BUG-24915 fix unhashble Series.name in df aggregation

[lofoyet]

• closes Cannot do agg if dataframe has column "name" #24915
• tests added / passed
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

Before creating a pd.Series and pass {name} as argument, check if {name} is hashable. If not use default None.

[lofoyet]

@YaoquanYe @daisysu0310

[lofoyet]

1. not sure about what tests to add

[lofoyet]

@youhealthy and I are the same person

[codecov]

Codecov Report

Merging #24920 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #24920      +/-   ##
==========================================
+ Coverage   92.38%   92.38%   +<.01%     
==========================================
  Files         166      166              
  Lines       52404    52406       +2     
==========================================
+ Hits        48412    48414       +2     
  Misses       3992     3992

Flag	Coverage Δ
#multiple	90.8% <100%> (ø)	⬆️
#single	42.9% <0%> (-0.01%)	⬇️

Impacted Files	Coverage Δ
pandas/core/base.py	97.76% <100%> (+0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 539c54f...b270371. Read the comment docs.

[codecov]

Codecov Report

Merging #24920 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #24920      +/-   ##
==========================================
+ Coverage   92.38%   92.38%   +<.01%     
==========================================
  Files         166      166              
  Lines       52404    52406       +2     
==========================================
+ Hits        48412    48414       +2     
  Misses       3992     3992

Flag	Coverage Δ
#multiple	90.8% <100%> (ø)	⬆️
#single	42.9% <0%> (-0.01%)	⬇️

Impacted Files	Coverage Δ
pandas/core/base.py	97.76% <100%> (+0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 539c54f...b270371. Read the comment docs.

[mroeschke]

Needs:

• Whatsnew entry for 0.24.1
• Test: Add a test based on the example in the original issue. Construct the DataFrame/Series that replicates the issue and the expected DataFrame/Series

[jreback]

this needs a test. I am also not sure this is actually the correct fix. constructio of a Series from scalars with an invalid name with raise now. IOW this should be raising already, so not sure this is a good remedy.

`

In [1]: Series([1,2],name=[1,2])                                                                                                                                                                                                                                        
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-1-360525e04c11> in <module>
----> 1 Series([1,2],name=[1,2])

~/pandas/pandas/core/series.py in __init__(self, data, index, dtype, name, copy, fastpath)
    266         generic.NDFrame.__init__(self, data, fastpath=True)
    267 
--> 268         self.name = name
    269         self._set_axis(0, index, fastpath=True)
    270 

~/pandas/pandas/core/generic.py in __setattr__(self, name, value)
   5077             object.__setattr__(self, name, value)
   5078         elif name in self._metadata:
-> 5079             object.__setattr__(self, name, value)
   5080         else:
   5081             try:

~/pandas/pandas/core/series.py in name(self, value)
    400     def name(self, value):
    401         if value is not None and not is_hashable(value):
--> 402             raise TypeError('Series.name must be a hashable type')
    403         object.__setattr__(self, '_name', value)
    404 

TypeError: Series.name must be a hashable type

[lofoyet]

@jreback I agree with you. You can refer to #24915 for the problem.

There are 2 ways that I have in mind to fix this:

1. This PR.
2. Make change to how getattr works.

For 2. What I looked at is

#pandas.core.generic.py
if (name in self._internal_names_set or name in self._metadata or
    name in self._accessors):
        return object.__getattribute__(self, name)
else:
    if self._info_axis._can_hold_identifiers_and_holds_name(name):
        return self[name]
    return object.__getattribute__(self, name)

getattr(self, "name") will go to else clause, and if "name" is a column, it will return in if clause. But df.name is really an attribute rather not a column name. It should skip that "if" in "else" and goes to object.getattribute().

How do you propose to change? @jreback

[lofoyet]

@jreback any idea?

[jreback]

@lofoyet first thing you need here is a test. Then step thru it until you get to the source of the error, see if you patch fixes.

I would rather shy away from changing the way getattr works. Also I am not sure your change is the right way.

[WillAyd]

Closing as stale. @lofoyet ping if you'd like to continue though as mentioned above this needs a test first and foremost