replace substrings in pandas columns while skipping rows with value: None

Question

I stole this from here Remove/replace columns values based on another columns using pandas

[a.replace(b,'') for a,b in zip(df1['asker'], df1['party']) if a != None]

I added if a != None because it always threw error: AttributeError: 'NoneType' object has no attribute 'replace'

Here are different solutions to the same problem: replacing substring df1['party'] within col asker

df1['new_column'] = df1['asker'].replace(to_replace=r'\b'+df1['party']+r'\b', value='',regex=True)

df1['asker'] = df1.apply(lambda x: x['asker'].replace(x['party'], ''), axis = 1)

None of them worked as soon as i added the exception for None values

example of df1 columns party

[QQQ,
None,
RRR-Fraktion]

example of df1 columns asker

[Konrad Munch QQQ,
None,
Heiko Baer RRR-Fraktion]

Some problem with solution?

jezrael
– jezrael

2021-10-25 12:56:04 +00:00
Commented Oct 25, 2021 at 12:56 — jezrael
– jezrael, Commented Oct 25, 2021 at 12:56

jezrael · Accepted Answer · 2021-10-25 13:34:27Z

1

Use:

[a.replace(b,'') if (a != None) and (b != None) 
                 else a
                 for a,b in zip(df1['asker'], df1['party'])]

If need test NaNs or Nones use notna:

df1 = pd.DataFrame({"asker": ["Heiko Baer RRR-Fraktion", "a", "b", 
                               np.nan, None, None, np.nan], 
                    "party": ['RRR-Fraktion', None, np.nan, 'a', 's', None, np.nan]})
    
df1['asker'] = [a.replace(b,'') if pd.notna(a) and pd.notna(b) 
                                else a 
                                for a,b in zip(df1['asker'], df1['party'])]
print (df1)
         asker         party
0  Heiko Baer   RRR-Fraktion
1            a          None
2            b           NaN
3          NaN             a
4         None             s
5         None          None
6          NaN           NaN

edited Oct 25, 2021 at 13:34

answered Oct 25, 2021 at 10:25

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

id345678 Over a year ago

thanks for the suggestion. I had something like this in mind. It doesent remove the substring of party from the asker thouhg. I does nothing. I just tested your second suggestion. It doesent do anything at all, unfortunately. I cannot explain why

jezrael Over a year ago

@id345678 - Can you be more specific?

id345678 Over a year ago

I dont know how to be more specific. I could give you an example maybe?

jezrael Over a year ago

@id345678 - added to sample data in answer. Konrad Munch QQQ is not substring in Fraktion der QQQ so no replacement, so not added to sample data

Collectives™ on Stack Overflow

replace substrings in pandas columns while skipping rows with value: None

1 Answer 1

4 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related