BUG: erroneous initialization of a DataFrame with Series objects

[raphaelquast]

• 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

x = pd.Series(["a", "b", "c"])
y = pd.Series([1, 2, 3])

pd.DataFrame(y, x)
>>>     0
>>> a NaN
>>> b NaN
>>> c NaN

pd.DataFrame(x, y)
>>>      0
>>> 1    b
>>> 2    c
>>> 3  NaN

pd.DataFrame(x.values, y.values)
>>>    0
>>> 1  a
>>> 2  b
>>> 3  c

Problem description

I would expect pd.Series objects to be valid inputs for the DataFrame constructor.
If this is not the case a warning (or even raising an error) would be nice...

Output of pd.show_versions()

INSTALLED VERSIONS

commit : c7f7443
python : 3.9.6.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.17763
machine : AMD64
processor : AMD64 Family 25 Model 33 Stepping 0, AuthenticAMD
byteorder : little
LC_ALL : None
LANG : en
LOCALE : German_Austria.1252

pandas : 1.3.1
numpy : 1.21.1
pytz : 2021.1
dateutil : 2.8.2
pip : 21.2.1
setuptools : 49.6.0.post20210108
Cython : None
pytest : 6.2.4
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader: None
bs4 : None
bottleneck : None
fsspec : None
fastparquet : None
gcsfs : None
matplotlib : 3.4.2
numexpr : 2.7.3
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pyxlsb : None
s3fs : None
scipy : 1.7.0
sqlalchemy : None
tables : 3.6.1
tabulate : None
xarray : None
xlrd : None
xlwt : None
numba : None

[jreback]

use a list of the Series

what u r passing is actually valid but thr 2nd arg is the index

[raphaelquast]

@jreback thanks for responding!

I don't fully get what you mean "it is actually valid"... the result is really not what I've expected
(and i know that the 2nd arg is supposed to be the index... )

let me clarify this a bit more:
i was just expecting that passing a Series object would be similar to passing a list or a 1D numpy array
to summarize:

x = [1,2,3,4,5]
pd.DataFrame(x, index=x)                         # OK
pd.DataFrame(np.array(x), index=x)               # OK
pd.Series(x).to_frame().set_index(pd.Series(x))  # OK

pd.DataFrame(pd.Series(x), index=x)              # NOT OK
pd.DataFrame(dict(a=pd.Series(x)), index=x)      # NOT OK
pd.DataFrame([pd.Series(x)], index=x)            # NOT OK  (generates a 2D grid of values)

[phofl]

The index of the series and the DataFrame Index are aligned, this causes the nans, since you do not have matching entries in there.

x = pd.Series([2, 1, 0])
y = pd.Series([1, 2, 3])

pd.DataFrame(y, x)

returns

2  3
1  2
0  1

which should make clear what happens here

[raphaelquast]

@phofl
aaaah, OK i got it 😄 sorry for the confusion!

so basically if the data (e.g. y) already has an index defined, the index gets aligned with respect to the new index (instead of being replaced with the new index)

I consider myself as a long-term user of pandas but this was so far not clear to me!
(also the doc of pd.DataFrame does not mention this special treatment of Series and DataFrame compared to other iterables)

data ndarray (structured or homogeneous), Iterable, dict, or DataFrame
Dict can contain Series, arrays, constants, dataclass or list-like objects. If data is a dict, column order follows insertion-order.

not sure, but how about a comment in the doc of pd.DataFrame to clarify this?
...or maybe just an Example that uses Series or mixed objects to construct a DataFrame... something like:

>>> pd.DataFrame(dict(a=[1,2,3,4,5], 
>>>                   b=pd.Series([3,4,5], index=[3,4,5])), 
>>>              index=[1,2,3,4,5])
   a    b
1  1  NaN
2  2  NaN
3  3  3.0
4  4  4.0
5  5  5.0

[simonjayhawkins]

PRs always welcome to clarify/enhance the documentation.

[tyuyoshi]

Hi, can I pick this issue as my first OSS contribution?

[simonjayhawkins]

@tyuyoshi sure. go for it!

[dhivyadharshin]

Can I fix the issues?

[Stark-developer01]

@tyuyoshi sure. go for it!

I have proposed a PR at #43271 can the PR be approved for awaiting workflows?

[rajaryanbanka]

Hi, I would like to take up this problem as my first open source contribution.

[ankitasankars]

Hello, Can I use this issue as my first contribution?

[goyaldhara]

take

[Tarunssd]

take

[amanj7820]

df = pd.DataFrame(prcpSeries, columns=['prcp'])
df['tmax'] = tmaxSeries
df = prcpSeries.to_frame(name='prcp’)
An alternative here is to create each as a DataFrame and then perform an outer join (using concat):

df1 = pd.DataFrame(prcpSeries, columns=['prcp'])
df2 = pd.DataFrame(tmaxSeries, columns=['tmax'])
...

df = pd.concat([df1, df2, ...], join='outer', axis=1)

In [21]: dfA = pd.DataFrame([1,2], columns=['A'])

In [22]: dfB = pd.DataFrame([1], columns=['B'])

In [23]: pd.concat([dfA, dfB], join='outer', axis=1)
Out[23]:
A B
0 1 1
1 2 NaN

[tyuyoshi]

Sorry, but I have already open PR #42960
I'm going to wait for review, so please take other issues.

@goyaldhara @Tarunssd @simonjayhawkins