TST/CLN: Remove makeTime methods (#56264)

* Remove makeTimeDataFrame

* Remove makeTimeSeriesData

* Remove getCols

* Remove use makeCustomIndex usage

* Fix tests

Author: mroeschke

pandas/_testing/__init__.py

@@ -6,7 +6,6 @@
 import operator
 import os
 import re
-import string
 from sys import byteorder
 from typing import (
     TYPE_CHECKING,
@@ -109,7 +108,6 @@
     from pandas.core.arrays import ArrowExtensionArray
 
 _N = 30
-_K = 4
 
 UNSIGNED_INT_NUMPY_DTYPES: list[NpDtype] = ["uint8", "uint16", "uint32", "uint64"]
 UNSIGNED_INT_EA_DTYPES: list[Dtype] = ["UInt8", "UInt16", "UInt32", "UInt64"]
@@ -341,10 +339,6 @@ def to_array(obj):
 # Others
 
 
-def getCols(k) -> str:
-    return string.ascii_uppercase[:k]
-
-
 def makeTimeSeries(nper=None, freq: Frequency = "B", name=None) -> Series:
     if nper is None:
         nper = _N
@@ -355,16 +349,6 @@ def makeTimeSeries(nper=None, freq: Frequency = "B", name=None) -> Series:
     )
 
 
-def getTimeSeriesData(nper=None, freq: Frequency = "B") -> dict[str, Series]:
-    return {c: makeTimeSeries(nper, freq) for c in getCols(_K)}
-
-
-# make frame
-def makeTimeDataFrame(nper=None, freq: Frequency = "B") -> DataFrame:
-    data = getTimeSeriesData(nper, freq)
-    return DataFrame(data)
-
-
 def makeCustomIndex(
     nentries,
     nlevels,
@@ -887,21 +871,18 @@ def shares_memory(left, right) -> bool:
     "external_error_raised",
     "FLOAT_EA_DTYPES",
     "FLOAT_NUMPY_DTYPES",
-    "getCols",
     "get_cython_table_params",
     "get_dtype",
     "getitem",
     "get_locales",
     "get_finest_unit",
     "get_obj",
     "get_op_from_name",
-    "getTimeSeriesData",
     "iat",
     "iloc",
     "loc",
     "makeCustomDataframe",
     "makeCustomIndex",
-    "makeTimeDataFrame",
     "makeTimeSeries",
     "maybe_produces_warning",
     "NARROW_NP_DTYPES",

pandas/conftest.py

@@ -550,7 +550,11 @@ def multiindex_year_month_day_dataframe_random_data():
     DataFrame with 3 level MultiIndex (year, month, day) covering
     first 100 business days from 2000-01-01 with random data
     """
-    tdf = tm.makeTimeDataFrame(100)
+    tdf = DataFrame(
+        np.random.default_rng(2).standard_normal((100, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=100, freq="B"),
+    )
     ymd = tdf.groupby([lambda x: x.year, lambda x: x.month, lambda x: x.day]).sum()
     # use int64 Index, to make sure things work
     ymd.index = ymd.index.set_levels([lev.astype("i8") for lev in ymd.index.levels])

pandas/tests/frame/conftest.py

@@ -7,7 +7,6 @@
     NaT,
     date_range,
 )
-import pandas._testing as tm
 
 
 @pytest.fixture
@@ -16,27 +15,12 @@ def datetime_frame() -> DataFrame:
     Fixture for DataFrame of floats with DatetimeIndex
 
     Columns are ['A', 'B', 'C', 'D']
-
-                       A         B         C         D
-    2000-01-03 -1.122153  0.468535  0.122226  1.693711
-    2000-01-04  0.189378  0.486100  0.007864 -1.216052
-    2000-01-05  0.041401 -0.835752 -0.035279 -0.414357
-    2000-01-06  0.430050  0.894352  0.090719  0.036939
-    2000-01-07 -0.620982 -0.668211 -0.706153  1.466335
-    2000-01-10 -0.752633  0.328434 -0.815325  0.699674
-    2000-01-11 -2.236969  0.615737 -0.829076 -1.196106
-    ...              ...       ...       ...       ...
-    2000-02-03  1.642618 -0.579288  0.046005  1.385249
-    2000-02-04 -0.544873 -1.160962 -0.284071 -1.418351
-    2000-02-07 -2.656149 -0.601387  1.410148  0.444150
-    2000-02-08 -1.201881 -1.289040  0.772992 -1.445300
-    2000-02-09  1.377373  0.398619  1.008453 -0.928207
-    2000-02-10  0.473194 -0.636677  0.984058  0.511519
-    2000-02-11 -0.965556  0.408313 -1.312844 -0.381948
-
-    [30 rows x 4 columns]
     """
-    return DataFrame(tm.getTimeSeriesData())
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((100, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=100, freq="B"),
+    )
 
 
 @pytest.fixture

pandas/tests/frame/indexing/test_indexing.py

@@ -584,7 +584,7 @@ def test_fancy_getitem_slice_mixed(
             tm.assert_frame_equal(float_frame, original)
 
     def test_getitem_setitem_non_ix_labels(self):
-        df = tm.makeTimeDataFrame()
+        df = DataFrame(range(20), index=date_range("2020-01-01", periods=20))
 
         start, end = df.index[[5, 10]]
 

pandas/tests/frame/methods/test_cov_corr.py

@@ -6,7 +6,9 @@
 import pandas as pd
 from pandas import (
     DataFrame,
+    Index,
     Series,
+    date_range,
     isna,
 )
 import pandas._testing as tm
@@ -325,8 +327,12 @@ def test_corrwith(self, datetime_frame, dtype):
             tm.assert_almost_equal(correls[row], df1.loc[row].corr(df2.loc[row]))
 
     def test_corrwith_with_objects(self):
-        df1 = tm.makeTimeDataFrame()
-        df2 = tm.makeTimeDataFrame()
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
+        df2 = df1.copy()
         cols = ["A", "B", "C", "D"]
 
         df1["obj"] = "foo"

pandas/tests/frame/methods/test_first_and_last.py

@@ -1,12 +1,15 @@
 """
 Note: includes tests for `last`
 """
+import numpy as np
 import pytest
 
 import pandas as pd
 from pandas import (
     DataFrame,
+    Index,
     bdate_range,
+    date_range,
 )
 import pandas._testing as tm
 
@@ -16,13 +19,21 @@
 
 class TestFirst:
     def test_first_subset(self, frame_or_series):
-        ts = tm.makeTimeDataFrame(freq="12h")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=100, freq="12h"),
+        )
         ts = tm.get_obj(ts, frame_or_series)
         with tm.assert_produces_warning(FutureWarning, match=deprecated_msg):
             result = ts.first("10d")
             assert len(result) == 20
 
-        ts = tm.makeTimeDataFrame(freq="D")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=100, freq="D"),
+        )
         ts = tm.get_obj(ts, frame_or_series)
         with tm.assert_produces_warning(FutureWarning, match=deprecated_msg):
             result = ts.first("10d")
@@ -64,13 +75,21 @@ def test_first_last_raises(self, frame_or_series):
             obj.last("1D")
 
     def test_last_subset(self, frame_or_series):
-        ts = tm.makeTimeDataFrame(freq="12h")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((100, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=100, freq="12h"),
+        )
         ts = tm.get_obj(ts, frame_or_series)
         with tm.assert_produces_warning(FutureWarning, match=last_deprecated_msg):
             result = ts.last("10d")
         assert len(result) == 20
 
-        ts = tm.makeTimeDataFrame(nper=30, freq="D")
+        ts = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=30, freq="D"),
+        )
         ts = tm.get_obj(ts, frame_or_series)
         with tm.assert_produces_warning(FutureWarning, match=last_deprecated_msg):
             result = ts.last("10d")

pandas/tests/frame/methods/test_truncate.py

@@ -60,7 +60,7 @@ def test_truncate(self, datetime_frame, frame_or_series):
         truncated = ts.truncate(before=ts.index[-1] + ts.index.freq)
         assert len(truncated) == 0
 
-        msg = "Truncate: 2000-01-06 00:00:00 must be after 2000-02-04 00:00:00"
+        msg = "Truncate: 2000-01-06 00:00:00 must be after 2000-05-16 00:00:00"
         with pytest.raises(ValueError, match=msg):
             ts.truncate(
                 before=ts.index[-1] - ts.index.freq, after=ts.index[0] + ts.index.freq

pandas/tests/frame/test_arithmetic.py

@@ -1523,8 +1523,12 @@ def test_combineFunc(self, float_frame, mixed_float_frame):
         [operator.eq, operator.ne, operator.lt, operator.gt, operator.ge, operator.le],
     )
     def test_comparisons(self, simple_frame, float_frame, func):
-        df1 = tm.makeTimeDataFrame()
-        df2 = tm.makeTimeDataFrame()
+        df1 = DataFrame(
+            np.random.default_rng(2).standard_normal((30, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=pd.date_range("2000-01-01", periods=30, freq="B"),
+        )
+        df2 = df1.copy()
 
         row = simple_frame.xs("a")
         ndim_5 = np.ones(df1.shape + (1, 1, 1))

pandas/tests/generic/test_generic.py

@@ -10,7 +10,9 @@
 
 from pandas import (
     DataFrame,
+    Index,
     Series,
+    date_range,
 )
 import pandas._testing as tm
 
@@ -328,12 +330,16 @@ def test_squeeze_series_noop(self, ser):
 
     def test_squeeze_frame_noop(self):
         # noop
-        df = tm.makeTimeDataFrame()
+        df = DataFrame(np.eye(2))
         tm.assert_frame_equal(df.squeeze(), df)
 
     def test_squeeze_frame_reindex(self):
         # squeezing
-        df = tm.makeTimeDataFrame().reindex(columns=["A"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        ).reindex(columns=["A"])
         tm.assert_series_equal(df.squeeze(), df["A"])
 
     def test_squeeze_0_len_dim(self):
@@ -345,7 +351,11 @@ def test_squeeze_0_len_dim(self):
 
     def test_squeeze_axis(self):
         # axis argument
-        df = tm.makeTimeDataFrame(nper=1).iloc[:, :1]
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((1, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=1, freq="B"),
+        ).iloc[:, :1]
         assert df.shape == (1, 1)
         tm.assert_series_equal(df.squeeze(axis=0), df.iloc[0])
         tm.assert_series_equal(df.squeeze(axis="index"), df.iloc[0])
@@ -360,14 +370,22 @@ def test_squeeze_axis(self):
             df.squeeze(axis="x")
 
     def test_squeeze_axis_len_3(self):
-        df = tm.makeTimeDataFrame(3)
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((3, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=3, freq="B"),
+        )
         tm.assert_frame_equal(df.squeeze(axis=0), df)
 
     def test_numpy_squeeze(self):
         s = Series(range(2), dtype=np.float64)
         tm.assert_series_equal(np.squeeze(s), s)
 
-        df = tm.makeTimeDataFrame().reindex(columns=["A"])
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        ).reindex(columns=["A"])
         tm.assert_series_equal(np.squeeze(df), df["A"])
 
     @pytest.mark.parametrize(
@@ -382,11 +400,19 @@ def test_transpose_series(self, ser):
         tm.assert_series_equal(ser.transpose(), ser)
 
     def test_transpose_frame(self):
-        df = tm.makeTimeDataFrame()
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
         tm.assert_frame_equal(df.transpose().transpose(), df)
 
     def test_numpy_transpose(self, frame_or_series):
-        obj = tm.makeTimeDataFrame()
+        obj = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
         obj = tm.get_obj(obj, frame_or_series)
 
         if frame_or_series is Series:
@@ -419,7 +445,11 @@ def test_take_series(self, ser):
 
     def test_take_frame(self):
         indices = [1, 5, -2, 6, 3, -1]
-        df = tm.makeTimeDataFrame()
+        df = DataFrame(
+            np.random.default_rng(2).standard_normal((10, 4)),
+            columns=Index(list("ABCD"), dtype=object),
+            index=date_range("2000-01-01", periods=10, freq="B"),
+        )
         out = df.take(indices)
         expected = DataFrame(
             data=df.values.take(indices, axis=0),
@@ -431,7 +461,7 @@ def test_take_frame(self):
     def test_take_invalid_kwargs(self, frame_or_series):
         indices = [-3, 2, 0, 1]
 
-        obj = tm.makeTimeDataFrame()
+        obj = DataFrame(range(5))
         obj = tm.get_obj(obj, frame_or_series)
 
         msg = r"take\(\) got an unexpected keyword argument 'foo'"

pandas/tests/groupby/aggregate/test_aggregate.py

@@ -161,7 +161,11 @@ def test_agg_apply_corner(ts, tsframe):
 
 
 def test_agg_grouping_is_list_tuple(ts):
-    df = tm.makeTimeDataFrame()
+    df = DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=pd.date_range("2000-01-01", periods=30, freq="B"),
+    )
 
     grouped = df.groupby(lambda x: x.year)
     grouper = grouped.grouper.groupings[0].grouping_vector

pandas/tests/groupby/conftest.py

@@ -1,7 +1,11 @@
 import numpy as np
 import pytest
 
-from pandas import DataFrame
+from pandas import (
+    DataFrame,
+    Index,
+    date_range,
+)
 import pandas._testing as tm
 from pandas.core.groupby.base import (
     reduction_kernels,
@@ -48,7 +52,11 @@ def ts():
 
 @pytest.fixture
 def tsframe():
-    return DataFrame(tm.getTimeSeriesData())
+    return DataFrame(
+        np.random.default_rng(2).standard_normal((30, 4)),
+        columns=Index(list("ABCD"), dtype=object),
+        index=date_range("2000-01-01", periods=30, freq="B"),
+    )
 
 
 @pytest.fixture