BUG: Pandas any() returning false with true values present (GH #23070) (#24434)

Author: makbigc

doc/source/whatsnew/v0.24.0.rst

@@ -1413,6 +1413,7 @@ Timezones
 - Bug in :class:`Timestamp` constructor where a ``dateutil.tz.tzutc`` timezone passed with a ``datetime.datetime`` argument would be converted to a ``pytz.UTC`` timezone (:issue:`23807`)
 - Bug in :func:`to_datetime` where ``utc=True`` was not respected when specifying a ``unit`` and ``errors='ignore'`` (:issue:`23758`)
 - Bug in :func:`to_datetime` where ``utc=True`` was not respected when passing a :class:`Timestamp` (:issue:`24415`)
+- Bug in :meth:`DataFrame.any` returns wrong value when ``axis=1`` and the data is of datetimelike type (:issue:`23070`)
 
 Offsets
 ^^^^^^^

pandas/core/frame.py

@@ -7291,7 +7291,7 @@ def f(x):
  if filter_type is None or filter_type == 'numeric':
  data = self._get_numeric_data()
  elif filter_type == 'bool':
- data = self._get_bool_data()
+ data = self
  else: # pragma: no cover
  msg = ("Generating numeric_only data with filter_type {f}"
  "not supported.".format(f=filter_type))

pandas/tests/frame/test_analytics.py

@@ -228,13 +228,6 @@ def assert_bool_op_api(opname, bool_frame_with_na, float_string_frame,
  getattr(mixed, opname)(axis=0)
  getattr(mixed, opname)(axis=1)
 
- class NonzeroFail(object):
-
- def __nonzero__(self):
- raise ValueError
-
- mixed['_nonzero_fail_'] = NonzeroFail()
-
  if has_bool_only:
  getattr(mixed, opname)(axis=0, bool_only=True)
  getattr(mixed, opname)(axis=1, bool_only=True)
@@ -1374,25 +1367,22 @@ def test_any_all_extra(self):
  result = df[['C']].all(axis=None).item()
  assert result is True
 
- # skip pathological failure cases
- # class CantNonzero(object):
-
- # def __nonzero__(self):
- # raise ValueError
+ def test_any_datetime(self):
 
- # df[4] = CantNonzero()
-
- # it works!
- # df.any(1)
- # df.all(1)
- # df.any(1, bool_only=True)
- # df.all(1, bool_only=True)
+ # GH 23070
+ float_data = [1, np.nan, 3, np.nan]
+ datetime_data = [pd.Timestamp('1960-02-15'),
+ pd.Timestamp('1960-02-16'),
+ pd.NaT,
+ pd.NaT]
+ df = DataFrame({
+ "A": float_data,
+ "B": datetime_data
+ })
 
- # df[4][4] = np.nan
- # df.any(1)
- # df.all(1)
- # df.any(1, bool_only=True)
- # df.all(1, bool_only=True)
+ result = df.any(1)
+ expected = Series([True, True, True, False])
+ tm.assert_series_equal(result, expected)
 
  @pytest.mark.parametrize('func, data, expected', [
  (np.any, {}, False),