I got the following df

How can i get the following df

I filtered the alphabetic values from it with replace function. And then using the update function but its not working. Does someone know a better way

housenumber = df['huisnummer'].str.replace('[^a-zA-Z.,-/]', '')

df.set_index('serial', inplace=True)
df.update(huisnummer)
df.reset_index(inplace=True)  # to recover the initial structure

CodePudding user response：

Use .str.extract:

df['housenumber_extra'] = df['housenumber'].str.extract(r'\d ([^\d]. )$')[0].str.strip().fillna(df['housenumber_extra']).str.upper()

Output:

>>> df
  housenumber housenumber_extra
0           1               NaN
1           2                 A
2         55a                 A
3        55-4                -4
4      3455 C                 C

CodePudding user response：

It seems that you're actually not using the second column and can extract all info directly from the first column:

df[['housenumber', 'housenumber_extra']] = df['housenumber'].str.extract('(\d )\W*(\w*)')

output:

  housenumber housenumber_extra
0           1                  
1           2                  
2          55                 a
3          55                 4
4        3455                 C

Or even without assignment using named capturing groups:

df['housenumber'].str.extract('(?P<housenumber>\d )\W*(?P<housenumber_extra>\w*)')