WIP/TST: troubleshoot xfails

pandas/tests/indexes/datetimes/test_indexing.py

@@ -651,10 +651,6 @@ def test_get_indexer_mixed_dtypes(self, target):
             ([date(9999, 1, 1), date(9999, 1, 1)], [-1, -1]),
         ],
     )
-    # FIXME: these warnings are flaky GH#36131
-    @pytest.mark.filterwarnings(
-        "ignore:Comparison of Timestamp with datetime.date:FutureWarning"
-    )
     def test_get_indexer_out_of_bounds_date(self, target, positions):
         values = DatetimeIndex([Timestamp("2020-01-01"), Timestamp("2020-01-02")])
 

pandas/tests/io/json/test_pandas.py

@@ -1741,11 +1741,6 @@ def test_json_multiindex(self, dataframe, expected):
         result = series.to_json(orient="index")
         assert result == expected
 
-    @pytest.mark.xfail(
-        is_platform_windows(),
-        reason="localhost connection rejected",
-        strict=False,
-    )
     def test_to_s3(self, s3_resource, s3so):
         import time
 

pandas/tests/io/parser/common/test_chunksize.py

@@ -162,7 +162,6 @@ def test_chunk_begins_with_newline_whitespace(all_parsers):
 
 
 @pytest.mark.slow
-@pytest.mark.xfail(reason="GH38630, sometimes gives ResourceWarning", strict=False)
 def test_chunks_have_consistent_numerical_type(all_parsers):
     parser = all_parsers
     integers = [str(i) for i in range(499999)]
@@ -176,7 +175,7 @@ def test_chunks_have_consistent_numerical_type(all_parsers):
     assert result.a.dtype == float
 
 
-def test_warn_if_chunks_have_mismatched_type(all_parsers, request):
+def test_warn_if_chunks_have_mismatched_type(all_parsers):
     warning_type = None
     parser = all_parsers
     size = 10000
@@ -193,24 +192,8 @@ def test_warn_if_chunks_have_mismatched_type(all_parsers, request):
 
     buf = StringIO(data)
 
-    try:
-        with tm.assert_produces_warning(warning_type):
-            df = parser.read_csv(buf)
-    except AssertionError as err:
-        # 2021-02-21 this occasionally fails on the CI with an unexpected
-        #  ResourceWarning that we have been unable to track down,
-        #  see GH#38630
-        if "ResourceWarning" not in str(err) or parser.engine != "python":
-            raise
-
-        # Check the main assertion of the test before re-raising
-        assert df.a.dtype == object
-
-        mark = pytest.mark.xfail(
-            reason="ResourceWarning for unclosed SSL Socket, GH#38630"
-        )
-        request.node.add_marker(mark)
-        raise
+    with tm.assert_produces_warning(warning_type):
+        df = parser.read_csv(buf)
 
     assert df.a.dtype == object
 

pandas/tests/io/parser/test_compression.py

@@ -89,7 +89,7 @@ def test_zip_error_invalid_zip(parser_and_data):
 
 @skip_pyarrow
 @pytest.mark.parametrize("filename", [None, "test.{ext}"])
-def test_compression(parser_and_data, compression_only, buffer, filename):
+def test_compression(parser_and_data, compression_only, buffer, filename, request):
     parser, data, expected = parser_and_data
     compress_type = compression_only
 

pandas/tests/io/parser/test_network.py

@@ -212,7 +212,6 @@ def test_read_s3_fails(self, s3so):
         with pytest.raises(OSError, match=msg):
             read_csv("s3://cant_get_it/file.csv")
 
-    @pytest.mark.xfail(reason="GH#39155 s3fs upgrade", strict=False)
     def test_write_s3_csv_fails(self, tips_df, s3so):
         # GH 32486
         # Attempting to write to an invalid S3 path should raise
@@ -228,7 +227,6 @@ def test_write_s3_csv_fails(self, tips_df, s3so):
                 "s3://an_s3_bucket_data_doesnt_exit/not_real.csv", storage_options=s3so
             )
 
-    @pytest.mark.xfail(reason="GH#39155 s3fs upgrade", strict=False)
     @td.skip_if_no("pyarrow")
     def test_write_s3_parquet_fails(self, tips_df, s3so):
         # GH 27679

pandas/tests/io/pytables/test_append.py

@@ -581,8 +581,7 @@ def check_col(key, name, size):
             & (df_new.A > 0)
             & (df_new.B < 0)
         ]
-        tm.assert_frame_equal(result, expected, check_freq=False)
-        # FIXME: 2020-05-07 freq check randomly fails in the CI
+        tm.assert_frame_equal(result, expected)
 
         # yield an empty frame
         result = store.select("df", "string='foo' and string2='cool'")
@@ -611,9 +610,7 @@ def check_col(key, name, size):
 
         result = store.select("df_dc", ["B > 0", "C > 0", "string == foo"])
         expected = df_dc[(df_dc.B > 0) & (df_dc.C > 0) & (df_dc.string == "foo")]
-        tm.assert_frame_equal(result, expected, check_freq=False)
-        # FIXME: 2020-12-07 intermittent build failures here with freq of
-        #  None instead of BDay(4)
+        tm.assert_frame_equal(result, expected)
 
     with ensure_clean_store(setup_path) as store:
         # doc example part 2
@@ -896,9 +893,6 @@ def test_append_to_multiple_dropna(setup_path):
         tm.assert_index_equal(store.select("df1").index, store.select("df2").index)
 
 
-@pytest.mark.xfail(
-    run=False, reason="append_to_multiple_dropna_false is not raising as failed"
-)
 def test_append_to_multiple_dropna_false(setup_path):
     df1 = tm.makeTimeDataFrame()
     df2 = tm.makeTimeDataFrame().rename(columns="{}_2".format)
@@ -912,8 +906,7 @@ def test_append_to_multiple_dropna_false(setup_path):
             {"df1a": ["A", "B"], "df2a": None}, df, selector="df1a", dropna=False
         )
 
-        # TODO Update error message to desired message for this case
-        msg = "Cannot select as multiple after appending with dropna=False"
+        msg = "all tables must have exactly the same nrows!"
         with pytest.raises(ValueError, match=msg):
             store.select_as_multiple(["df1a", "df2a"])
 

pandas/tests/io/pytables/test_select.py

@@ -838,8 +838,7 @@ def test_select_as_multiple(setup_path):
         )
         expected = concat([df1, df2], axis=1)
         expected = expected[(expected.A > 0) & (expected.B > 0)]
-        tm.assert_frame_equal(result, expected, check_freq=False)
-        # FIXME: 2021-01-20 this is failing with freq None vs 4B on some builds
+        tm.assert_frame_equal(result, expected)
 
         # multiple (diff selector)
         result = store.select_as_multiple(

pandas/tests/io/pytables/test_store.py

@@ -646,9 +646,7 @@ def test_coordinates(setup_path):
 
         expected = concat([df1, df2], axis=1)
         expected = expected[(expected.A > 0) & (expected.B > 0)]
-        tm.assert_frame_equal(result, expected, check_freq=False)
-        # FIXME: 2021-01-18 on some (mostly windows) builds we get freq=None
-        #  but expect freq="18B"
+        tm.assert_frame_equal(result, expected)
 
     # pass array/mask as the coordinates
     with ensure_clean_store(setup_path) as store:

pandas/tests/io/test_parquet.py

@@ -13,7 +13,6 @@
 
 from pandas._config import get_option
 
-from pandas.compat import is_platform_windows
 from pandas.compat.pyarrow import (
     pa_version_under1p0,
     pa_version_under2p0,
@@ -735,11 +734,6 @@ def test_categorical(self, pa):
 
         check_round_trip(df, pa)
 
-    @pytest.mark.xfail(
-        is_platform_windows(),
-        reason="localhost connection rejected",
-        strict=False,
-    )
     def test_s3_roundtrip_explicit_fs(self, df_compat, s3_resource, pa, s3so):
         s3fs = pytest.importorskip("s3fs")
         s3 = s3fs.S3FileSystem(**s3so)

pandas/tests/plotting/frame/test_frame.py

@@ -494,11 +494,6 @@ def test_line_lim(self):
             assert xmin <= lines[0].get_data()[0][0]
             assert xmax >= lines[0].get_data()[0][-1]
 
-    @pytest.mark.xfail(
-        strict=False,
-        reason="2020-12-01 this has been failing periodically on the "
-        "ymin==0 assertion for a week or so.",
-    )
     @pytest.mark.parametrize("stacked", [True, False])
     def test_area_lim(self, stacked):
         df = DataFrame(np.random.rand(6, 4), columns=["x", "y", "z", "four"])

pandas/tests/plotting/test_converter.py

@@ -10,10 +10,7 @@
 
 import pandas._config.config as cf
 
-from pandas.compat import (
-    is_platform_windows,
-    np_datetime64_compat,
-)
+from pandas.compat import np_datetime64_compat
 import pandas.util._test_decorators as td
 
 from pandas import (
@@ -92,11 +89,6 @@ def test_registering_no_warning(self):
         ax.plot(s.index, s.values)
         plt.close()
 
-    @pytest.mark.xfail(
-        is_platform_windows(),
-        reason="Getting two warnings intermittently, see GH#37746",
-        strict=False,
-    )
     def test_pandas_plots_register(self):
         plt = pytest.importorskip("matplotlib.pyplot")
         s = Series(range(12), index=date_range("2017", periods=12))

pandas/tests/plotting/test_groupby.py

@@ -4,7 +4,6 @@
 import numpy as np
 import pytest
 
-from pandas.compat import is_platform_windows
 import pandas.util._test_decorators as td
 
 from pandas import (
@@ -20,11 +19,6 @@
 
 @td.skip_if_no_mpl
 class TestDataFrameGroupByPlots(TestPlotBase):
-    @pytest.mark.xfail(
-        is_platform_windows(),
-        reason="Looks like LinePlot._is_ts_plot is wrong",
-        strict=False,
-    )
     def test_series_groupby_plotting_nominally_works(self):
         n = 10
         weight = Series(np.random.normal(166, 20, size=n))

pandas/tests/tseries/offsets/test_dst.py

@@ -177,9 +177,6 @@ def test_all_offset_classes(self, tup):
         assert first == second
 
 
-@pytest.mark.xfail(
-    strict=False, reason="'Africa/Kinshasa' test case fails under pytz=2017.3"
-)
 @pytest.mark.parametrize(
     "original_dt, target_dt, offset, tz",
     [

pandas/tests/tseries/offsets/test_offsets_properties.py

@@ -14,7 +14,6 @@
     given,
     strategies as st,
 )
-from hypothesis.errors import Flaky
 from hypothesis.extra.dateutil import timezones as dateutil_timezones
 from hypothesis.extra.pytz import timezones as pytz_timezones
 import pytest
@@ -111,7 +110,6 @@ def test_on_offset_implementations(dt, offset):
     assert offset.is_on_offset(dt) == (compare == dt)
 
 
-@pytest.mark.xfail(strict=False, raises=Flaky, reason="unreliable test timings")
 @given(gen_yqm_offset)
 def test_shift_across_dst(offset):
     # GH#18319 check that 1) timezone is correctly normalized and

pandas/tests/window/test_rolling.py

@@ -1634,7 +1634,6 @@ def test_rolling_quantile_np_percentile():
     tm.assert_almost_equal(df_quantile.values, np.array(np_percentile))
 
 
-@pytest.mark.xfail(reason="GH#44343", strict=False)
 @pytest.mark.parametrize("quantile", [0.0, 0.1, 0.45, 0.5, 1])
 @pytest.mark.parametrize(
     "interpolation", ["linear", "lower", "higher", "nearest", "midpoint"]