[FIXED] I want to replace strings from a column by names with numbers in crescent order

Issue

I have a dataframe with a column named ID, which contains strings like: "10387-14_S91_L001" and "1590-13H-caso14_S76", a total of 20 different IDs which repeats itself along a 107 rows dataframe.

I want to replace the strings that have the substring "caso", to only "caso_00", the number that accompanies must be in crescent order; and the strings that don't contains "caso", must be replaced by "controle_00", also with the crescent order.

I tried to list only the unique() values in the column but can't continue from that

An example:

BEFORE                          AFTER 
0387-14_S91_L001                controle_01
10694-14_S86_L001               controle_02
590-13H-caso14_S76_L001         caso_01
1692-15G-caso20_S74_L001        caso_02

Solution

Unfortunately, a bug prevents me to update my other answer. So here is an alternative.

First create a boolean mask to determine if the row is a "controle" or "caso", then use a groupby.cumcount to increment your values:

m = df['BEFORE'].str.contains(r'caso\d+_')

df['AFTER'] = (m.replace({True: 'caso_', False: 'controle_'})
              +m.groupby(m).cumcount().add(1).astype(str).str.zfill(2)
              )

Output:

                     BEFORE        AFTER
0          0387-14_S91_L001  controle_01
1         10694-14_S86_L001  controle_02
2   590-13H-caso14_S76_L001      caso_01
3  1692-15G-caso20_S74_L001      caso_02

Answered By - mozway