate a function with annotated function parameters, so DuckDB can infer that the function should be provided with pyarrow arrays and should expect pyarrow array(s) as output."""
     import types
     from inspect import signature

duckdb_packaging/pypi_cleanup.py

@@ -178,31 +178,25 @@ def __init__(self, target: str, contains_input: bool | None = None) -> None:  #
         self._in_form = False  # Currently parsing a form with an action we're interested in
         self._input_contained = False  # Input field requested is contained in the current form
 
-    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]):  # noqa: D102
+    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:  # noqa: D102
         if tag == "form":
             attrs = dict(attrs)
             action = attrs.get("action")  # Might be None.
             if action and (action == self._target or action.startswith(self._target)):
                 self._in_form = True
-            return
-
-        if self._in_form and tag == "input":
+        elif self._in_form and tag == "input":
             attrs = dict(attrs)
             if attrs.get("name") == "csrf_token":
                 self._csrf = attrs["value"]
-
             if self._contains_input and attrs.get("name") == self._contains_input:
                 self._input_contained = True
 
-            return
-
-    def handle_endtag(self, tag: str):  # noqa: D102
+    def handle_endtag(self, tag: str) -> None:  # noqa: D102
         if tag == "form":
             self._in_form = False
             # If we're in a right form that contains the requested input and csrf is not set
             if (not self._contains_input or self._input_contained) and not self.csrf:
                 self.csrf = self._csrf
-            return
 
 
 class PyPICleanup:

duckdb_packaging/setuptools_scm_version.py

@@ -76,7 +76,7 @@ def _bump_version(base_version: str, distance: int, dirty: bool = False) -> str:
     return f"{format_version(major, minor, patch + 1)}.dev{distance}"
 
 
-def forced_version_from_env():
+def forced_version_from_env() -> str:
     """Handle getting versions from environment variables.
 
     Only supports a single way of manually overriding the version through
@@ -132,7 +132,7 @@ def _git_describe_override_to_pep_440(override_value: str) -> str:
     return pep440_version
 
 
-def _remove_unsupported_env_var(env_var: str):
+def _remove_unsupported_env_var(env_var: str) -> None:
     """Remove an unsupported environment variable with a warning."""
     print(f"[versioning] WARNING: We do not support {env_var}! Removing.")
     del os.environ[env_var]