Ruff TD002 and TD003: added noqa for author and issue links in existing TODOs

evertlammerts · evertlammerts · commit 2ac017620a2d · 2025-09-18T10:46:31.000+02:00
diff --git a/duckdb/experimental/spark/sql/column.py b/duckdb/experimental/spark/sql/column.py
@@ -173,7 +173,7 @@ def __getitem__(self, k: Any) -> "Column":  # noqa: ANN401
             #    raise ValueError("Using a slice with a step value is not supported")
             # return self.substr(k.start, k.stop)
         else:
-            # TODO: this is super hacky
+            # TODO: this is super hacky  # noqa: TD002, TD003
             expr_str = str(self.expr) + "." + str(k)
             return Column(ColumnExpression(expr_str))
 
diff --git a/duckdb/experimental/spark/sql/functions.py b/duckdb/experimental/spark/sql/functions.py
@@ -704,7 +704,7 @@ def asin(col: "ColumnOrName") -> Column:
     +--------+
     """
     col = _to_column_expr(col)
-    # TODO: ConstantExpression(float("nan")) gives NULL and not NaN
+    # TODO: ConstantExpression(float("nan")) gives NULL and not NaN  # noqa: TD002, TD003
     return Column(
         CaseExpression((col < -1.0) | (col > 1.0), ConstantExpression(float("nan"))).otherwise(
             FunctionExpression("asin", col)
diff --git a/duckdb/experimental/spark/sql/group.py b/duckdb/experimental/spark/sql/group.py
@@ -418,4 +418,4 @@ def agg(self, *exprs: Union[Column, dict[str, str]]) -> DataFrame:
             rel = self._df.relation.select(*expressions, groups=group_by)
         return DataFrame(rel, self.session)
 
-    # TODO: add 'pivot'
+    # TODO: add 'pivot'  # noqa: TD002, TD003
diff --git a/duckdb/experimental/spark/sql/type_utils.py b/duckdb/experimental/spark/sql/type_utils.py
@@ -79,7 +79,7 @@ def convert_nested_type(dtype: DuckDBPyType) -> DataType:  # noqa: D103
     if id == "list" or id == "array":
         children = dtype.children
         return ArrayType(convert_type(children[0][1]))
-    # TODO: add support for 'union'
+    # TODO: add support for 'union'  # noqa: TD002, TD003
     if id == "struct":
         children: list[tuple[str, DuckDBPyType]] = dtype.children
         fields = [StructField(x[0], convert_type(x[1])) for x in children]
diff --git a/duckdb/query_graph/__main__.py b/duckdb/query_graph/__main__.py
@@ -87,7 +87,7 @@ def calculate_percentage(self, total_time: float) -> None:  # noqa: D102
         self.percentage = self.time / total_time
 
     def combine_timing(l: "NodeTiming", r: "NodeTiming") -> "NodeTiming":  # noqa: D102
-        # TODO: can only add timings for same-phase nodes
+        # TODO: can only add timings for same-phase nodes  # noqa: TD002, TD003
         total_time = l.time + r.time
         return NodeTiming(l.phase, total_time)
 
@@ -161,7 +161,7 @@ def get_node_body(name: str, result: str, cpu_time: float, card: int, est: int,
         body += f"<p>cardinality: {card}</p>"
         body += f"<p>estimate: {est}</p>"
         body += f"<p>width: {width} bytes</p>"
-    # TODO: Expand on timing. Usually available from a detailed profiling
+    # TODO: Expand on timing. Usually available from a detailed profiling  # noqa: TD002, TD003
     body += "</div>"
     body += "</span>"
     return body
@@ -250,7 +250,7 @@ def generate_tree_html(graph_json: object) -> str:  # noqa: D103
     tree_prefix = '<div class="tf-tree tf-gap-sm"> \n <ul>'
     tree_suffix = "</ul> </div>"
     # first level of json is general overview
-    # TODO: make sure json output first level always has only 1 level
+    # TODO: make sure json output first level always has only 1 level  # noqa: TD002, TD003
     tree_body = generate_tree_recursive(json_graph["children"][0], cpu_time)
     return tree_prefix + tree_body + tree_suffix
 
diff --git a/duckdb/value/constant/__init__.py b/duckdb/value/constant/__init__.py
@@ -235,7 +235,7 @@ def __init__(self, object: Any, members: dict[str, DuckDBPyType]) -> None:
         super().__init__(object, union_type)
 
 
-# TODO: add EnumValue once `duckdb.enum_type` is added
+# TODO: add EnumValue once `duckdb.enum_type` is added  # noqa: TD002, TD003
 
 __all__ = [
     "BinaryValue",
diff --git a/tests/coverage/test_pandas_categorical_coverage.py b/tests/coverage/test_pandas_categorical_coverage.py
@@ -68,7 +68,7 @@ def check_create_table(category, pandas):
     conn.execute("DROP TABLE t1")
 
 
-# TODO: extend tests with ArrowPandas
+# TODO: extend tests with ArrowPandas  # noqa: TD002, TD003
 class TestCategory:
     @pytest.mark.parametrize("pandas", [NumpyPandas()])
     def test_category_string_uint16(self, duckdb_cursor, pandas):
diff --git a/tests/extensions/test_httpfs.py b/tests/extensions/test_httpfs.py
@@ -8,7 +8,7 @@
 import duckdb
 
 # We only run this test if this env var is set
-# TODO: we can add a custom command line argument to pytest to provide an extension directory
+# TODO: we can add a custom command line argument to pytest to provide an extension directory  # noqa: TD002, TD003
 # We can use that instead of checking this environment variable inside of conftest.py's 'require' method
 pytestmark = mark.skipif(
     not os.getenv("DUCKDB_PYTHON_TEST_EXTENSION_REQUIRED", False),
diff --git a/tests/fast/api/test_duckdb_connection.py b/tests/fast/api/test_duckdb_connection.py
@@ -123,7 +123,7 @@ def test_execute(self):
 
     def test_executemany(self):
         # executemany does not keep an open result set
-        # TODO: shouldn't we also have a version that executes a query multiple times with different parameters,
+        # TODO: shouldn't we also have a version that executes a query multiple times with different parameters,  # noqa: TD002, TD003
         #  returning all of the results?
         duckdb.execute("create table tbl (i integer, j varchar)")
         duckdb.executemany("insert into tbl VALUES (?, ?)", [(5, "test"), (2, "duck"), (42, "quack")])
diff --git a/tests/fast/api/test_duckdb_execute.py b/tests/fast/api/test_duckdb_execute.py
@@ -41,7 +41,7 @@ def generator(rowcount):
                 yield min(2048, rowcount - count)
                 count += 2048
 
-        # TODO: perhaps we want to test with different buffer sizes?
+        # TODO: perhaps we want to test with different buffer sizes?  # noqa: TD002, TD003
         # duckdb_cursor.execute("set streaming_buffer_size='1mb'")
         duckdb_cursor.execute(f"create table tbl as from range({rowcount})")
         duckdb_cursor.execute("select * from tbl")
diff --git a/tests/fast/api/test_read_csv.py b/tests/fast/api/test_read_csv.py
@@ -519,7 +519,7 @@ def test_read_csv_names_mixed_with_dtypes(self, tmp_path):
         assert rel.types == ["BIGINT", "BOOLEAN", "VARCHAR", "BIGINT"]
 
         # dtypes and names dont match
-        # TODO: seems the order columns are named in this error is non-deterministic
+        # TODO: seems the order columns are named in this error is non-deterministic  # noqa: TD002, TD003
         # so for now I'm excluding the list of columns from the expected error
         expected_error = """do not exist in the CSV File"""
         with pytest.raises(duckdb.BinderException, match=expected_error):
diff --git a/tests/fast/arrow/test_arrow_offsets.py b/tests/fast/arrow/test_arrow_offsets.py
@@ -228,7 +228,7 @@ def test_struct_of_blobs(self, duckdb_cursor, col1_null, col2_null):
     def test_struct_of_time(self, duckdb_cursor, constructor, unit, expected, col1_null, col2_null):
         size = MAGIC_ARRAY_SIZE
         if unit == "s":
-            # TODO: We limit the size because we don't support time values > 24 hours
+            # TODO: We limit the size because we don't support time values > 24 hours  # noqa: TD002, TD003
             size = 86400  # The amount of seconds in a day
 
         col1 = [i for i in range(size)]
diff --git a/tests/fast/arrow/test_integration.py b/tests/fast/arrow/test_integration.py
@@ -15,7 +15,7 @@ def test_parquet_roundtrip(self, duckdb_cursor):
         parquet_filename = str(Path(__file__).parent / "data" / "userdata1.parquet")
         cols = "id, first_name, last_name, email, gender, ip_address, cc, country, birthdate, salary, title, comments"
 
-        # TODO timestamp
+        # TODO timestamp  # noqa: TD002, TD003, TD004
 
         userdata_parquet_table = pq.read_table(parquet_filename)
         userdata_parquet_table.validate(full=True)
diff --git a/tests/fast/pandas/test_create_table_from_pandas.py b/tests/fast/pandas/test_create_table_from_pandas.py
@@ -31,7 +31,7 @@ class TestCreateTableFromPandas:
     def test_integer_create_table(self, duckdb_cursor, pandas):
         if sys.version_info.major < 3:
             return
-        # TODO: This should work with other data types e.g., int8...
+        # TODO: This should work with other data types e.g., int8...  # noqa: TD002, TD003
         data_types = ["Int8", "Int16", "Int32", "Int64"]
         internal_data = [1, 2, 3, 4]
         expected_result = [(1,), (2,), (3,), (4,)]
@@ -40,4 +40,4 @@ def test_integer_create_table(self, duckdb_cursor, pandas):
             assert_create_register(internal_data, expected_result, data_type, pandas)
             assert_create(internal_data, expected_result, data_type, pandas)
 
-    # TODO: Also test other data types
+    # TODO: Also test other data types  # noqa: TD002, TD003
diff --git a/tests/fast/pandas/test_df_object_resolution.py b/tests/fast/pandas/test_df_object_resolution.py
@@ -85,7 +85,7 @@ def check_struct_upgrade(expected_type: str, creation_method, pair: ObjectPair,
 
 
 class TestResolveObjectColumns:
-    # TODO: add support for ArrowPandas
+    # TODO: add support for ArrowPandas  # noqa: TD002, TD003
     @pytest.mark.parametrize("pandas", [NumpyPandas()])
     def test_integers(self, pandas, duckdb_cursor):
         data = [5, 0, 3]
diff --git a/tests/fast/pandas/test_pandas_types.py b/tests/fast/pandas/test_pandas_types.py
@@ -64,7 +64,7 @@ def test_pandas_numeric(self):
         out_df = conn.execute("select * from df").df()
 
         # Verify that the types in the out_df are correct
-        # TODO: we don't support outputting pandas specific types (i.e UInt64)
+        # TODO: we don't support outputting pandas specific types (i.e UInt64)  # noqa: TD002, TD003
         for letter, item in zip(string.ascii_lowercase, data_types):
             column_name = letter
             assert str(out_df[column_name].dtype) == item.lower()
diff --git a/tests/fast/pandas/test_timedelta.py b/tests/fast/pandas/test_timedelta.py
@@ -80,7 +80,7 @@ def create_python_interval(
         equality = equality.format(value=duck_interval)
         res, a, b = duckdb_cursor.execute(equality, [val]).fetchone()
         if res != True:
-            # TODO: in some cases intervals that are identical don't compare equal.
+            # TODO: in some cases intervals that are identical don't compare equal.  # noqa: TD002, TD003
             assert a == b
         else:
             assert res == True
diff --git a/tests/fast/relational_api/test_rapi_aggregations.py b/tests/fast/relational_api/test_rapi_aggregations.py
@@ -292,9 +292,9 @@ def test_sum(self, table):
         assert len(result) == len(expected)
         assert all(r == e for r, e in zip(result, expected))
 
-    # TODO: Approximate aggregate functions
+    # TODO: Approximate aggregate functions  # noqa: TD002, TD003
 
-    # TODO: Statistical aggregate functions
+    # TODO: Statistical aggregate functions  # noqa: TD002, TD003
     def test_median(self, table):
         result = table.median("v").execute().fetchall()
         expected = [(2.0,)]
diff --git a/tests/fast/spark/test_spark_catalog.py b/tests/fast/spark/test_spark_catalog.py
@@ -47,7 +47,7 @@ def test_list_columns(self, spark):
             Column(name="b", description=None, dataType="BOOLEAN", nullable=True, isPartition=False, isBucket=False),
         ]
 
-        # TODO: should this error instead?
+        # TODO: should this error instead?  # noqa: TD002, TD003
         non_existant_columns = spark.catalog.listColumns("none_existant")
         assert non_existant_columns == []
 
diff --git a/tests/fast/spark/test_spark_column.py b/tests/fast/spark/test_spark_column.py
@@ -14,7 +14,7 @@ class TestSparkColumn:
     def test_struct_column(self, spark):
         df = spark.createDataFrame([Row(a=1, b=2, c=3, d=4)])
 
-        # TODO: column names should be set explicitly using the Row, rather than letting duckdb assign defaults
+        # TODO: column names should be set explicitly using the Row, rather than letting duckdb assign defaults  # noqa: TD002, TD003
         #  (col0, col1, etc..)
         if USE_ACTUAL_SPARK:
             df = df.withColumn("struct", struct(df.a, df.b))
diff --git a/tests/fast/spark/test_spark_dataframe.py b/tests/fast/spark/test_spark_dataframe.py
@@ -61,7 +61,7 @@ def test_dataframe_from_list_of_tuples(self, spark):
 
         # Empty list
         if not USE_ACTUAL_SPARK:
-            # TODO: Spark raises PySparkValueError [CANNOT_INFER_EMPTY_SCHEMA]
+            # TODO: Spark raises PySparkValueError [CANNOT_INFER_EMPTY_SCHEMA]  # noqa: TD002, TD003
             df = spark.createDataFrame([], ["id", "address", "test"])
             res = df.collect()
             assert res == []
@@ -83,13 +83,13 @@ def test_dataframe_from_list_of_tuples(self, spark):
 
         # Not enough column names
         if not USE_ACTUAL_SPARK:
-            # TODO: Spark does not raise this error
+            # TODO: Spark does not raise this error  # noqa: TD002, TD003
             with pytest.raises(PySparkValueError, match="number of columns in the DataFrame don't match"):
                 df = spark.createDataFrame(address, ["id", "address"])
 
         # Empty column names list
         # Columns are filled in with default names
-        # TODO: check against Spark behavior
+        # TODO: check against Spark behavior  # noqa: TD002, TD003
         df = spark.createDataFrame(address, [])
         res = df.collect()
         assert res == [
@@ -172,7 +172,7 @@ def test_empty_df(self, spark):
         )
         df = spark.createDataFrame([], schema=schema)
         res = df.collect()
-        # TODO: assert that the types and column names are correct
+        # TODO: assert that the types and column names are correct  # noqa: TD002, TD003
         assert res == []
 
     def test_df_from_pandas(self, spark):
diff --git a/tests/fast/spark/test_spark_functions_array.py b/tests/fast/spark/test_spark_functions_array.py
@@ -222,7 +222,7 @@ def test_arrays_zip(self, spark):
         df = spark.createDataFrame([([1, 2, 3], [2, 4, 6], [3, 6])], ["vals1", "vals2", "vals3"])
 
         res = df.select(sf.arrays_zip(df.vals1, df.vals2, df.vals3).alias("zipped")).collect()
-        # TODO: The structure of the results should be the same
+        # TODO: The structure of the results should be the same  # noqa: TD002, TD003
         if USE_ACTUAL_SPARK:
             assert res == [
                 Row(
diff --git a/tests/fast/spark/test_spark_functions_numeric.py b/tests/fast/spark/test_spark_functions_numeric.py
@@ -291,7 +291,7 @@ def test_asin(self, spark):
         if USE_ACTUAL_SPARK:
             assert np.isnan(res[1].asin_value)
         else:
-            # TODO: DuckDB should return NaN here. Reason is that
+            # TODO: DuckDB should return NaN here. Reason is that  # noqa: TD002, TD003
             # ConstantExpression(float("nan")) gives NULL and not NaN
             assert res[1].asin_value is None
 
diff --git a/tests/fast/spark/test_spark_functions_string.py b/tests/fast/spark/test_spark_functions_string.py
@@ -177,7 +177,7 @@ def test_encode(self, spark):
         df = spark.createDataFrame([("abcd",)], ["c"])
 
         res = df.select(F.encode("c", "UTF-8").alias("encoded")).collect()
-        # TODO: Should return the same type
+        # TODO: Should return the same type  # noqa: TD002, TD003
         if USE_ACTUAL_SPARK:
             assert res == [Row(encoded=bytearray(b"abcd"))]
         else:
diff --git a/tests/fast/spark/test_spark_group_by.py b/tests/fast/spark/test_spark_group_by.py
@@ -299,7 +299,7 @@ def test_group_by_product(self, spark):
     def test_group_by_skewness(self, spark):
         df = spark.createDataFrame([[1, "A"], [1, "A"], [2, "A"]], ["c", "group"])
         res = df.groupBy("group").agg(skewness(df.c).alias("v")).collect()
-        # TODO: Why is this different?
+        # TODO: Why is this different?  # noqa: TD002, TD003
         if USE_ACTUAL_SPARK:
             assert pytest.approx(res[0].v) == 0.7071067811865475
         else:
diff --git a/tests/fast/test_insert.py b/tests/fast/test_insert.py
@@ -27,6 +27,6 @@ def test_insert_with_schema(self, duckdb_cursor):
         res = duckdb_cursor.table("not_main.tbl").fetchall()
         assert len(res) == 10
 
-        # TODO: This is not currently supported
+        # TODO: This is not currently supported  # noqa: TD002, TD003
         with pytest.raises(duckdb.CatalogException, match="Table with name tbl does not exist"):
             duckdb_cursor.table("not_main.tbl").insert([42, 21, 1337])
diff --git a/tests/fast/test_replacement_scan.py b/tests/fast/test_replacement_scan.py
@@ -200,7 +200,7 @@ def test_replacement_scan_after_creation(self, duckdb_cursor):
         duckdb_cursor.execute("drop table df")
         df = pd.DataFrame({"b": [1, 2, 3]})
         res = rel.fetchall()
-        # TODO: this should error instead, the 'df' table we relied on has been removed and replaced with a
+        # TODO: this should error instead, the 'df' table we relied on has been removed and replaced with a  # noqa: TD002, TD003
         #  replacement scan
         assert res == [(1,), (2,), (3,)]
 
@@ -263,7 +263,7 @@ def test_cte(self, duckdb_cursor, df_create):
         if df_create == from_arrow:
             # Because the RecordBatchReader is destructive, it's empty after the first scan
             # But we reference it multiple times, so the subsequent reads have no data to read
-            # TODO: this should probably throw an error...
+            # TODO: this should probably throw an error...  # noqa: TD002, TD003
             assert len(res) >= 0
         else:
             assert res == [([1, 2, 3],), ([1, 2, 3],), ([1, 2, 3],)]
@@ -351,7 +351,7 @@ def create_view_in_func(con) -> None:
 
         create_view_in_func(duckdb_cursor)
 
-        # TODO: this should be fixed in the future, likely by unifying the behavior of .sql and .execute
+        # TODO: this should be fixed in the future, likely by unifying the behavior of .sql and .execute  # noqa: TD002, TD003
         with pytest.raises(duckdb.CatalogException, match="Table with name df does not exist"):
             rel = duckdb_cursor.sql("select * from v1")
 
diff --git a/tests/fast/test_type.py b/tests/fast/test_type.py
@@ -40,7 +40,7 @@
 class TestType:
     def test_sqltype(self):
         assert str(duckdb.sqltype("struct(a VARCHAR, b BIGINT)")) == "STRUCT(a VARCHAR, b BIGINT)"
-        # TODO: add tests with invalid type_str
+        # TODO: add tests with invalid type_str  # noqa: TD002, TD003
 
     def test_primitive_types(self):
         assert str(SQLNULL) == '"NULL"'
@@ -83,7 +83,7 @@ def test_struct_type(self):
         type = duckdb.struct_type({"a": BIGINT, "b": BOOLEAN})
         assert str(type) == "STRUCT(a BIGINT, b BOOLEAN)"
 
-        # TODO: create an unnamed struct when fields are provided as a list
+        # TODO: create an unnamed struct when fields are provided as a list  # noqa: TD002, TD003
         type = duckdb.struct_type([BIGINT, BOOLEAN])
         assert str(type) == "STRUCT(v1 BIGINT, v2 BOOLEAN)"
 
diff --git a/tests/fast/udf/test_scalar_native.py b/tests/fast/udf/test_scalar_native.py
@@ -198,7 +198,7 @@ def add_extra_column(original):
         )
         # added extra column to the struct
         assert len(res.fetchone()[0].keys()) == 3
-        # TODO: this is needed, otherwise the old transaction is still active when we try to start a new transaction
+        # TODO: this is needed, otherwise the old transaction is still active when we try to start a new transaction  # noqa: TD002, TD003
         #  inside of 'create_function', which means the call would fail
         res.fetchall()
 
