.replace() raises when replacing strings

[carlosdanielcsantos]

Code Sample, a copy-pastable example if possible

pd.Series(['a', 'b', 'c']).replace('a', 'x')

Problem description

This raises the following error:

TypeError: Type aliases cannot be used with isinstance().

This behavior is exclusive to the new version 0.23. Previous versions work as expected.

Full traceback

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-75-761515b509b4> in <module>()
----> 1 pd.Series(['a','b','c']).replace('a', 'x')

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/series.py in replace(self, to_replace, value, inplace, limit, regex, method)
   3427         return super(Series, self).replace(to_replace=to_replace, value=value,
   3428                                            inplace=inplace, limit=limit,
-> 3429                                            regex=regex, method=method)
   3430
   3431     @Appender(generic._shared_docs['shift'] % _shared_doc_kwargs)

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/generic.py in replace(self, to_replace, value, inplace, limit, regex, method)
   5877                     new_data = self._data.replace(to_replace=to_replace,
   5878                                                   value=value, inplace=inplace,
-> 5879                                                   regex=regex)
   5880                 else:
   5881                     msg = ('Invalid "to_replace" type: '

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/internals.py in replace(self, **kwargs)
   3718
   3719     def replace(self, **kwargs):
-> 3720         return self.apply('replace', **kwargs)
   3721
   3722     def replace_list(self, src_list, dest_list, inplace=False, regex=False,

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/internals.py in apply(self, f, axes, filter, do_integrity_check, consolidate, **kwargs)
   3579
   3580             kwargs['mgr'] = self
-> 3581             applied = getattr(b, f)(**kwargs)
   3582             result_blocks = _extend_blocks(applied, result_blocks)
   3583

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/internals.py in replace(self, to_replace, value, inplace, filter, regex, convert, mgr)
   2430         blocks = [self]
   2431
-> 2432         if not either_list and is_re(to_replace):
   2433             return self._replace_single(to_replace, value, inplace=inplace,
   2434                                         filter=filter, regex=True,

~/.virtualenvs/new/lib/python3.5/site-packages/pandas/core/dtypes/inference.py in is_re(obj)
    217     """
    218
--> 219     return isinstance(obj, re_type)
    220
    221

/usr/lib/python3.5/typing.py in __instancecheck__(self, obj)
    256
    257     def __instancecheck__(self, obj):
--> 258         raise TypeError("Type aliases cannot be used with isinstance().")
    259
    260     def __subclasscheck__(self, cls):

TypeError: Type aliases cannot be used with isinstance().

Expected Output

0    x
1    b
2    c
dtype: object

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.5.1.final.0
python-bits: 64
OS: Linux
OS-release: 4.13.13-5-pve
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: en_US.UTF-8
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8

pandas: 0.23.0
pytest: None
pip: 8.1.1
setuptools: 20.7.0
Cython: None
numpy: 1.14.3
scipy: 1.1.0
pyarrow: None
xarray: None
IPython: 6.4.0
sphinx: None
patsy: None
dateutil: 2.6.1
pytz: 2018.4
blosc: None
bottleneck: None
tables: None
numexpr: None
feather: None
matplotlib: 2.2.2
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: None
html5lib: 0.999
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[TomAugspurger]

Hmm I'm not able to reproduce on Python 3.6.

Can you past the full traceback?

[TomAugspurger]

I don't understand the bit about "type aliases". Are you overriding a builting with something from the typing module?

[carlosdanielcsantos]

@TomAugspurger I added the full traceback

[TomAugspurger]

Does this raise for you?

In [10]: import re

In [11]: import typing

In [12]: isinstance(re.compile(''), typing.re.Pattern)
Out[12]: True

If so, can you check the Python issue tracker to see if this has been filed?

[TomAugspurger]

That works fine on 3.5.4 for me.

Python 3.5.4 | packaged by conda-forge | (default, Aug 10 2017, 01:41:15)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import re
>>> import typing
>>> isinstance(re.compile(''), typing.re.Pattern)
True
>>>

Can you try upgrading from 3.5.1 to a later point release?

[carlosdanielcsantos]

It definitely works with 3.5.4 but doesn't work with 3.5.2 either

[TomAugspurger]

OK. See #20553 for reference.

If you want, you can make a PR using the old way for Python 3.5.2 (or 3) and older... Or upgrade to the latest point release

[carlosdanielcsantos]

Will do, thanks!

[adamabk]

I had the same issue using the same Python 3.5.2. I just opened up the PR for it to revert it back.
Was that what you mean @TomAugspurger?