BUG: pandas.DataFrame().stack() raise an error, while expected is empty

[evyasonov]

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

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

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Code Sample, a copy-pastable example

pandas.DataFrame().stack()
pandas.DataFrame().stack(dropna = True)

Problem description

I expect that stack() of empty dataframe will be empty dataframe and not an exception

[TomAugspurger]

Thanks for the report. I think the expected output should be an empty Series rather than an empty DataFrame. What do you think?

[evyasonov]

@TomAugspurger
I expect a dataframe, because if input is not empty, then the method returns a dataframe

[TomAugspurger]

Doesn't DataFrame.stack return a Series? Or do I misunderstand?

[ivanovmg]

From the code on stack method:

from pandas.core.reshape.reshape import stack, stack_multiple

if isinstance(level, (tuple, list)):
    return stack_multiple(self, level, dropna=dropna)
else:
    return stack(self, level, dropna=dropna)

By default level=-1.
Meanwhile pandas.core.reshape.reshape.stack returns a Series.

[evyasonov]

@TomAugspurger
All of my use cases were about multiindex, so .stack() returns a dataframe for my cases.
But I checked the documentation and it returns a Series if it's just one level.
So let's use the logic @ivanovmg described