Does pandas really perform index alignment?

In a long code, I found myself at a point where I had to compare two series. I can’t share the full code but I made example :

series1 = pd.Series(['a', 'b', 'c'], index=[2, 1, 3])
series2 = pd.Series(['b', 'a', 'c'], index=[3, 2, 1])

print(series1)
2    a
1    b
3    c
dtype: object

print(series2)
3    b
2    a
1    c
dtype: object

For the comparison it’s a simple series1 == series2 but surprisingly I got an error :

ValueError: Can only compare identically-labeled Series objects

MEDevel.com: Open-source for Healthcare and Education

Collecting and validating open-source software for healthcare, education, enterprise, development, medical imaging, medical records, and digital pathology.

Visit Medevel

To solve the problem I searched in stackoveflow and people say it’s because of the different indexes which appear to be true because the code below solved my issue :
Pandas "Can only compare identically-labeled DataFrame objects" error

series1.sort_index() == series2.sort_index()

1    False
2     True
3    False
dtype: bool

My question is : do pandas really do index alignment ? Or is it just a myth ? Because I feel like sometimes it does and sometimes not.

Do you guys have an explanation to my error and an answer to my question ?

>Solution :

Prefer the use of eq that is more powerful than == (likewise le/lt/ge/gt in place of <=/</>=/>):

series1.eq(series2)

Output:

1    False
2     True
3    False
dtype: bool

Alternative with a manual reindexing:

series1 == series2.reindex_like(series1)

2     True
1    False
3    False
dtype: bool