BUG: pd._libs.lib.is_iterator reports Timestamp as iterable on Cygwin

[DWesl]

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 master branch of pandas.

Reproducible Example

>>> import pandas as pd
>>> stamp = pd.Timestamp("2010-01-03T07:23:48")
>>> pd._libs.lib.is_iterator(stamp)
True
>>> list(stamp)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'Timestamp' object is not iterable
>>> pd.__version__
'1.3.4'
>>> import collections.abc
>>> isinstance(stamp, collections.abc.Iterator)
False
>>> next(stamp)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'Timestamp' object is not an iterator

$ for ver in 3.{6,7,8}; do 
>     python${ver} -c "import collections.abc, pandas as pd; stamp = pd.Timestamp('1987-06-05T04:32:10'); 
> print(pd.__version__, pd._libs.lib.is_iterator(stamp), isinstance(stamp, collections.abc.Iterator))"; 
> done
1.1.4 True False
1.3.4 True False
1.3.4 True False

Issue Description

I have noticed two functions in pandas that use the above pattern:

pandas.core.indexing._LocationIndexer.__getitem__(self, tuple)
pandas.core.indexing._LocIndexer._getitem_axis(self, tuple)

This is causing tests to fail as I try to build 1.3.5. Both the attempt shown above and the attempt with 1.3.5 are on Cygwin. The problem does not occur with pandas 1.0.5, 1.1.2, or 1.3.5 on Windows. All attempts were made with the CPython implementation.

Expected Behavior

I would expect either is_iterator to return False or list to succeed.

Installed Versions

>>> pd.show_versions()
/usr/lib/python3.8/site-packages/h5py/__init__.py:36: UserWarning: h5py is running against HDF5 1.10.8 when it was built against 1.10.7, this may cause problems
  _warn(("h5py is running against HDF5 {0} when it was built against {1}, "

INSTALLED VERSIONS
------------------
commit           : 945c9ed766a61c7d2c0a7cbb251b6edebf9cb7d5
python           : 3.8.12.final.0
python-bits      : 64
OS               : CYGWIN_NT-10.0-19043
OS-release       : 3.3.3-341.x86_64
Version          : 2021-12-03 16:35 UTC
machine          : x86_64
processor        :
byteorder        : little
LC_ALL           : None
LANG             : en_US.UTF-8
LOCALE           : en_US.UTF-8

pandas           : 1.3.4
numpy            : 1.21.4
pytz             : 2021.3
dateutil         : 2.8.2
pip              : 21.3.1
setuptools       : 59.6.0
Cython           : 0.29.25
pytest           : 6.2.5
hypothesis       : 6.31.3
sphinx           : 4.0.3
blosc            : None
feather          : None
xlsxwriter       : None
lxml.etree       : 4.7.1
html5lib         : 1.1
pymysql          : None
psycopg2         : None
jinja2           : 3.0.3
IPython          : 7.30.1
pandas_datareader: None
bs4              : 4.10.0
bottleneck       : 1.3.2
fsspec           : 2021.06.0
fastparquet      : None
gcsfs            : None
matplotlib       : 3.5.1
numexpr          : 2.8.0
odfpy            : None
openpyxl         : None
pandas_gbq       : None
pyarrow          : None
pyxlsb           : None
s3fs             : None
scipy            : 1.7.1
sqlalchemy       : 1.4.28
tables           : None
tabulate         : 0.8.7
xarray           : 0.18.2
xlrd             : None
xlwt             : None
numba            : None

[mroeschke]

On OSX I get this on master

In [1]: >>> import pandas as pd
   ...: >>> stamp = pd.Timestamp("2010-01-03T07:23:48")
   ...: >>> pd._libs.lib.is_iterator(stamp)
Out[1]: False

This is causing tests to fail as I try to build 1.3.5

Could you show a full traceback of the build logs?

[DWesl]

I'm not sure what you mean by "full traceback from build logs." The full build log is bigger than the maximum 65536-character limit for a comment, and I can't figure out how to attach.

[mroeschke]

You could pipe the build log to a text file and attach the file to a Github comment

[DWesl]

build.log
Found the attach button. I think this is the full build log for python 3.7

[DWesl]

I got a Cygwin compile running with GitHub Actions; this looks like one instance where this error arises. You should be able to see the build log from earlier on that same page.

[DWesl]

I updated the Cygwin compile to merge the latest main branch from pandas-dev/pandas: the same error still shows up in the test suite. You can see the full build log here.

What information do you still need?

[mroeschke]

Thanks for confirming @DWesl.

[DWesl]

I just checked a few other cases, and it looks like PyIter_Check(obj) disagrees with isinstance(obj, collections.abc.Iterable) for os.environ and fractions.Fraction(0, 1), so I'm suspecting this is exposing an underlying CPython bug.

[mroeschke]

Given #46477 (comment), it appears that this is an issue that should be fixed upstream. Closing