BUG: [unit test] pandas/core/indexes/multi.py:3655: TypeError

[adrienpacifico]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

pytest --pdb pandas/tests/test_multilevel.py

Issue Description

The unit-test

pandas/pandas/tests/test_multilevel.py

Line 138 in 65af4ef

def test_alignment(self):

does not pass.

I am using VSCode with the .devcontainer file (thus in a container), with a linux distribution (pop-os).

Stack trace:

self = <pandas.tests.test_multilevel.TestMultiLevel object at 0x7fc319bae370>

    def test_alignment(self):
        x = Series(
            data=[1, 2, 3], index=MultiIndex.from_tuples([("A", 1), ("A", 2), ("B", 3)])
        )
    
        y = Series(
            data=[4, 5, 6], index=MultiIndex.from_tuples([("Z", 1), ("Z", 2), ("B", 3)])
        )
    
>       res = x - y

pandas/tests/test_multilevel.py:147: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
pandas/core/ops/common.py:72: in new_method
    return method(self, other)
pandas/core/arraylike.py:111: in __sub__
    return self._arith_method(other, operator.sub)
pandas/core/series.py:6273: in _arith_method
    self, other = ops.align_method_SERIES(self, other)
pandas/core/ops/__init__.py:168: in align_method_SERIES
    left, right = left.align(right, copy=False)
pandas/core/series.py:4865: in align
    return super().align(
pandas/core/generic.py:9478: in align
    return self._align_series(
pandas/core/generic.py:9585: in _align_series
    join_index, lidx, ridx = self.index.join(
pandas/util/_decorators.py:317: in wrapper
    return func(*args, **kwargs)
pandas/core/indexes/base.py:230: in join
    join_index, lidx, ridx = meth(self, other, how=how, level=level, sort=sort)
pandas/core/indexes/base.py:4725: in join
    return self._join_via_get_indexer(other, how, sort)
pandas/core/indexes/base.py:4747: in _join_via_get_indexer
    join_index = self.union(other)
pandas/core/indexes/base.py:3351: in union
    result = self._union(other, sort=sort)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

self = MultiIndex([('A', 1),
            ('A', 2),
            ('B', 3)],
           )
other = MultiIndex([('Z', 1),
            ('Z', 2),
            ('B', 3)],
           ), sort = None

    def _union(self, other, sort) -> MultiIndex:
        other, result_names = self._convert_can_do_setop(other)
        if (
            any(-1 in code for code in self.codes)
            and any(-1 in code for code in other.codes)
            or other.has_duplicates
        ):
            # This is only necessary if both sides have nans or other has dups,
            # fast_unique_multiple is faster
            result = super()._union(other, sort)
    
            if isinstance(result, MultiIndex):
                return result
            return MultiIndex.from_arrays(
                zip(*result), sortorder=None, names=result_names
            )
    
        else:
            rvals = other._values.astype(object, copy=False)
>           right_missing = lib.fast_unique_multiple(self._values, rvals)
E           TypeError: Argument 'arrays' has incorrect type (expected list, got numpy.ndarray)

pandas/core/indexes/multi.py:3655: TypeError

Expected Behavior

The test should pass

Installed Versions

INSTALLED VERSIONS ------------------ commit : aea824f python : 3.8.13.final.0 python-bits : 64 OS : Linux OS-release : 5.8.0-7630-generic Version : #32~1609193707~20.10~781bb80-Ubuntu SMP Tue Jan 5 21:29:56 UTC 2 machine : x86_64 processor : x86_64 byteorder : little LC_ALL : C.UTF-8 LANG : C.UTF-8 LOCALE : en_US.UTF-8

pandas : 1.6.0.dev0+119.gaea824f9aa.dirty
numpy : 1.22.4
pytz : 2022.2.1
dateutil : 2.8.2
setuptools : 65.3.0
pip : 22.2.2
Cython : 0.29.32
pytest : 7.1.3
hypothesis : 6.54.5
sphinx : 4.5.0
blosc : None
feather : None
xlsxwriter : 3.0.3
lxml.etree : 4.9.1
html5lib : 1.1
pymysql : 1.0.2
psycopg2 : 2.9.3
jinja2 : 3.0.3
IPython : 8.5.0
pandas_datareader: 0.10.0
bs4 : 4.11.1
bottleneck : 1.3.5
brotli :
fastparquet : 0.8.3
fsspec : 2021.11.0
gcsfs : 2021.11.0
matplotlib : 3.5.3
numba : 0.55.2
numexpr : 2.8.3
odfpy : None
openpyxl : 3.0.10
pandas_gbq : 0.17.8
pyarrow : 9.0.0
pyreadstat : 1.1.9
pyxlsb : 1.0.9
s3fs : 2021.11.0
scipy : 1.9.1
snappy :
sqlalchemy : 1.4.41
tables : 3.7.0
tabulate : 0.8.10
xarray : 2022.6.0
xlrd : 2.0.1
xlwt : 1.3.0
zstandard : 0.18.0
tzdata : None

[phofl]

Hi, thanks for your report. You have to rebuild the c extensions. lib.fast_unique_multiple was changed like an hour ago

[adrienpacifico]

Thanx @phofl phofl!
Won't it be possible to have a test that check that the c extension has been rebuilt if there is a change to the pandas/_libs file since the last compilation to avoid future mistake like mine? I ran the test_fast.sh file at first, maybe a clever git command could be included there?

[phofl]

Hm not sure, could you open another issue about this? Theoretically, you could rebuild the extensions from within that file I think

[adrienpacifico]

Tests are passing after rebuilding the c extension.