Operation between DataFrame with non-numeric types and incomplete series #22663
Description
Code Sample, a copy-pastable example if possible
In [2]: df = pd.DataFrame([[1, 2, 3], [4, 5, 6]], columns=list('abc'))
In [3]: df + pd.Series([-1, -2], index=list('ab'))
Out[3]:
a b c
0 0.0 0.0 NaN
1 3.0 3.0 NaN
In [4]: df['c'] = 'c'
In [5]: df + pd.Series([-1, -2], index=list('ab'))
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
~/nobackup/repo/pandas/pandas/core/ops.py in na_op(x, y)
1714 try:
-> 1715 result = expressions.evaluate(op, str_rep, x, y, **eval_kwargs)
1716 except TypeError:
~/nobackup/repo/pandas/pandas/core/computation/expressions.py in evaluate(op, op_str, a, b, use_numexpr, **eval_kwargs)
204 if use_numexpr:
--> 205 return _evaluate(op, op_str, a, b, **eval_kwargs)
206 return _evaluate_standard(op, op_str, a, b)
~/nobackup/repo/pandas/pandas/core/computation/expressions.py in _evaluate_numexpr(op, op_str, a, b, truediv, reversed, **eval_kwargs)
119 if result is None:
--> 120 result = _evaluate_standard(op, op_str, a, b)
121
~/nobackup/repo/pandas/pandas/core/computation/expressions.py in _evaluate_standard(op, op_str, a, b, **eval_kwargs)
64 with np.errstate(all='ignore'):
---> 65 return op(a, b)
66
TypeError: Can't convert 'float' object to str implicitly
During handling of the above exception, another exception occurred:
ValueError Traceback (most recent call last)
<ipython-input-5-c60f1e54c070> in <module>()
----> 1 df + pd.Series([-1, -2], index=list('ab'))
~/nobackup/repo/pandas/pandas/core/ops.py in f(self, other, axis, level, fill_value)
1737 return _combine_series_frame(self, other, na_op,
1738 fill_value=fill_value, axis=axis,
-> 1739 level=level, try_cast=True)
1740 else:
1741 if fill_value is not None:
~/nobackup/repo/pandas/pandas/core/ops.py in _combine_series_frame(self, other, func, fill_value, axis, level, try_cast)
1655 # default axis is columns
1656 return self._combine_match_columns(other, func, level=level,
-> 1657 try_cast=try_cast)
1658
1659
~/nobackup/repo/pandas/pandas/core/frame.py in _combine_match_columns(self, other, func, level, try_cast)
4832 new_data = left._data.eval(func=func, other=right,
4833 axes=[left.columns, self.index],
-> 4834 try_cast=try_cast)
4835 return self._constructor(new_data)
4836
~/nobackup/repo/pandas/pandas/core/internals/managers.py in eval(self, **kwargs)
527
528 def eval(self, **kwargs):
--> 529 return self.apply('eval', **kwargs)
530
531 def quantile(self, **kwargs):
~/nobackup/repo/pandas/pandas/core/internals/managers.py in apply(self, f, axes, filter, do_integrity_check, consolidate, **kwargs)
421
422 kwargs['mgr'] = self
--> 423 applied = getattr(b, f)(**kwargs)
424 result_blocks = _extend_blocks(applied, result_blocks)
425
~/nobackup/repo/pandas/pandas/core/internals/blocks.py in eval(self, func, other, errors, try_cast, mgr)
1416 try:
1417 with np.errstate(all='ignore'):
-> 1418 result = get_result(other)
1419
1420 # if we have an invalid shape/broadcast error
~/nobackup/repo/pandas/pandas/core/internals/blocks.py in get_result(other)
1384 result = func(values, other)
1385 else:
-> 1386 result = func(values, other)
1387
1388 # mask if needed
~/nobackup/repo/pandas/pandas/core/ops.py in na_op(x, y)
1715 result = expressions.evaluate(op, str_rep, x, y, **eval_kwargs)
1716 except TypeError:
-> 1717 result = masked_arith_op(x, y, op)
1718
1719 result = missing.fill_zeros(result, x, y, op_name, fill_zeros)
~/nobackup/repo/pandas/pandas/core/ops.py in masked_arith_op(x, y, op)
819 # errors in Py3 (TypeError) vs Py2 (ValueError)
820 # Note: Only = an issue in DataFrame case
--> 821 raise ValueError('Cannot broadcast operands together.')
822
823 if mask.any():
ValueError: Cannot broadcast operands together.Problem description
df + df[['a', 'b']] works, and the above also worked until recently (it works for instance in 0.19, although it raises a VisibleDeprecationWarning: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 2 - which however makes no reference to the mixed dtype).
Expected Output
In 0.19, In [3] and In [5] would yield the same result.
Output of pd.show_versions()
INSTALLED VERSIONS
commit: None
python: 3.5.3.final.0
python-bits: 64
OS: Linux
OS-release: 4.9.0-7-amd64
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: it_IT.UTF-8
LOCALE: it_IT.UTF-8
pandas: 0.24.0.dev0+465.ge775f9ab6
pytest: 3.5.0
pip: 9.0.1
setuptools: 39.2.0
Cython: 0.28.4
numpy: 1.14.3
scipy: 0.19.0
pyarrow: None
xarray: None
IPython: 6.2.1
sphinx: 1.5.6
patsy: 0.5.0
dateutil: 2.7.3
pytz: 2018.4
blosc: None
bottleneck: 1.2.0dev
tables: 3.3.0
numexpr: 2.6.1
feather: 0.3.1
matplotlib: 2.2.2.post1634.dev0+ge8120cf6d
openpyxl: 2.3.0
xlrd: 1.0.0
xlwt: 1.3.0
xlsxwriter: 0.9.6
lxml: 4.1.1
bs4: 4.5.3
html5lib: 0.999999999
sqlalchemy: 1.0.15
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: 0.2.1
gcsfs: None