[Python] Search pattern not unique? - Regular expression

I want to write a function to clean the index column of the dataframe.

1. 
   Delete the whole row that has high-level IDs. For example, delete
   
   East Kootenay (5901) 01010
   
   
2. 
   Tailor the index into 7-digit number for low-level IDs. For example, turn
   
   East Kootenay A (5901017) RDA 02020 into 5901017
   
   
3. 
   If it has two parenthesis keep only the 7-digit number in the second parenthesis. For example,
   
   Sechelt (Part) (5929803) IGD 02020 to 5929803
   
   Capital H (Part 1) (5917054) RDA 01020 to 5917054
   
   Capital H (Part 2) (5917056) RDA 02030 to 5917056
   
   T'Sou-ke 1 (Sooke 1) (5917817) IRI 01010 to 5917817
   
   T'Sou-ke 2 (Sooke 2) (5917818) IRI 00000 to 5917818
   

An example of code only works for one bracket is

def extract_id(s):
m = re.search('\((.*)\)', s)
if m:
i = int(m.group(0)[1:-1])
return i

if __name__ == '__main__':
# Read data
census_subdivision_profile = pd.read_excel('../data/census_subdivision_profile.xlsx', sheetname='Data',
index_col='Geography', encoding='utf-8').T
print(census_subdivision_profile.head())
print(census_subdivision_profile.shape)

census_subdivision_profile.index = census_subdivision_profile.index.map(extract_id)
print(census_subdivision_profile.index)


To see the full code, see another question I posted earlier

Merge dataframes that have indices that one contains another (but not the same)

Continue reading...