Apply ruff code formatter

Author: christophscheuch

tests/test_add_lagged_columns.py

@@ -18,9 +18,7 @@ def test_exact_lag_with_by_returns_correct_lagged_values():
     data = pd.DataFrame(
         {
             "permno": [1] * 4 + [2] * 4,
-            "date": list(
-                pd.date_range("2023-01-01", periods=4, freq="MS")
-            ) * 2,
+            "date": list(pd.date_range("2023-01-01", periods=4, freq="MS")) * 2,
             "size": [float(i) for i in range(1, 9)],
         }
     )
@@ -41,9 +39,7 @@ def test_exact_lag_without_by_returns_correct_values():
             "size": [1.0, 2.0, 3.0],
         }
     )
-    result = add_lagged_columns(
-        data, cols="size", lag=pd.DateOffset(months=1)
-    )
+    result = add_lagged_columns(data, cols="size", lag=pd.DateOffset(months=1))
     assert pd.isna(result["size_lag"].iloc[0])
     assert result["size_lag"].iloc[1] == 1
     assert result["size_lag"].iloc[2] == 2
@@ -53,9 +49,7 @@ def test_window_lag_handles_all_src_date_conditions():
     """Test window lag: NA, in-window, and below-lower-bound cases."""
     data = pd.DataFrame(
         {
-            "date": pd.to_datetime(
-                ["2023-01-01", "2023-02-01", "2023-06-01"]
-            ),
+            "date": pd.to_datetime(["2023-01-01", "2023-02-01", "2023-06-01"]),
             "size": [1.0, 2.0, 3.0],
         }
     )
@@ -104,9 +98,7 @@ def test_ff_adjustment_without_by_uses_year_grouping_only():
     """Test ff_adjustment without by uses year grouping only."""
     data = pd.DataFrame(
         {
-            "date": pd.to_datetime(
-                ["2022-06-01", "2022-12-01", "2023-06-01"]
-            ),
+            "date": pd.to_datetime(["2022-06-01", "2022-12-01", "2023-06-01"]),
             "size": [10.0, 20.0, 30.0],
         }
     )
@@ -152,18 +144,14 @@ def test_error_when_date_column_is_absent_from_data():
 
 def test_error_when_lag_is_negative():
     """Test error when lag is negative."""
-    data = pd.DataFrame(
-        {"date": [pd.Timestamp("2023-01-01")], "size": [1.0]}
-    )
+    data = pd.DataFrame({"date": [pd.Timestamp("2023-01-01")], "size": [1.0]})
     with pytest.raises(ValueError, match="non-negative"):
         add_lagged_columns(data, cols="size", lag=-1)
 
 
 def test_error_when_max_lag_is_less_than_lag():
     """Test error when max_lag is less than lag."""
-    data = pd.DataFrame(
-        {"date": [pd.Timestamp("2023-01-01")], "size": [1.0]}
-    )
+    data = pd.DataFrame({"date": [pd.Timestamp("2023-01-01")], "size": [1.0]})
     with pytest.raises(ValueError, match="max_lag"):
         add_lagged_columns(
             data,
@@ -175,9 +163,7 @@ def test_error_when_max_lag_is_less_than_lag():
 
 def test_error_when_requested_column_is_absent_from_data():
     """Test error when a requested column is absent from data."""
-    data = pd.DataFrame(
-        {"date": [pd.Timestamp("2023-01-01")], "size": [1.0]}
-    )
+    data = pd.DataFrame({"date": [pd.Timestamp("2023-01-01")], "size": [1.0]})
     with pytest.raises(ValueError, match="missing"):
         add_lagged_columns(
             data, cols="no_such_col", lag=pd.DateOffset(months=1)
@@ -186,9 +172,7 @@ def test_error_when_requested_column_is_absent_from_data():
 
 def test_error_when_by_column_is_absent_from_data():
     """Test error when a by column is absent from data."""
-    data = pd.DataFrame(
-        {"date": [pd.Timestamp("2023-01-01")], "size": [1.0]}
-    )
+    data = pd.DataFrame({"date": [pd.Timestamp("2023-01-01")], "size": [1.0]})
     with pytest.raises(ValueError, match="missing"):
         add_lagged_columns(
             data,
@@ -207,9 +191,7 @@ def test_error_when_join_key_is_not_unique():
         }
     )
     with pytest.raises(ValueError, match="unique"):
-        add_lagged_columns(
-            data, cols="size", lag=pd.DateOffset(months=1)
-        )
+        add_lagged_columns(data, cols="size", lag=pd.DateOffset(months=1))
 
 
 def test_error_when_upper_helper_column_already_exists():
@@ -233,6 +215,7 @@ def test_error_when_upper_helper_column_already_exists():
 def test_data_options_specifies_date_column_name():
     """Test data_options dict specifies the date column name."""
     from tidyfinance.core import data_options
+
     data = pd.DataFrame(
         {
             "my_date": pd.date_range("2023-01-01", periods=3, freq="MS"),
@@ -249,6 +232,7 @@ def test_data_options_specifies_date_column_name():
     assert pd.isna(result["size_lag"].iloc[0])
     assert result["size_lag"].iloc[1] == 1
 
+
 if __name__ == "__main__":
     # Run all tests
     pytest.main([__file__])

tests/test_assign_portfolio.py

@@ -14,7 +14,6 @@
 
 from tidyfinance.core import assign_portfolio  # noqa: E402
 
-
 # %% mock breakpoint functions
 
 
@@ -31,9 +30,7 @@ def mock_breakpoint_percentiles(
     data, sorting_variable, bp_options=None, data_options=None
 ):
     """Percentile-based mock breakpoint function."""
-    percentiles = (bp_options or {}).get(
-        "percentiles", [0.2, 0.4, 0.6, 0.8]
-    )
+    percentiles = (bp_options or {}).get("percentiles", [0.2, 0.4, 0.6, 0.8])
     probs = [0] + list(percentiles) + [1]
     return np.quantile(data[sorting_variable].dropna().values, probs)
 
@@ -156,9 +153,7 @@ def test_constant_sorting_variable_returns_vector_of_correct_length():
 
 def test_two_distinct_values_with_2_portfolios_produces_two_groups():
     """Test two distinct values with 2 portfolios produces two groups."""
-    data = pd.DataFrame(
-        {"id": range(100), "value": [1] * 50 + [2] * 50}
-    )
+    data = pd.DataFrame({"id": range(100), "value": [1] * 50 + [2] * 50})
     result = assign_portfolio(
         data,
         "value",
@@ -189,9 +184,7 @@ def test_single_row_data_frame_with_constant_variable_triggers_warning():
 
 def test_warning_when_clusters_reduce_number_of_portfolios():
     """Test warning is issued when clusters reduce the number of portfolios."""
-    data = pd.DataFrame(
-        {"id": range(100), "value": [1] * 50 + [100] * 50}
-    )
+    data = pd.DataFrame({"id": range(100), "value": [1] * 50 + [100] * 50})
 
     def mock_bp_5(data, sv, bp_options=None, data_options=None):
         return np.array([1, 20, 40, 60, 80, 100])
@@ -335,9 +328,7 @@ def test_function_works_with_large_datasets():
     """Test function works with large datasets."""
     rng = np.random.default_rng(1)
     n = 100_000
-    data = pd.DataFrame(
-        {"id": np.arange(n), "value": rng.standard_normal(n)}
-    )
+    data = pd.DataFrame({"id": np.arange(n), "value": rng.standard_normal(n)})
     result = assign_portfolio(
         data,
         "value",
@@ -416,9 +407,7 @@ def test_percentile_based_breakpoints_produce_correct_number_of_groups():
 def test_two_portfolios_split_data_roughly_in_half():
     """Test two portfolios split data roughly in half."""
     rng = np.random.default_rng(10)
-    data = pd.DataFrame(
-        {"id": range(1000), "value": rng.standard_normal(1000)}
-    )
+    data = pd.DataFrame({"id": range(1000), "value": rng.standard_normal(1000)})
     result = assign_portfolio(
         data,
         "value",
@@ -444,9 +433,7 @@ def spy_bp(data, sorting_variable, bp_options=None, data_options=None):
         received["data_options"] = data_options
         return np.array([0.0, 0.5, 1.0])
 
-    data = pd.DataFrame(
-        {"id": range(10), "value": np.linspace(0, 1, 10)}
-    )
+    data = pd.DataFrame({"id": range(10), "value": np.linspace(0, 1, 10)})
     my_bp_opts = {"n_portfolios": 2}
     my_data_opts = {"date": "date_col"}
     assign_portfolio(

tests/test_backend.py

@@ -35,6 +35,7 @@ def _restore_backend():
 
 # %% set_backend / get_backend
 
+
 def test_default_backend_is_pandas():
     assert get_backend() == "pandas"
 
@@ -51,6 +52,7 @@ def test_set_backend_rejects_invalid_value():
 
 # %% _convert_output
 
+
 def test_convert_output_passthrough_for_pandas_backend():
     df = pd.DataFrame({"a": [1, 2]})
     assert _convert_output(df) is df
@@ -92,6 +94,7 @@ def test_convert_output_leaves_dict_alone():
 
 # %% _to_pandas_input
 
+
 def test_to_pandas_input_converts_polars_frame():
     out = _to_pandas_input(pl.DataFrame({"a": [1, 2]}))
     assert isinstance(out, pd.DataFrame)
@@ -109,6 +112,7 @@ def test_to_pandas_input_passes_through_pandas():
 
 # %% download_data integration (network-free pseudo domain)
 
+
 def test_download_data_returns_pandas_by_default():
     with pytest.warns(UserWarning, match="pseudo data"):
         out = tf.download_data("Pseudo Data", "crsp_monthly")
@@ -124,12 +128,12 @@ def test_download_data_returns_polars_when_configured():
 
 # %% core function honors the backend (input + output round-trip)
 
+
 def _lag_input():
     return pd.DataFrame(
         {
             "permno": [1] * 4 + [2] * 4,
-            "date": list(pd.date_range("2023-01-01", periods=4, freq="MS"))
-            * 2,
+            "date": list(pd.date_range("2023-01-01", periods=4, freq="MS")) * 2,
             "size": [float(i) for i in range(1, 9)],
         }
     )
@@ -174,6 +178,7 @@ def test_series_returning_function_stays_pandas_under_polars():
 
 # %% boundary-only wrapping protects internal cross-calls
 
+
 def test_in_module_implementations_remain_unwrapped():
     """The wrapping is applied only at the public package boundary. The
     in-module implementations (which core functions call internally,
@@ -193,6 +198,7 @@ def test_in_module_implementations_remain_unwrapped():
 # argument through the polars backend without raising and produce the
 # expected output type.
 
+
 def _panel_with_returns():
     """Five-asset, two-year panel — enough per cross-section to fit
     three breakpoints without ties."""
@@ -353,13 +359,9 @@ def test_process_trace_data_round_trips_polars():
             "cusip_id": ["00077D1AA"] * 2,
             "msg_seq_nb": [1, 2],
             "orig_msg_seq_nb": [1, 2],
-            "trd_rpt_dt": pd.to_datetime(
-                ["2015-01-05", "2015-01-06"]
-            ),
+            "trd_rpt_dt": pd.to_datetime(["2015-01-05", "2015-01-06"]),
             "trd_rpt_tm": ["09:31:00", "09:36:00"],
-            "trd_exctn_dt": pd.to_datetime(
-                ["2015-01-04", "2015-01-05"]
-            ),
+            "trd_exctn_dt": pd.to_datetime(["2015-01-04", "2015-01-05"]),
             "trd_exctn_tm": ["09:30:00", "09:35:00"],
             "rptd_pr": [100.0, 100.5],
             "entrd_vol_qt": [1000, 1500],
@@ -370,9 +372,7 @@ def test_process_trace_data_round_trips_polars():
             "asof_cd": [None, None],
             "wis_fl": ["N", "N"],
             "days_to_sttl_ct": [2, 2],
-            "stlmnt_dt": pd.to_datetime(
-                ["2015-01-06", "2015-01-07"]
-            ),
+            "stlmnt_dt": pd.to_datetime(["2015-01-06", "2015-01-07"]),
             "spcl_trd_fl": [None, None],
         }
     )

tests/test_breakpoint_options.py

@@ -2,6 +2,7 @@
 
 import os
 import sys
+
 import pytest
 
 sys.path.insert(