BUG: pandas merge suffixes accepting set can interchange right and left suffix order

[PuneethaPai]

• 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.

---

df_b = pd.DataFrame(dict(p=[1,2,3], q=[2,3,4]))
df_a = pd.DataFrame(dict(p=[2,3,4], q=[1,2,3]))
pd.merge(left=df_b, right=df_a, on=['p'], how='outer', suffixes={'_b', '_a'})

Current output:

	p	q_a	q_b
0	1	2.0	NaN
1	2	3.0	1.0
2	3	4.0	2.0
3	4	NaN	3.0

Expected Output:

	p	q_b	q_a
0	1	2.0	NaN
1	2	3.0	1.0
2	3	4.0	2.0
3	4	NaN	3.0

Problem description

This line is causing the issue:

lsuf, rsuf = self.suffixes

When you unpack set it returns elements in sorted order:

l, r = (2, 1); print(l, r) # l=2, r=1
l, r = [2, 1]; print(l, r) # l=2, r=1
l, r = {2, 1}; print(l, r) # l=1, r=2

Output of pd.show_versions()

[INSTALLED VERSIONS ------------------ commit : None python : 3.8.2.final.0 python-bits : 64 OS : Linux OS-release : 4.14.165-133.209.amzn2.x86_64 machine : x86_64 processor : x86_64 byteorder : little LC_ALL : None LANG : en_US.UTF-8 LOCALE : en_US.UTF-8

pandas : 1.0.3
numpy : 1.18.3
pytz : 2019.3
dateutil : 2.8.1
pip : 20.0.2
setuptools : 46.1.3.post20200330
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : 7.13.0
pandas_datareader: None
bs4 : None
bottleneck : None
fastparquet : None
gcsfs : None
lxml.etree : None
matplotlib : 3.2.1
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pytables : None
pytest : None
pyxlsb : None
s3fs : 0.4.2
scipy : 1.4.1
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
xlsxwriter : None
numba : None]

[jreback]

this should be an error
suffixes must be a tuple / list

a PR would be accepted for this patch

[PuneethaPai]

I would like to contribute. @jreback we should restrict suffixes parameter type to be either tuple/list right?
Else raise InvalidInputException or ValueError something like this right?

[jreback]

yep ValueError